diff --git a/Ready/BUILD.txt b/BUILD.txt
similarity index 100%
rename from Ready/BUILD.txt
rename to BUILD.txt
diff --git a/Brandeisator/CMakeLists.txt b/Brandeisator/CMakeLists.txt
deleted file mode 100644
index 882e8ea4f..000000000
--- a/Brandeisator/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(Brandeisator)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(Brandeisator 
-  brandeisator.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(Brandeisator ${OpenCV_LIBS} )
diff --git a/Brandeisator/brandeisator.cpp b/Brandeisator/brandeisator.cpp
deleted file mode 100644
index f55fd79a1..000000000
--- a/Brandeisator/brandeisator.cpp
+++ /dev/null
@@ -1,172 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float D_a,float D_b,float A,float B,
-             float speed,
-             bool parameter_space);
-
-int main()
-{
-    // Here we implement the Brandeisator; the Lengyel-Epstein model for the 
-    // chlorite-iodide-malonic acid (CIMA) reaction 
-    // http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.34.9043&rep=rep1&type=pdf
-
-    // But I can't get this to work at the moment. Perhaps we need a more sophisticated solver?
-    // Or the initial conditions need to be right to get interesting behaviour?
-
-    // -- parameters --
-    float A = 12.371f;
-    float B = 16.0f;
-    float D_a = 0.1f;
-    float D_b = 1.0f;
-    float speed = 0.01f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,da,db,D_a,D_b,A,B,speed,false);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Brandeisator - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,b,iteration,true,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            // start with a uniform field with an approximate circle in the middle
-            //if(hypot(i%20-10/*-X/2*/,j%20-10/*-Y/2*/)<=frand(2,5)) {
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5))
-            {
-                a[i][j] = 0.0f;
-                b[i][j] = 1.0f;
-            }
-            else {
-                a[i][j] = 1;
-                b[i][j] = 0;
-            }
-            a[i][j] = frand(-2.0f,2.0f);
-            b[i][j] = frand(-2.0f,2.0f);
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float D_a,float D_b,float A,float B,float speed,
-             bool parameter_space)
-{
-    const bool toroidal = false;
-
-    int iprev,inext,jprev,jnext;
-
-    // compute change in each cell
-    for(int i = 0; i < X; i++) {
-        if(toroidal) {
-            iprev = (i + X - 1) % X;
-            inext = (i + 1) % X;
-        }
-        else {
-            iprev = max(0,i-1);
-            inext = min(X-1,i+1);
-        }
-
-        for(int j = 0; j < Y; j++) {
-            if(toroidal) {
-                jprev = (j + Y - 1) % Y;
-                jnext = (j + 1) % Y;
-            }
-            else {
-                jprev = max(0,j-1);
-                jnext = min(Y-1,j+1);
-            }
-
-            if(parameter_space) {
-                /*const float kmin=0.03f,kmax=0.07f,fmin=0.00f,fmax=0.06f;
-                // set f and k for this location (ignore the provided values of f and k)
-                k = kmin + i*(kmax-kmin)/X;
-                f = fmin + j*(fmax-fmin)/Y;*/
-            }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = D_a * dda + A - aval - (aval*bval) / (1+aval*aval);
-            db[i][j] = D_b * ddb + 4*B*aval - B * (aval*bval) / (1+aval*aval);
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
-
diff --git a/Brusselator/CMakeLists.txt b/Brusselator/CMakeLists.txt
deleted file mode 100644
index fbb280370..000000000
--- a/Brusselator/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(Brusselator)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(Brusselator
-  brusselator.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(Brusselator ${OpenCV_LIBS} )
diff --git a/Brusselator/brusselator.cpp b/Brusselator/brusselator.cpp
deleted file mode 100644
index fd2618bb7..000000000
--- a/Brusselator/brusselator.cpp
+++ /dev/null
@@ -1,147 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y],float A,float B);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float A,float B,float D1,float D2,
-             float speed,
-             bool parameter_space);
-
-int main()
-{
-    // -- parameters --
-    float A = 3.0f;
-    float B = 10.0f;
-    float D1 = 5.0f;
-    float D2 = 12.0f;
-    float speed = 0.001f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b,A,B);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,da,db,A,B,D1,D2,speed,false);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Brusselator - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,b,a,iteration,true,20.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y],float A,float B)
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) 
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] = A + frand(-0.01f,0.01f);
-            b[i][j] = B/A + frand(-0.01f,0.01f);
-        }
-    }
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float A,float B,float D1,float D2,
-             float speed,
-             bool parameter_space)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        //int iprev = (i + X - 1) % X;
-        //int inext = (i + 1) % X;
-        int iprev = max(0,i-1);
-        int inext = min(X-1,i+1);
-
-        for(int j = 0; j < Y; j++) 
-        {
-            //int jprev = (j + Y - 1) % Y;
-            //int jnext = (j + 1) % Y;
-            int jprev = max(0,j-1);
-            int jnext = min(Y-1,j+1);
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            if(parameter_space)
-            {
-                const float A1=0.0f,A2=4.0f,B1=0.0f,B2=15.0f;
-                A = A1+(A2-A1)*i/X;
-                B = B1+(B2-B1)*j/Y;
-            }
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = A-(B+1)*aval + aval*aval*bval + D1*dda;
-            db[i][j] = B*aval - aval*aval*bval + D2*ddb;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
-
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 53c03e73e..b5a025f0a 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,18 +1,459 @@
-cmake_minimum_required(VERSION 2.6)
-
-project(ReactionDiffusion)
-
-add_subdirectory(TuringSpots)
-add_subdirectory(MeinhardtSpots)
-add_subdirectory(MeinhardtStripes)
-add_subdirectory(GrayScott)
-add_subdirectory(Brusselator)
-add_subdirectory(Oregonator)
-add_subdirectory(Brandeisator)
-add_subdirectory(FitzHughNagumo)
-add_subdirectory(FitzHughNagumo3D)
-add_subdirectory(ComplexGinzbergLandau)
-add_subdirectory(Schlogl)
-add_subdirectory(Schnakenberg)
-add_subdirectory(EdblomOrbanEpstein)
-add_subdirectory(Linear)
+#  Copyright 2011, 2012, 2013 The Ready Bunch
+#
+#  This file is part of Ready.
+#
+#  Ready is free software: you can redistribute it and/or modify
+#  it under the terms of the GNU General Public License as published by
+#  the Free Software Foundation, either version 3 of the License, or
+#  (at your option) any later version.
+#
+#  Ready is distributed in the hope that it will be useful,
+#  but WITHOUT ANY WARRANTY; without even the implied warranty of
+#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+#  GNU General Public License for more details.
+#
+#  You should have received a copy of the GNU General Public License
+#  along with Ready. If not, see <http://www.gnu.org/licenses/>.
+#
+#--------------------------------------------------------------------------
+    
+cmake_minimum_required( VERSION 2.6 )
+cmake_policy( SET CMP0003 NEW )
+
+project( Ready )
+
+set( READY_VERSION 0.6 )   # check matches Help/about.html
+add_definitions( -D READY_VERSION=${READY_VERSION} )
+
+if( APPLE OR WIN32 )
+  # app names are usually capitalized on Mac OS X and Windows
+  set( APP_NAME Ready )
+else()
+  # Linux binaries are usually all lowercase
+  set( APP_NAME ready )
+endif()
+set( CMD_NAME rdy ) # command-line version
+
+#-------------------------------------------source files----------------------------------------------
+
+set( BASE_SOURCES      # low-level code used in all executables
+  src/readybase/AbstractRD.hpp                src/readybase/AbstractRD.cpp
+  src/readybase/ImageRD.hpp                   src/readybase/ImageRD.cpp
+  src/readybase/GrayScottImageRD.hpp          src/readybase/GrayScottImageRD.cpp
+  src/readybase/OpenCLImageRD.hpp             src/readybase/OpenCLImageRD.cpp
+  src/readybase/FormulaOpenCLImageRD.hpp      src/readybase/FormulaOpenCLImageRD.cpp
+  src/readybase/FullKernelOpenCLImageRD.hpp   src/readybase/FullKernelOpenCLImageRD.cpp
+  src/readybase/MeshRD.hpp                    src/readybase/MeshRD.cpp
+  src/readybase/GrayScottMeshRD.hpp           src/readybase/GrayScottMeshRD.cpp
+  src/readybase/OpenCLMeshRD.hpp              src/readybase/OpenCLMeshRD.cpp
+  src/readybase/FormulaOpenCLMeshRD.hpp       src/readybase/FormulaOpenCLMeshRD.cpp
+  src/readybase/FullKernelOpenCLMeshRD.hpp    src/readybase/FullKernelOpenCLMeshRD.cpp
+  src/readybase/OpenCL_MixIn.hpp              src/readybase/OpenCL_MixIn.cpp
+  src/readybase/OpenCL_utils.hpp              src/readybase/OpenCL_utils.cpp
+  src/readybase/IO_XML.hpp                    src/readybase/IO_XML.cpp
+  src/readybase/overlays.hpp                  src/readybase/overlays.cpp
+  src/readybase/Properties.hpp                src/readybase/Properties.cpp
+  src/readybase/utils.hpp                     src/readybase/utils.cpp
+  src/readybase/OpenCL_Dyn_Load.h             src/readybase/OpenCL_Dyn_Load.c 
+  src/readybase/MeshGenerators.hpp            src/readybase/MeshGenerators.cpp
+  src/readybase/SystemFactory.hpp             src/readybase/SystemFactory.cpp
+  src/readybase/scene_items.hpp               src/readybase/scene_items.cpp
+)
+include_directories( src/readybase )
+
+set( GUI_SOURCES      # high-level GUI code used only in Ready
+  src/gui/IDs.hpp
+  src/gui/wxutils.hpp                      src/gui/wxutils.cpp
+  src/gui/dialogs.hpp                      src/gui/dialogs.cpp
+  src/gui/prefs.hpp                        src/gui/prefs.cpp
+  src/gui/app.hpp                          src/gui/app.cpp 
+  src/gui/frame.hpp                        src/gui/frame.cpp 
+  src/gui/HelpPanel.hpp                    src/gui/HelpPanel.cpp
+  src/gui/InfoPanel.hpp                    src/gui/InfoPanel.cpp
+  src/gui/PatternsPanel.hpp                src/gui/PatternsPanel.cpp
+  src/gui/vtk_pipeline.hpp                 src/gui/vtk_pipeline.cpp
+  src/gui/InteractorStylePainter.hpp       src/gui/InteractorStylePainter.cpp
+  src/gui/wxVTKRenderWindowInteractor.h    src/gui/wxVTKRenderWindowInteractor.cxx
+  src/gui/RecordingDialog.hpp              src/gui/RecordingDialog.cpp
+)
+include_directories( src/gui )
+
+set( CMD_SOURCES      # code used for the command-line version
+  src/cmd/main.cpp
+)
+
+set( RESOURCES
+  resources/ready.rc
+  resources/appicon.ico 
+  resources/appicon16.ico 
+  resources/appicon32.ico 
+  resources/appicon48.ico 
+  resources/appicon.xpm
+  resources/Info.plist.in
+  resources/app.icns 
+  resources/file.icns
+)
+include_directories( resources )
+
+set( PATTERN_FILES
+  Patterns/Meinhardt1982/stripes.vti       Patterns/Meinhardt1982/zebra.vtu
+  Patterns/Schlogl.vti 
+  Patterns/heat_equation.vti 
+  Patterns/Turing1952/spots.vti            Patterns/Turing1952/spots_noisy.vti
+  Patterns/kernel_test.vti 
+  Patterns/parameter_modulation_demo.vti
+  Patterns/parameter_modulation_demo2.vti
+  Patterns/bunny.vtu
+  Patterns/lion.vtu
+  Patterns/heat_equation_interpolation.vti
+  Patterns/Ginzburg-Landau/complex_Ginzburg-Landau.vti
+  Patterns/Ginzburg-Landau/complex_Ginzburg-Landau_magnitude.vti
+  Patterns/wave_equation.vti
+  Patterns/oregonator.vti
+  Patterns/Brusselator.vti
+  Patterns/SmoothLife/smoothglider.vti     Patterns/SmoothLife/smoothlifeL.vti     Patterns/SmoothLife/glider_3D.vti
+  Patterns/Purwins1999/glider.vti          Patterns/Purwins1999/glider_3D.vti      Patterns/Purwins1999/multiGlider.vti
+  Patterns/CPU-only/grayscott_1D.vti 
+  Patterns/CPU-only/grayscott_2D.vti 
+  Patterns/CPU-only/grayscott_3D.vti
+  Patterns/FitzHugh-Nagumo/tip-splitting.vti 
+  Patterns/FitzHugh-Nagumo/spiral_turbulence.vti
+  Patterns/FitzHugh-Nagumo/pulsate.vti 
+  Patterns/FitzHugh-Nagumo/squid_axon.vti
+  Patterns/FitzHugh-Nagumo/Ising_regime.vti
+  Patterns/Gray-Scott/Lesmes_noisy.vti
+  Patterns/Gray-Scott/noisy_solitons_mitosis.vti
+  Patterns/Gray-Scott/parameter-map.vti
+  Patterns/Gray-Scott/Pearson1993.vti
+  Patterns/Gray-Scott/self-replicating_spots.vti 
+  Patterns/Gray-Scott/U-Skate/Hutton-and-helix-gliders.vti
+  Patterns/Gray-Scott/U-Skate/Munafo_glider.vti
+  Patterns/Gray-Scott/U-Skate/o-ring_2D.vti
+  Patterns/CellularAutomata/Bays_3D.vti
+  Patterns/CellularAutomata/Conway_life.vti 
+  Patterns/CellularAutomata/life_torus.vtu
+  Patterns/CellularAutomata/larger-than-life.vti
+  Patterns/CellularAutomata/Buss_hex.vtu
+  Patterns/CellularAutomata/tri_life.vtu
+  Patterns/CellularAutomata/hex_B2oS2m34_gliders.vtu
+  Patterns/CellularAutomata/PenroseTilings/life.vtu
+  Patterns/CellularAutomata/PenroseTilings/life_oscillators.vtu
+  Patterns/CellularAutomata/PenroseTilings/Goucher_glider.vtu
+  Patterns/CellularAutomata/PenroseTilings/Imai_glider_B2SC4.vtu
+  Patterns/CellularAutomata/PenroseTilings/Goucher_loops.vtu
+  Patterns/CellularAutomata/Salt/salt2D_demo.vti
+  Patterns/CellularAutomata/Salt/salt3D_circular330.vti
+  Patterns/Yang2002/Yang_1.vti    Patterns/Yang2002/Yang_2b.vti  Patterns/Yang2002/Yang_2c.vti
+  Patterns/Yang2002/Yang_2d.vti   Patterns/Yang2002/Yang_3a.vti  Patterns/Yang2002/Yang_3b.vti
+  Patterns/Yang2002/Yang_3c.vti   Patterns/Yang2002/Yang_3d.vti  Patterns/Yang2002/Yang_4.vti
+  Patterns/Yang2003/Fig2.vti
+  Patterns/Yang2003/Fig3a.vti     Patterns/Yang2003/Fig3b.vti    Patterns/Yang2003/Fig3c.vti
+  Patterns/McCabe/McCabe.vti               Patterns/McCabe/McCabe_simple.vti
+  Patterns/McCabe/McCabe_additive2b.vti    Patterns/McCabe/McCabe_additive2a.vti
+  Patterns/Kytta2007/Fig5.7a.vti  Patterns/Kytta2007/Fig5.7c.vti
+  Patterns/Kytta2007/Fig5.8c.vti  Patterns/Kytta2007/Fig5.8d.vti Patterns/Kytta2007/Fig5.8e.vti
+  Patterns/Kytta2007/Fig5.8f.vti  Patterns/Kytta2007/Fig5.8g.vti
+  Patterns/Yang2006/jumping.vti   Patterns/Yang2006/jumping_cGL.vti
+  Patterns/Schrodinger1926/packet.vti             Patterns/Schrodinger1926/packet_reflect.vti
+  Patterns/Schrodinger1926/packet_pass.vti        Patterns/Schrodinger1926/quantum_tunnelling.vti
+  Patterns/Schrodinger1926/packet_reflect2D.vti
+  Patterns/Experiments/mutually-catalytic_spots.vti
+  Patterns/Experiments/cglrd_ramps_example_djw.vti  
+  Patterns/Experiments/grayscott-historyWave_fuseWorms.vti
+  Patterns/Experiments/grayscott-historyWave_moreLifelike.vti
+  Patterns/Experiments/grayscott-historyWaveDC_solitonsAndWorms_init.vti
+  Patterns/Experiments/orbits_explodey_init.djw.vti
+  Patterns/Experiments/orbits_sharpWaves-init_djw.vti
+  Patterns/Experiments/grayscott-historyWave_coralGrow_djw.vti
+  Patterns/Experiments/gladman_vermiformSolitons.vti
+)
+
+set( HELP_FILES
+  Help/about.gif              Help/about.html
+  Help/action.html            Help/credits.html 
+  Help/file.html              Help/help.html 
+  Help/mouse.html             Help/quickstart.html 
+  Help/tips.html              Help/changes.html 
+  Help/edit.html              Help/formats.html 
+  Help/index.html             Help/problems.html 
+  Help/view.html              Help/introduction.html
+)
+
+set( OTHER_FILES
+  ./README.txt 
+  ./COPYING.txt 
+  ./TODO.txt 
+  ./BUILD.txt 
+  ./CMakeLists.txt 
+  src/FindOpenCL.cmake 
+  src/Doxyfile.in
+  resources/logo.png
+  resources/Icons/22px/icon-pointer.png
+  resources/Icons/22px/draw-freehand.png
+  resources/Icons/22px/draw-brush.png
+  resources/Icons/22px/color-picker.png
+  resources/Icons/22px/document-new.png
+  resources/Icons/22px/document-open.png
+  resources/Icons/22px/document-save.png
+  resources/Icons/22px/document-revert.png
+  resources/Icons/22px/media-playback-start_green.png
+  resources/Icons/22px/media-playback-pause_red.png
+  resources/Icons/22px/media-seek-forward.png
+  resources/Icons/22px/media-seek-backward.png
+  resources/Icons/22px/media-skip-backward_modified.png
+  resources/Icons/22px/media-record.png
+  resources/Icons/22px/system-run.png
+  resources/Icons/22px/list-add_gray.png
+  resources/Icons/22px/camera-photo.png
+  resources/Icons/32px/icon-pointer.png
+  resources/Icons/32px/draw-freehand.png
+  resources/Icons/32px/draw-brush.png
+  resources/Icons/32px/color-picker.png
+  resources/Icons/32px/document-new.png
+  resources/Icons/32px/document-open.png
+  resources/Icons/32px/document-save.png
+  resources/Icons/32px/document-revert.png
+  resources/Icons/32px/media-playback-start_green.png
+  resources/Icons/32px/media-playback-pause_red.png
+  resources/Icons/32px/media-seek-forward.png
+  resources/Icons/32px/media-seek-backward.png
+  resources/Icons/32px/media-skip-backward_modified.png
+  resources/Icons/32px/media-record.png
+  resources/Icons/32px/system-run.png
+  resources/Icons/32px/list-add_gray.png
+  resources/Icons/32px/camera-photo.png
+  resources/Cursors/pencil-cursor.png
+  resources/Cursors/brush-cursor.png
+  resources/Cursors/picker-cursor.png
+)
+
+#-------------------------------------------VTK----------------------------------------------
+
+find_package( VTK )
+if( VTK_FOUND )
+  include( ${VTK_USE_FILE} )
+else()
+  message(FATAL_ERROR "Cannot build the executable without VTK. Please set the VTK variables.")
+endif()
+
+#-------------------------------------------wxVTK----------------------------------------------
+
+# The following allows you to access wxGLCanvas for GTK
+IF(WIN32)
+  SET(GUI_EXECUTABLE WIN32)
+ELSE(WIN32)
+  IF(APPLE)
+    SET(GUI_EXECUTABLE MACOSX_BUNDLE)
+    IF(VTK_USE_COCOA)
+      SET_SOURCE_FILES_PROPERTIES(
+        src/gui/wxVTKRenderWindowInteractor.cxx
+        PROPERTIES COMPILE_FLAGS "-ObjC++")
+    ENDIF(VTK_USE_COCOA)
+  ELSE(APPLE)
+    # Ok X11 for sure, but just check:
+    IF(NOT VTK_USE_X)
+      MESSAGE(FATAL_ERROR "You need to have VTK_USE_X")
+    ENDIF(NOT VTK_USE_X)
+    # CMake 2.6:
+    # technically those packages are not required since one can still use the Motif/X11 version and not the gtk one:
+    FIND_PACKAGE(PkgConfig)
+    pkg_check_modules (GTK2 gtk+-2.0)
+    #MESSAGE("${GTK2_INCLUDE_DIRS}")
+    INCLUDE_DIRECTORIES(${GTK2_INCLUDE_DIRS})
+    LINK_LIBRARIES(${GTK2_LIBRARIES})
+    # Can I require all my user to have the gl lib on linux, even if they do not really need it...
+    SET(WXGLCANVASLIBS "gl")
+  ENDIF(APPLE)
+ENDIF(WIN32)
+
+#-------------------------------------------wxWidgets----------------------------------------------
+
+if( APPLE )
+  # on Mac OS X it's better to use locally installed wxWidgets headers and libs
+  # (the pre-installed stuff tends to be out of date; eg. 10.6 has wxMac 2.8.8 and it's a 32-bit debug build)
+  set( wxWidgets_CONFIG_EXECUTABLE /usr/local/bin/wx-config )
+  set( wxWidgets_wxrc_EXECUTABLE /usr/local/bin/wxrc )         # not used, but no harm leaving it in
+elseif(UNIX)
+  # remove -rdynamic from link options on Linux to reduce size by about 1.2MB
+  set( CMAKE_SHARED_LIBRARY_LINK_CXX_FLAGS "" )
+endif()
+
+# wxWidgets is required to build the project
+FIND_PACKAGE( wxWidgets COMPONENTS html aui ${WXGLCANVASLIBS} core adv base )
+
+IF( wxWidgets_FOUND )
+  INCLUDE( ${wxWidgets_USE_FILE} )
+ELSE()
+  MESSAGE( FATAL_ERROR "Cannot build the executable without wxWidgets. Please set the wxWidgets variables." )
+ENDIF()
+
+if( WIN32 )
+  # prevent link errors with wxMSW 2.9.x
+  add_definitions( -DwxDEBUG_LEVEL=0 )
+endif()
+
+#-------------------------------------------OpenCL----------------------------------------------
+
+set( CMAKE_MODULE_PATH ${Ready_SOURCE_DIR}/src )
+# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
+
+# we need to build against OpenCL
+find_package( OpenCL REQUIRED )
+include_directories( ${OPENCL_INCLUDE_DIRS} )
+if( APPLE )
+  link_libraries( ${OPENCL_LIBRARIES} ) # on MacOSX we assume that OpenCL is available (might need to rethink for versions before 10.6)
+endif()
+  
+#---------------copy installation files to build folder (helps with testing)--------------------
+
+foreach( file ${PATTERN_FILES} ${HELP_FILES} ${RESOURCES} ${OTHER_FILES} )
+  add_custom_command( 
+    OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/${file}"
+    COMMAND cmake -E copy "${CMAKE_CURRENT_SOURCE_DIR}/${file}" "${CMAKE_CURRENT_BINARY_DIR}/${file}"
+    DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/${file}"
+  )
+  list( APPEND files_dest "${CMAKE_CURRENT_BINARY_DIR}/${file}" )
+endforeach() 
+
+add_custom_target( CopyFiles ALL DEPENDS ${files_dest} )
+
+#-------------------------------- build ------------------------------------------------------
+
+# ensure we link the C runtime statically (N.B. still appears as /MD in the CMake gui but ignore this)
+# see: http://www.cmake.org/Wiki/CMake_FAQ#Dynamic_Replace
+foreach( var CMAKE_CXX_FLAGS CMAKE_CXX_FLAGS_DEBUG CMAKE_CXX_FLAGS_RELEASE CMAKE_CXX_FLAGS_MINSIZEREL CMAKE_CXX_FLAGS_RELWITHDEBINFO )
+  string( REGEX REPLACE "/MD" "/MT" ${var} "${${var}}" )
+endforeach()
+
+set( USE_SSE "YES" )
+if( USE_SSE )
+  # enable SSE to allow us to set flags to avoid denormals, and relax the floating-point accuracy for speed
+  if( MSVC )
+    if( CMAKE_CL_64 )
+      set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} /fp:fast" )
+      set( CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /fp:fast" )
+    else() # avoid warning on x64, which always comes with SSE2
+      set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} /arch:SSE2 /fp:fast" )
+      set( CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:SSE2 /fp:fast" )
+    endif()
+  else()
+    add_definitions( -msse2 -ffast-math )
+  endif()
+  add_definitions( -DUSE_SSE )
+endif()
+
+if( APPLE )
+  # support Mac OS 10.5 or later
+  add_definitions( -mmacosx-version-min=10.5 )
+endif()
+if( APPLE OR UNIX )
+  # use same settings as in makefiles
+  add_definitions( -D_LARGE_FILES )
+endif()
+
+# create base library used by all executables
+add_library( readybase STATIC ${BASE_SOURCES} )
+if( ${VTK_MAJOR_VERSION} GREATER 5 )
+  target_link_libraries( 
+    readybase 
+    vtkCommonCore 
+    vtkFiltersModeling
+    vtkFiltersTexture
+    vtkInteractionStyle 
+    vtkIOXML 
+    vtkRenderingCore 
+    vtkRenderingAnnotation 
+    vtkRenderingFreeType 
+    vtkRenderingFreeTypeOpenGL 
+  )
+else()
+  target_link_libraries( readybase vtkCommon vtkGraphics vtkIO vtkRendering vtkHybrid )
+endif()
+
+# create command-line utility
+add_executable( ${CMD_NAME} ${CMD_SOURCES} )
+target_link_libraries( ${CMD_NAME} readybase )
+
+# create GUI application
+add_executable( ${APP_NAME} ${GUI_EXECUTABLE} ${GUI_SOURCES} ${RESOURCES} )
+target_link_libraries( ${APP_NAME} readybase ${wxWidgets_LIBRARIES} )
+
+if( APPLE )
+  # create Info.plist (using Info.plist.in) and PkgInfo files inside .app bundle
+  add_custom_target( app_bundle
+    COMMAND sed -e "s/VERSION/${READY_VERSION}/" ${CMAKE_SOURCE_DIR}/resources/Info.plist.in >Ready.app/Contents/Info.plist
+    COMMAND echo -n "APPLReDy" >Ready.app/Contents/PkgInfo
+  )
+  add_dependencies( ${APP_NAME} app_bundle )
+   
+  # copy *.icns files into Resources directory inside .app bundle
+  set_source_files_properties( ${CMAKE_SOURCE_DIR}/resources/app.icns PROPERTIES MACOSX_PACKAGE_LOCATION Resources )
+  set_source_files_properties( ${CMAKE_SOURCE_DIR}/resources/file.icns PROPERTIES MACOSX_PACKAGE_LOCATION Resources )
+   
+  # remove unreachable functions and data, and don't add debug info (reduces app size by about 12MB)
+  target_link_libraries( ${APP_NAME} -Wl,-dead_strip -Wl,-S )
+endif()
+
+# (Visual Studio only) put the executable in the root binary folder, not in "Debug" or "Release"
+# http://stackoverflow.com/questions/543203/cmake-runtime-output-directory-on-windows
+if( MSVC_IDE )
+  set_target_properties( ${APP_NAME} PROPERTIES PREFIX "../" )
+  set_target_properties( ${CMD_NAME} PROPERTIES PREFIX "../" )
+endif()
+
+# avoid security warnings
+if( MSVC )
+  add_definitions( /D_CRT_SECURE_NO_WARNINGS /D_CRT_NONSTDC_NO_WARNINGS )
+endif()
+
+if( CMAKE_COMPILER_IS_GNUCXX )
+  list( APPEND CMAKE_EXE_LINKER_FLAGS_RELEASE "-s" )  # strip release binary, for smaller file size
+endif()
+
+#----------------------------------------doxygen------------------------------------------------
+
+find_package( Doxygen )
+if(DOXYGEN_FOUND)
+  configure_file(${CMAKE_CURRENT_SOURCE_DIR}/src/Doxyfile.in ${CMAKE_CURRENT_BINARY_DIR}/Doxyfile @ONLY)
+  add_custom_target(doc
+    ${DOXYGEN_EXECUTABLE} ${CMAKE_CURRENT_BINARY_DIR}/Doxyfile
+    WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
+  )
+endif()
+# (if doxygen is found then 'make doc' should produce html documentation of the source code)
+
+#----------------------------------------install------------------------------------------------
+
+# put Ready in the root of the installation folder instead of in "bin"
+install( TARGETS ${APP_NAME} DESTINATION "." )      # (add ${CMD_NAME} if we want to distribute the command-line version too)
+
+# install our source files, resource files, pattern files, help files and text files
+foreach( source_file ${BASE_SOURCES} ${GUI_SOURCES} ${CMD_SOURCES} ${RESOURCES} ${PATTERN_FILES} ${HELP_FILES} ${OTHER_FILES} )
+  get_filename_component( path_name "${source_file}" PATH )
+  install( FILES "${source_file}" DESTINATION ${path_name} )
+endforeach()
+
+#----------------------------------------package----------------------------------------------
+
+if( APPLE )
+  set( CPACK_SYSTEM_NAME "Mac" )  # nicer than "Darwin"
+elseif( UNIX )
+  if( CMAKE_SIZEOF_VOID_P EQUAL 8 )
+    set( CPACK_SYSTEM_NAME "Linux-64bit" )  
+  else()
+    set( CPACK_SYSTEM_NAME "Linux-32bit" )  
+  endif()
+elseif( WIN32)
+  if( CMAKE_SIZEOF_VOID_P EQUAL 8 )
+    set( CPACK_SYSTEM_NAME "Windows-64bit" )  
+  else()
+    set( CPACK_SYSTEM_NAME "Windows-32bit" )  
+  endif()
+endif()
+if( NOT USE_SSE )
+  set( CPACK_SYSTEM_NAME "${CPACK_SYSTEM_NAME}-noSSE" )
+endif()
+set( CPACK_GENERATOR "ZIP" )
+set( CPACK_PACKAGE_VERSION "${READY_VERSION}" )
+set( CPACK_SOURCE_GENERATOR "ZIP" )
+include( CPack )
diff --git a/ComplexGinzbergLandau/CMakeLists.txt b/ComplexGinzbergLandau/CMakeLists.txt
deleted file mode 100644
index bf6ec5710..000000000
--- a/ComplexGinzbergLandau/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(ComplexGinzbergLandau)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(ComplexGinzbergLandau 
-  complex_ginzberg_landau.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(ComplexGinzbergLandau ${OpenCV_LIBS} )
diff --git a/ComplexGinzbergLandau/complex_ginzberg_landau.cpp b/ComplexGinzbergLandau/complex_ginzberg_landau.cpp
deleted file mode 100644
index 46b521c43..000000000
--- a/ComplexGinzbergLandau/complex_ginzberg_landau.cpp
+++ /dev/null
@@ -1,137 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float D_a,float D_b,float alpha,float beta,float gamma,float delta,
-             float speed);
-
-int main()
-{
-    // Here we implement the complex Ginzberg-Landau model:
-    // http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.31.529&rep=rep1&type=pdf
-
-    // -- parameters --
-    float alpha = 1.0f/16.0f;
-    float beta = 1.0f;
-    float delta = 1.0f; // 2.0f gives unstable spirals
-    float gamma = delta/16.0f;
-    float D_a = 0.2f;
-    float D_b = 0.2f;
-    float speed = 0.2f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,da,db,D_a,D_b,alpha,beta,gamma,delta,speed);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Complex Ginsberg-Landau - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,b,iteration,true,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] = frand(-1.0f,1.0f);
-            b[i][j] = frand(-1.0f,1.0f);
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float D_a,float D_b,float alpha,float beta,float gamma,float delta,
-             float speed)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        int iprev = (i + X - 1) % X;
-        int inext = (i + 1) % X;
-
-        for(int j = 0; j < Y; j++) 
-        {
-            int jprev = (j + Y - 1) % Y;
-            int jnext = (j + 1) % Y;
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = D_a * dda + alpha*aval - gamma*bval + (-beta*aval + delta*bval)*(aval*aval+bval*bval);
-            db[i][j] = D_b * ddb + alpha*bval + gamma*aval + (-beta*bval - delta*aval)*(aval*aval+bval*bval);
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
-
diff --git a/Display/defs.h b/Display/defs.h
deleted file mode 100644
index 62feba110..000000000
--- a/Display/defs.h
+++ /dev/null
@@ -1,3 +0,0 @@
-// the size of the world:
-#define X 256
-#define Y 256
diff --git a/Display/display.cpp b/Display/display.cpp
deleted file mode 100644
index 8f410638e..000000000
--- a/Display/display.cpp
+++ /dev/null
@@ -1,129 +0,0 @@
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib
-#include <stdio.h>
-
-// local:
-#include "display.h"
-
-bool display(float r[X][Y],float g[X][Y],float b[X][Y],
-             int iteration,bool auto_brighten,float manual_brighten,
-			 int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-	static bool write_video = false;
-
-    static IplImage *im,*im2,*im3;
-	static int border = 0;
-    static CvFont font;
-	static CvVideoWriter *video;
-	static const CvScalar white = cvScalar(255,255,255);
-
-	const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-        im3 = cvCreateImage(cvSize(X*scale+border*2,Y*scale+border),IPL_DEPTH_8U,3);
-        
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-        
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-		if(write_video)
-		{
-			video = cvCreateVideoWriter(title,CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im3),1);
-			border = 20;
-		}
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<X;i++)
-        {
-            for(int j=0;j<Y;j++)
-            {
-				if(r) {
-					val = r[i][j];
-					if(val<minR) minR=val; if(val>maxR) maxR=val;
-				}
-				if(g) {
-					val = g[i][j];
-					if(val<minG) minG=val; if(val>maxG) maxG=val;
-				}
-				if(b) {
-					val = b[i][j];
-					if(val<minB) minB=val; if(val>maxB) maxB=val;
-				}
-            }
-        }
-    }
-    for(int i=0;i<X;i++)
-    {
-        for(int j=0;j<Y;j++)
-        {
-			if(r) {
-				float val = r[i][Y-j-1];
-				if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-				else val *= manual_brighten;
-				if(val<0) val=0; if(val>255) val=255;
-				((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 2] = (uchar)val;
-			}
-			if(g) {
-				float val = g[i][Y-j-1];
-				if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-				else val *= manual_brighten;
-				if(val<0) val=0; if(val>255) val=255;
-				((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 1] = (uchar)val;
-			}
-			if(b) {
-				float val = b[i][Y-j-1];
-				if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-				else val *= manual_brighten;
-				if(val<0) val=0; if(val>255) val=255;
-				((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 0] = (uchar)val;
-			}
-        }
-    }
-
-    cvResize(im,im2);
-	cvCopyMakeBorder(im2,im3,cvPoint(border*2,0),IPL_BORDER_CONSTANT);
-
-	char txt[100];
-	if(!write_video)
-	{
-		sprintf(txt,"%d",iteration);
-		cvPutText(im3,txt,cvPoint(20,20),&font,white);
-	}
-
-	cvPutText(im3,message,cvPoint(20,40),&font,white);
-
-	if(write_video)
-		cvWriteFrame(video,im3);
-
-    cvShowImage(title,im3);
-    
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-		if(write_video)
-			cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
-
diff --git a/Display/display.h b/Display/display.h
deleted file mode 100644
index 91de43969..000000000
--- a/Display/display.h
+++ /dev/null
@@ -1,5 +0,0 @@
-#include "defs.h"
-
-bool display(float r[X][Y],float g[X][Y],float b[X][Y],
-             int iteration,bool auto_brighten,float manual_brighten,
-			 int scale,int delay_ms,const char* message);
diff --git a/EdblomOrbanEpstein/CMakeLists.txt b/EdblomOrbanEpstein/CMakeLists.txt
deleted file mode 100644
index 489a331ee..000000000
--- a/EdblomOrbanEpstein/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(EdblomOrbanEpstein)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(EdblomOrbanEpstein 
-  edblom_orban_epstein.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(EdblomOrbanEpstein ${OpenCV_LIBS} )
diff --git a/EdblomOrbanEpstein/edblom_orban_epstein.cpp b/EdblomOrbanEpstein/edblom_orban_epstein.cpp
deleted file mode 100644
index 378cf95b7..000000000
--- a/EdblomOrbanEpstein/edblom_orban_epstein.cpp
+++ /dev/null
@@ -1,173 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float A,float B,float C,float D,float F,
-             float speed,
-             bool parameter_space);
-
-int main()
-{
-    // http://www.robinengelhardt.info/speciale/
-
-    // -- parameters --
-    float A=17.00f;
-    float B=1.0f;
-    float C=1.0f;
-    float D=1.39f;
-    float F=7.65f;
-    float speed = 0.01f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,da,db,A,B,C,D,F,speed,false);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"EOE - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(b,b,b,iteration,true,200.0f,4,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            //a[i][j] = frand(0.0f,17.0f);
-            //b[i][j] = frand(0.0f,17.0f);
-            //if(abs(i-X/2)<2) // spreading from a vertical strip
-            if((abs(j-Y/2)<20 && abs(i-X/2+1)<7) || (abs(j-Y/2)>=20 && abs(i-X/2)<7) ) // spreading from a vertical strip with a 1 pixel wiggle
-            //if(abs(i-X/2)>10 && j>Y-2) // spreading from a broken horizontal strip at the top edge
-            //if(abs(i-X/2)>10 && j>Y-14) // spreading from a broken horizontal strip at the top edge
-            //if(abs(j-Y/2)<7) // spreading from a horizontal strip in the middle
-            {
-                a[i][j]=4.4f;
-                //b[i][j]=3.27f;
-                b[i][j]=4.5f;
-            }
-            else
-            {
-                a[i][j]=5.80f;
-                b[i][j]=1.85f;
-            }
-        }
-    }
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float A,float B,float C,float D,float F,
-             float speed,
-             bool parameter_space)
-{
-    const bool toroidal = true;
-
-    int iprev,inext,jprev,jnext;
-
-    // compute change in each cell
-    for(int i = 0; i < X; i++) {
-        if(toroidal) {
-            iprev = (i + X - 1) % X;
-            inext = (i + 1) % X;
-        }
-        else {
-            iprev = max(0,i-1);
-            inext = min(X-1,i+1);
-        }
-
-        for(int j = 0; j < Y; j++) {
-            if(toroidal) {
-                jprev = (j + Y - 1) % Y;
-                jnext = (j + 1) % Y;
-            }
-            else {
-                jprev = max(0,j-1);
-                jnext = min(Y-1,j+1);
-            }
-
-            if(parameter_space)	{
-                /*const float kmin=0.045f,kmax=0.07f,fmin=0.00f,fmax=0.14f;
-                // set f and k for this location (ignore the provided values of f and k)
-                k = kmin + i*(kmax-kmin)/X;
-                f = fmin + j*(fmax-fmin)/Y;*/
-            }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] =  C * ( - aval*bval*bval + A*bval - (1+B)*aval ) + D*dda;
-            db[i][j] = (1/C) * ( aval*bval*bval - (1+A)*bval + aval + F ) + ddb;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
-
diff --git a/FitzHughNagumo/CMakeLists.txt b/FitzHughNagumo/CMakeLists.txt
deleted file mode 100644
index 529e22336..000000000
--- a/FitzHughNagumo/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(FitzHughNagumo)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(FitzHughNagumo
-  fitz_hugh_nagumo.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(FitzHughNagumo ${OpenCV_LIBS} )
diff --git a/FitzHughNagumo/fitz_hugh_nagumo.cpp b/FitzHughNagumo/fitz_hugh_nagumo.cpp
deleted file mode 100644
index b9882ada6..000000000
--- a/FitzHughNagumo/fitz_hugh_nagumo.cpp
+++ /dev/null
@@ -1,243 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float a0,float a1,float epsilon,float delta,float k1,float k2,float k3,
-             float speed,
-             bool parameter_map);
-
-int main()
-{
-    // -- parameters --
-
-    // From Hagberg and Meron:
-    // http://arxiv.org/pdf/patt-sol/9401002
-
-    // for tip-splitting:
-    float a0 = -0.1f;
-    float a1 = 2.0f;
-    float epsilon = 0.05f;
-    float delta = 4.0f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;
-    bool spiral_waves = false;
-
-    // for spiral turbulence:
-    /*float a0 = -0.1f;
-    float a1 = 2.0f;
-    float epsilon = 0.014f;
-    float delta = 2.8f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;
-    bool spiral_waves = false;*/
-
-    // for spiral waves: http://thevirtualheart.org/java/2dfhn.html
-    /*float a0 = 0.0f;
-    float a1 = 1.0f;
-    float epsilon = 0.01f;
-    float delta = 0.0f;
-    float k1 = -0.1f;
-    float k2 = -1.1f;
-    float k3 = 0.5f;
-    bool spiral_waves = true;*/
-
-    // from Malevanets and Kapral (can't get these to work)
-
-    // for labyrinth:
-    /*float a0 = 0.146f;
-    float a1 = 3.05f;
-    float epsilon = 0.017f;
-    float delta = 4.0f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;*/
-    
-    // for bloch fronts:
-    /*float a0 = 0.0f;
-    float a1 = 4.88f;
-    float epsilon = 0.084f;
-    float delta = 0.0f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;*/
- 
-    float speed = 0.05f;
-    bool parameter_map = false;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,da,db,a0,a1,epsilon,delta,k1,k2,k3,speed,parameter_map);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"FitzHugh-Nagumo - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,b,b,iteration,true,200.0f,3,10,msg)) // did user ask to quit?
-            break;
-
-        // to make more interesting patterns we periodically reset part of the grid
-        if(spiral_waves && (iteration==1000 || (iteration>0 && iteration%2000==0)) )
-        {
-            int div = (int)(rand()*Y/(float)RAND_MAX);
-            for(int i = 0; i < X; i++) 
-            {
-                for(int j = 0; j < Y; j++) 
-                {
-                    if(j<div)
-                    {
-                        a[i][j] = 0.0f;
-                        b[i][j] = 0.0f;
-                    }
-                }
-            }
-        }
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    float pos;
-    for(int i = 0; i < X; i++) 
-    {
-        if(i%30==0)
-            pos = frand(-6.0f,6.0f);
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] = 0.0f;
-            b[i][j] = 0.0f;
-            /*if(i>10 && i<20) // start with a wave on the left edge
-            {
-                a[i][j] = 1.0f;
-            }
-            if(i>1 && i<17)
-            {
-                b[i][j] = 0.1f;
-            }*/
-            if(abs(i-X/2)<X/3 && abs(j-Y/2+pos)<10) // start with a wavy static line in the centre
-            //if(i<10)
-            {
-                a[i][j] = 1.0f;
-                b[i][j] = 0.0f;
-            }
-            // you can start from random conditions too, but it's harder to guarantee
-            // interesting waves
-            if(false)
-            {
-                a[i][j] = frand(-0.1f,0.1f);
-                b[i][j] = frand(-0.1f,0.1f);
-            }
-        }
-    }
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float a0,float a1,float epsilon,float delta,float k1,float k2,float k3,
-             float speed,
-             bool parameter_map)
-{
-    int iprev,inext,jprev,jnext;
-    bool toroidal = false;
-
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        if(toroidal) { iprev = (i + X - 1) % X; inext = (i + 1) % X; }
-        else { iprev=max(0,i-1); inext=min(X-1,i+1); }
-
-        for(int j = 0; j < Y; j++) 
-        {
-            if(toroidal) { jprev = (j + Y - 1) % Y; jnext = (j + 1) % Y; }
-            else { jprev=max(0,j-1); jnext=min(Y-1,j+1); }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            if(parameter_map)
-            {
-                const float min_a0=-0.6f,max_a0=0.6f;
-                const float min_a1=-2.0f,max_a1=10.0f;
-                const float min_epsilon=0.0f,max_epsilon=0.2f;
-                const float min_delta=0.0f,max_delta=5.0f;
-                //a0 = min_a0 + (max_a0-min_a0) * i/X;
-                //a1 = min_a1 + (max_a1-min_a1) * j/Y;
-                epsilon = min_epsilon + (max_epsilon-min_epsilon) * i/X;
-                delta = min_delta + (max_delta-min_delta) * j/Y;
-            }
-
-            // compute the new rate of change of a and b
-            da[i][j] = k1*aval - k2*aval*aval - aval*aval*aval - bval + dda;
-            db[i][j] = epsilon * (k3*aval - a1*bval - a0) + delta*ddb;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
-
diff --git a/FitzHughNagumo/fitz_hugh_nagumo2.cpp b/FitzHughNagumo/fitz_hugh_nagumo2.cpp
deleted file mode 100644
index 2f8071997..000000000
--- a/FitzHughNagumo/fitz_hugh_nagumo2.cpp
+++ /dev/null
@@ -1,209 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-Specifically following Wikipedia
-http://en.wikipedia.org/wiki/File:Reaction_diffusion_spiral.gif
-
-*/
-
-// Original copyright notice:
-
-/*
-
-Make spots and stripes with reaction-diffusion.
-
-The spot-formation system is described in the article:
-
-  "A Model for Generating Aspects of Zebra and Other Mammailian
-   Coat Patterns"
-  Jonathan B. L. Bard
-  Journal of Theoretical Biology, Vol. 93, No. 2, pp. 363-385
-  (November 1981)
-
-The stripe-formation system is described in the book:
-
-  Models of Biological Pattern Formation
-  Hans Meinhardt
-  Academic Press, 1982
-
-
-Permission is granted to modify and/or distribute this program so long
-as the program is distributed free of charge and this header is retained
-as part of the program.
-
-Copyright (c) Greg Turk, 1991
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float tau,float du2,float dv2,float kappa,float lambda,float sigma,
-             float speed,
-			 bool parameter_map);
-
-int main()
-{
-    // -- parameters --
-
-	// hex:
-	float tau = 0.1f;
-	float du2 = 0.00024f * 10.0f;
-	float dv2 = 0.005f * 10.0f;
-	float kappa = -0.05f;
-	float lambda = 1.0f;
-	float sigma = 1.0f;
-
-	// for spiral waves
-	/*float tau = 4.0f;
-	float du2 = 0.0015f * 10.0f;
-	float dv2 = 0.01f * 10.0f;
-	float kappa = -1.126f;
-	float lambda = 4.67f;
-	float sigma = -3.33f;*/
-
-	float speed = 0.001f;
-	bool parameter_map = false;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-    int iteration = 0;
-    while(true) 
-	{
-        // compute:
-        compute(a,b,da,db,tau,du2,dv2,kappa,lambda,sigma,speed,parameter_map);
-
-        // display:
-        if(iteration%100==0) 
-		{
-            if(display(a,b,b,iteration,true,200.0f,1,10,"FitzHughNagumo.avi")) // did user ask to quit?
-                break;
-        }
-
-		// to make more interesting patterns we periodically reset part of the grid
-		/*if(iteration==300 || (iteration>0 && iteration%3000==0))
-		{
-			int div = rand()*Y/(float)RAND_MAX;
-			for(int i = 0; i < X; i++) 
-			{
-				for(int j = 0; j < Y; j++) 
-				{
-					if(j<div)
-					{
-						a[i][j] = 0.0f;
-						b[i][j] = 0.0f;
-					}
-				}
-			}
-		}*/
-
-        iteration++;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-	float pos;
-    for(int j = 0; j < Y; j++) 
-	{
-		if(j%10==0)
-			pos = frand(8.0f,12.0f);
-	    for(int i = 0; i < X; i++) 
-		{
-			a[i][j] = 0.0f;
-			b[i][j] = 0.0f;
-			//if(abs(i-X/2+pos)<10) // start with a wavy static line in the centre
-			if(i>10 && i<20) // start with a wave on the left edge
-			{
-				a[i][j] = 1.0f;
-			}
-			if(i>1 && i<17)
-			{
-				b[i][j] = 0.1f;
-			}
-			// you can start from random conditions too, but it's harder to guarantee
-			// interesting waves
-		    a[i][j] = frand(-0.1f,0.1f);
-		    b[i][j] = frand(-0.1f,0.1f);
-        }
-    }
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float tau,float du2,float dv2,float kappa,float lambda,float sigma,
-			 float speed,
-			 bool parameter_map)
-{
-	int iprev,inext,jprev,jnext;
-	bool toroidal = false;
-
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-	{
-		if(toroidal) { iprev = (i + X - 1) % X; inext = (i + 1) % X; }
-		else { iprev=max(0,i-1); inext=min(X-1,i+1); }
-
-        for(int j = 0; j < Y; j++) 
-		{
-			if(toroidal) { jprev = (j + Y - 1) % Y; jnext = (j + 1) % Y; }
-			else { jprev=max(0,j-1); jnext=min(Y-1,j+1); }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-			// compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-			if(parameter_map)
-			{
-			}
-
-			// compute the new rate of change of a and b
-			da[i][j] = lambda*aval - aval*aval*aval - kappa - sigma*bval + du2*dda;
-			db[i][j] = (aval - bval - dv2*ddb ) / tau;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
diff --git a/FitzHughNagumo3D/CMakeLists.txt b/FitzHughNagumo3D/CMakeLists.txt
deleted file mode 100644
index 62432f33c..000000000
--- a/FitzHughNagumo3D/CMakeLists.txt
+++ /dev/null
@@ -1,17 +0,0 @@
-project(FitzHughNagumo3D)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR} )
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(FitzHughNagumo3D
-  fitz_hugh_nagumo_3d.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(FitzHughNagumo3D 
-  ${OpenCV_LIBS} 
-  ${VTK_LIBS} 
-)
diff --git a/FitzHughNagumo3D/fitz_hugh_nagumo_3d.cpp b/FitzHughNagumo3D/fitz_hugh_nagumo_3d.cpp
deleted file mode 100644
index 91202d788..000000000
--- a/FitzHughNagumo3D/fitz_hugh_nagumo_3d.cpp
+++ /dev/null
@@ -1,390 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <math.h>
-#include <stdio.h>
-
-// STL:
-#include <vector>
-using namespace std;
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-bool display(float *r,float *g,float *b,int S,
-             int iteration,bool auto_brighten,float manual_brighten,
-             float scale,int delay_ms,const char* window_title)
-{
-    static bool need_init = true;
-    
-    static bool write_video = true;
-    static bool is_parameter_map = false;
-
-    static IplImage *im,*im2,*im3;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    int k=S/2;
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(S,S),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(int(S*scale),int(S*scale)),IPL_DEPTH_8U,3);
-        
-        cvNamedWindow(window_title,CV_WINDOW_AUTOSIZE);
-        
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-        if(is_parameter_map)
-        {
-            border = 20;
-        }
-
-        im3 = cvCreateImage(cvSize(int(S*scale+border*2),int(S*scale+border)),IPL_DEPTH_8U,3);
-
-        if(write_video)
-        {
-            float fps=30.0f;
-            // try to find a codec that works
-            video = cvCreateVideoWriter("out.avi",CV_FOURCC('D','I','V','X'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('M','J','P','G'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('P','I','M','1'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('P','I','C','1'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('M','P','4','2'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('D','I','V','3'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('V','P','3','1'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('V','P','3','0'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('C','V','I','D'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",CV_FOURCC('f','f','d','s'),fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.mpg",CV_FOURCC_DEFAULT,fps,cvGetSize(im3));
-            if(video==NULL)
-                video = cvCreateVideoWriter("out.avi",-1,fps,cvGetSize(im3)); // ask usr to choose
-            if(video==NULL)
-                video = cvCreateVideoWriter("im_00001.png",0,fps,cvGetSize(im3)); // fall back on image output
-            // to get video output working on linux, you may need to recompile opencv to include ffmpeg,
-            // as described here: http://opencv.willowgarage.com/wiki/InstallGuide (I did on Ubuntu 10.04)
-        }
-
-    }
-
-    if(!im) return true;
-
-    // convert float arrays to IplImage for OpenCV to display
-    float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<S;i++)
-        {
-            for(int j=0;j<S;j++)
-            {
-                if(r) {
-                    val = r[(i*S+j)*S+k];
-                    if(val<minR) minR=val; if(val>maxR) maxR=val;
-                }
-                if(g) {
-                    val = g[(i*S+j)*S+k];
-                    if(val<minG) minG=val; if(val>maxG) maxG=val;
-                }
-                if(b) {
-                    val = b[(i*S+j)*S+k];
-                    if(val<minB) minB=val; if(val>maxB) maxB=val;
-                }
-            }
-        }
-    }
-    float rangeR = max(0.001f,maxR-minR);
-    float rangeG = max(0.001f,maxG-minG);
-    float rangeB = max(0.001f,maxB-minB);
-    for(int i=0;i<S;i++)
-    {
-        for(int j=0;j<S;j++)
-        {
-            if(r) {
-                val = r[(i*S+(S-j-1))*S+k];
-                if(auto_brighten) val = 255.0f * (val-minR) / rangeR;
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 2] = (uchar)val;
-            }
-            if(g) {
-                val = g[(i*S+(S-j-1))*S+k];
-                if(auto_brighten) val = 255.0f * (val-minG) / rangeG;
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 1] = (uchar)val;
-            }
-            if(b) {
-                val = b[(i*S+(S-j-1))*S+k];
-                if(auto_brighten) val = 255.0f * (val-minB) / rangeB;
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 0] = (uchar)val;
-            }
-        }
-    }
-
-    cvResize(im,im2);
-    cvCopyMakeBorder(im2,im3,cvPoint(border*2,0),IPL_BORDER_CONSTANT);
-
-    char txt[100];
-
-    // show the iteration count
-    sprintf(txt,"%d",iteration);
-    cvPutText(im3,txt,cvPoint(20,20),&font,white);
-
-    if(!write_video)
-    {
-        if(auto_brighten)
-        {
-            // show the range of chemical concentrations
-            sprintf(txt,"chemical 1 range: %.4f - %.4f",minR,maxR);
-            cvPutText(im3,txt,cvPoint(20,60),&font,white);
-            sprintf(txt,"chemical 2 range: %.4f - %.4f",minG,maxG);
-            cvPutText(im3,txt,cvPoint(20,80),&font,white);
-            sprintf(txt,"chemical 3 range: %.4f - %.4f",minB,maxB);
-            cvPutText(im3,txt,cvPoint(20,100),&font,white);
-        }
-    }
-
-    if(is_parameter_map)
-    {
-        // you'll need to change these labels to your needs
-        cvPutText(im3,"0.14",cvPoint(5,15),&font,white);
-        cvPutText(im3,"F",cvPoint(5,im2->height/2),&font,white);
-        cvPutText(im3,"0.00",cvPoint(5,im2->height),&font,white);
-        cvPutText(im3,"0.045",cvPoint(border*2-10,im2->height+15),&font,white);
-        cvPutText(im3,"K",cvPoint(border*2+im2->width/2,im2->height+15),&font,white);
-        cvPutText(im3,"0.07",cvPoint(im3->width-40,im2->height+15),&font,white);
-    }
-
-    if(write_video)
-        cvWriteFrame(video,im3);
-
-    cvShowImage(window_title,im3);
-    
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(window_title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        if(write_video)
-            cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
-
-int main()
-{
-    // -- parameters --
-
-    // From Hagberg and Meron:
-    // http://arxiv.org/pdf/patt-sol/9401002
-
-    // tip-splitting
-    float a0 = -0.1f;
-    float a1 = 2.0f;
-    float epsilon = 0.05f;
-    float delta = 4.0f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;
-    bool spiral_waves = false;
-
-    // for spiral turbulence:
-    /*float a0 = -0.1f;
-    float a1 = 2.0f;
-    float epsilon = 0.014f;
-    float delta = 2.8f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;
-    bool spiral_waves = false;*/
-
-    // for spiral waves: http://thevirtualheart.org/java/2dfhn.html
-    /*float a0 = 0.0f;
-    float a1 = 1.0f;
-    float epsilon = 0.01f;
-    float delta = 0.0f;
-    float k1 = -0.1f;
-    float k2 = -1.1f;
-    float k3 = 0.5f;
-    bool spiral_waves = true;
-    */
-
-    // from Malevanets and Kapral (can't get these to work)
-
-    // for labyrinth:
-    /*float a0 = 0.146f;
-    float a1 = 3.05f;
-    float epsilon = 0.017f;
-    float delta = 4.0f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;*/
-    
-    // for bloch fronts:
-    /*float a0 = 0.0f;
-    float a1 = 4.88f;
-    float epsilon = 0.084f;
-    float delta = 0.0f;
-    float k1 = 1.0f;
-    float k2 = 0.0f;
-    float k3 = 1.0f;*/
- 
-    float speed = 0.02f;
-    bool parameter_map = false;
-    // ----------------
-
-    const int S = 50;
-
-    // these arrays store the chemical concentrations:
-    float *a = new float[S*S*S],*b=new float[S*S*S];
-    // these arrays store the rate of change of those chemicals:
-    float *da = new float[S*S*S],*db=new float[S*S*S];
-
-    // put the initial conditions into each cell
-    srand((unsigned int)time(NULL));
-    for(int i = 0; i < S; i++) 
-    {
-        for(int j = 0; j < S; j++) 
-        {
-            for(int k = 0; k < S; k++) 
-            {
-                a[(i*S+j)*S+k] = 0.0f;
-                b[(i*S+j)*S+k] = 0.0f;
-                float x=i,y=j,z=k;
-                // rotate
-                {
-                    const float PI=3.1415926535f;
-                    float a1=PI/10,a2=PI/9,a3=PI/12; // angles
-                    x = i * cos(a2)*cos(a3) + j * (-cos(a1)*sin(a3)+sin(a1)*sin(a2)*cos(a3)) + k * (sin(a1)*sin(a3)+cos(a1)*sin(a2)*cos(a3));
-                    y = i * cos(a2)*sin(a3) + j * (cos(a1)*cos(a3)+sin(a1)*sin(a2)*sin(a3)) + k * (-sin(a1)*cos(a3)+cos(a1)*sin(a2)*sin(a3));
-                    z = i*(-sin(a2)) + j * (sin(a1)*cos(a2)) + k * (cos(a1)*cos(a2));
-                }
-                if( abs(x-S/2)<S/4 && abs(y-S/3)<4 && abs(z-S/4)<4 ) // start with a static line in the centre
-                {
-                    a[(i*S+j)*S+k] = frand(0.0f,1.0f);
-                    b[(i*S+j)*S+k] = 0.0f;
-                }
-            }
-        }
-    }
-    
-    int iteration = 0;
-    int output_frame = 0;
-    while(true) 
-    {
-        // compute:
-        int iprev,inext,jprev,jnext,kprev,knext;
-        bool toroidal = false;
-
-        // compute change in each cell
-        for(int i = 0; i < S; i++) 
-        {
-            if(toroidal) { iprev = (i + S - 1) % S; inext = (i + 1) % S; }
-            else { iprev=max(0,i-1); inext=min(S-1,i+1); }
-
-            for(int j = 0; j < S; j++) 
-            {
-                if(toroidal) { jprev = (j + S - 1) % S; jnext = (j + 1) % S; }
-                else { jprev=max(0,j-1); jnext=min(S-1,j+1); }
-
-                for(int k = 0; k < S; k++) 
-                {
-                    if(toroidal) { kprev = (k + S - 1) % S; knext = (k + 1) % S; }
-                    else { kprev=max(0,k-1); knext=min(S-1,k+1); }
-
-                    float aval = a[(i*S+j)*S+k];
-                    float bval = b[(i*S+j)*S+k];
-
-                    // compute the Laplacians of a and b
-                    float dda = a[(i*S+jprev)*S+k] + a[(i*S+jnext)*S+k] + a[(iprev*S+j)*S+k] + a[(inext*S+j)*S+k] + a[(i*S+j)*S+kprev] + a[(i*S+j)*S+knext] - 6*aval;
-                    float ddb = b[(i*S+jprev)*S+k] + b[(i*S+jnext)*S+k] + b[(iprev*S+j)*S+k] + b[(inext*S+j)*S+k] + b[(i*S+j)*S+kprev] + b[(i*S+j)*S+knext] - 6*bval;
-
-                    // compute the new rate of change of a and b
-                    da[(i*S+j)*S+k] = k1*aval - k2*aval*aval - aval*aval*aval - bval + dda;
-                    db[(i*S+j)*S+k] = epsilon * (k3*aval - a1*bval - a0) + delta*ddb;
-                }
-            }
-        }
-
-        // effect change
-        for(int i = 0; i < S; i++) {
-            for(int j = 0; j < S; j++) {
-                for(int k = 0; k < S; k++) {
-                    a[(i*S+j)*S+k] += (speed * da[(i*S+j)*S+k]);
-                    b[(i*S+j)*S+k] += (speed * db[(i*S+j)*S+k]);
-                }
-            }
-        }
-
-        // display
-        if(iteration%50==0)
-        {
-            if(display(a,b,b,S,iteration,true,1.0f,3.0f,10,"FHN 3D (Esc to quit)"))
-                break;
-
-            // output volume in VTK format
-            {
-                char txt[100];
-                sprintf(txt,"vol_%05d.vtk",output_frame);
-                FILE *out = fopen(txt,"wt");
-                fprintf(out,"# vtk DataFile Version 2.0\nVolume example\nASCII\nDATASET STRUCTURED_POINTS\nDIMENSIONS %d %d %d\n",S,S,S);
-                fprintf(out,"ASPECT_RATIO 1 1 1\nORIGIN 0 0 0\nPOINT_DATA %d\nSCALARS volume_scalars float 1\nLOOKUP_TABLE default\n",S*S*S);
-                for(int i=0;i<S*S*S;i++)
-                    fprintf(out,"%.3f ",a[i]);
-                fclose(out);
-                output_frame++;
-            }
-        }
-
-        iteration++;
-    }
-
-    // clean up
-    delete []a;
-    delete []b;
-    delete []da;
-    delete []db;
-}
diff --git a/FitzHughNagumo3D/render_volume.py b/FitzHughNagumo3D/render_volume.py
deleted file mode 100644
index d488090a4..000000000
--- a/FitzHughNagumo3D/render_volume.py
+++ /dev/null
@@ -1,81 +0,0 @@
-#!/usr/bin/env python
-
-import vtk
-import os
-
-# create a rendering window and renderer
-ren = vtk.vtkRenderer()
-renWin = vtk.vtkRenderWindow()
-renWin.AddRenderer(ren)
-renWin.SetSize(300,300)
-iren = vtk.vtkRenderWindowInteractor()
-iren.SetRenderWindow(renWin)
-
-# read the data
-sp = vtk.vtkStructuredPointsReader()
-build_folder = "../../../build/"
-sp.SetFileName(build_folder + "FitzHughNagumo3D/vol_00000.vtk")
-sp.Update()
-dims = sp.GetOutput().GetDimensions()
-
-# contour
-cont = vtk.vtkContourFilter()
-cont.SetInput(sp.GetOutput())
-cont.SetValue(0,0.5)
-
-# make an actor
-mapper = vtk.vtkPolyDataMapper()
-mapper.SetInput(cont.GetOutput())
-actor = vtk.vtkActor()
-actor.SetMapper(mapper)
-actor.GetProperty().SetDiffuse(0.6)
-actor.GetProperty().SetAmbient(0.1)
-actor.GetProperty().SetSpecular(0.5)
-actor.GetProperty().SetSpecularPower(100.0)
-
-# assign our actor to the renderer
-ren.AddActor(actor)
-
-# add a cube outline actor
-cube = vtk.vtkCubeSource()
-cube.SetXLength(dims[0])
-cube.SetYLength(dims[1])
-cube.SetZLength(dims[2])
-cube.SetCenter(dims[0]/2,dims[1]/2,dims[2]/2)
-cube_mapper = vtk.vtkPolyDataMapper()
-cube_mapper.SetInput(cube.GetOutput())
-cube_actor = vtk.vtkActor()
-cube_actor.SetMapper(cube_mapper)
-cube_actor.GetProperty().SetRepresentationToWireframe()
-cube_actor.GetProperty().SetAmbient(1)
-ren.AddActor(cube_actor)
-
-wif = vtk.vtkWindowToImageFilter()
-wif.SetInput(renWin)
-png = vtk.vtkPNGWriter()
-png.SetInput(wif.GetOutput())
-
-iren.Initialize()
-renWin.Render()
-iren.Start()
-
-#pov = vtk.vtkPOVExporter()
-#pov.SetRenderWindow(renWin)
-#pov.SetFileName("out.pov")
-#pov.Write()
-
-exit()
-
-# after user has positioned the scene, run through the frames
-renWin.OffScreenRenderingOn()
-i=0
-while True:
-    filename = build_folder + "FitzHughNagumo3D/vol_"+'0'*(5-len(str(i)))+str(i)+".vtk"
-    if not os.path.exists(filename):
-        break
-    sp.SetFileName(filename)
-    renWin.Render()
-    wif.Modified()
-    png.SetFileName("out_"+str(i)+".png")
-    png.Write()
-    i+=1
diff --git a/GrayScott/CMakeLists.txt b/GrayScott/CMakeLists.txt
deleted file mode 100644
index a9986d64a..000000000
--- a/GrayScott/CMakeLists.txt
+++ /dev/null
@@ -1,15 +0,0 @@
-project(GrayScott)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(GrayScott 
-  gray_scott.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-  ../Display/defs.h
-)
-
-TARGET_LINK_LIBRARIES(GrayScott ${OpenCV_LIBS} )
diff --git a/GrayScott/gray_scott.cpp b/GrayScott/gray_scott.cpp
deleted file mode 100644
index 686f2b567..000000000
--- a/GrayScott/gray_scott.cpp
+++ /dev/null
@@ -1,179 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float r_a,float r_b,float f,float k,
-             float speed);
-
-int main()
-{
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    // for spots:
-    float k = 0.064f;
-    float f = 0.035f;
-    // for stripes:
-    //float k = 0.06f;
-    //float f = 0.035f;
-    // for long stripes
-    //float k = 0.065f;
-    //float f = 0.056f;
-    // for dots and stripes
-    //float k = 0.064f;
-    //float f = 0.04f;
-    // for spiral waves:
-    //float k = 0.0475f;
-    //float f = 0.0118f;
-    float speed = 1.0f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-    clock_t start,end;
-
-    const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-        start = clock();
-
-        // compute:
-        for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-        {
-            compute(a,b,da,db,r_a,r_b,f,k,speed);
-            iteration++;
-        }
-
-        end = clock();
-
-        char msg[1000];
-        sprintf(msg,"GrayScott - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,a,iteration,false,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            
-            //if(hypot(i%10-5/*-X/2*/,j%10-5/*-Y/2*/)<=2.0f)//frand(2,3))
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5)) // start with a uniform field with an approximate circle in the middle
-            {
-                a[i][j] = 0.0f;
-                b[i][j] = 1.0f;
-            }
-            else {
-                a[i][j] = 1;
-                b[i][j] = 0;
-            }
-            //float v = frand(0.0f,1.0f);
-            //a[i][j] = v;
-            //b[i][j] = 1.0f-v;
-            //a[i][j] += frand(-0.01f,0.01f);
-            //b[i][j] += frand(-0.01f,0.01f);
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float r_a,float r_b,float f,float k,float speed)
-{
-    //const bool toroidal = false;
-
-    //int iprev,inext,jprev,jnext;
-
-    // compute change in each cell
-    for(int i = 0; i < X; i++) {
-        int iprev,inext;
-        /*if(toroidal) {
-            iprev = (i + X - 1) % X;
-            inext = (i + 1) % X;
-        }
-        else*/ {
-            iprev = max(0,i-1);
-            inext = min(X-1,i+1);
-        }
-
-        for(int j = 0; j < Y; j++) {
-            int jprev,jnext;
-            /*if(toroidal) {
-                jprev = (j + Y - 1) % Y;
-                jnext = (j + 1) % Y;
-            }
-            else*/ {
-                jprev = max(0,j-1);
-                jnext = min(Y-1,j+1);
-            }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = r_a * dda - aval*bval*bval + f*(1-aval);
-            db[i][j] = r_b * ddb + aval*bval*bval - (f+k)*bval;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += speed * da[i][j];
-            b[i][j] += speed * db[i][j];
-        }
-    }
-}
-
diff --git a/Ready/Help/about.gif b/Help/about.gif
similarity index 100%
rename from Ready/Help/about.gif
rename to Help/about.gif
diff --git a/Ready/Help/about.html b/Help/about.html
similarity index 100%
rename from Ready/Help/about.html
rename to Help/about.html
diff --git a/Ready/Help/action.html b/Help/action.html
similarity index 100%
rename from Ready/Help/action.html
rename to Help/action.html
diff --git a/Ready/Help/changes.html b/Help/changes.html
similarity index 100%
rename from Ready/Help/changes.html
rename to Help/changes.html
diff --git a/Ready/Help/credits.html b/Help/credits.html
similarity index 100%
rename from Ready/Help/credits.html
rename to Help/credits.html
diff --git a/Ready/Help/edit.html b/Help/edit.html
similarity index 100%
rename from Ready/Help/edit.html
rename to Help/edit.html
diff --git a/Ready/Help/file.html b/Help/file.html
similarity index 100%
rename from Ready/Help/file.html
rename to Help/file.html
diff --git a/Ready/Help/formats.html b/Help/formats.html
similarity index 100%
rename from Ready/Help/formats.html
rename to Help/formats.html
diff --git a/Ready/Help/help.html b/Help/help.html
similarity index 100%
rename from Ready/Help/help.html
rename to Help/help.html
diff --git a/Ready/Help/index.html b/Help/index.html
similarity index 100%
rename from Ready/Help/index.html
rename to Help/index.html
diff --git a/Ready/Help/introduction.html b/Help/introduction.html
similarity index 100%
rename from Ready/Help/introduction.html
rename to Help/introduction.html
diff --git a/Ready/Help/mouse.html b/Help/mouse.html
similarity index 100%
rename from Ready/Help/mouse.html
rename to Help/mouse.html
diff --git a/Ready/Help/problems.html b/Help/problems.html
similarity index 100%
rename from Ready/Help/problems.html
rename to Help/problems.html
diff --git a/Ready/Help/quickstart.html b/Help/quickstart.html
similarity index 100%
rename from Ready/Help/quickstart.html
rename to Help/quickstart.html
diff --git a/Ready/Help/tips.html b/Help/tips.html
similarity index 100%
rename from Ready/Help/tips.html
rename to Help/tips.html
diff --git a/Ready/Help/view.html b/Help/view.html
similarity index 100%
rename from Ready/Help/view.html
rename to Help/view.html
diff --git a/Linear/CMakeLists.txt b/Linear/CMakeLists.txt
deleted file mode 100644
index 09587ceb2..000000000
--- a/Linear/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(Linear)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(Linear 
-  linear.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(Linear ${OpenCV_LIBS} )
diff --git a/Linear/linear.cpp b/Linear/linear.cpp
deleted file mode 100644
index c3ce53db3..000000000
--- a/Linear/linear.cpp
+++ /dev/null
@@ -1,138 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],float da[X][Y],float db[X][Y]);
-
-int main()
-{
-    // http://www.nature.com/ncomms/journal/v1/n6/full/ncomms1071.html
-
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true)
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,da,db);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Linear - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,b,iteration,true,30.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    for(int i = 0; i < X; i++)
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            float m = 0.01f;
-            a[i][j] = frand(-m,m);
-            b[i][j] = frand(-m,m);
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y])
-{
-    // -- parameters --
-    float A = 0.08f;
-    float B = 0.08f;
-    float C = 0.15f;
-    float D = 0.03f;
-    float E = 0.10f;
-    float F = 0.12f;
-    float G = 0.06f;
-    float D_a = 0.5f;
-    float D_b = 10.0f;
-    float R = 20.0f;
-    float syn_a_max = 0.23f;
-    float syn_b_max = 0.50f;
-    float speed = 0.001f;
-    // ----------------
-    
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        int iprev = (i + X - 1) % X;
-        int inext = (i + 1) % X;
-        for(int j = 0; j < Y; j++) 
-        {
-            int jprev = (j + Y - 1) % Y;
-            int jnext = (j + 1) % Y;
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4 * aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4 * bval;
-
-            // compute the synthesis rates of a and b
-            float syn_a = A*aval + B*bval + C;
-            float syn_b = E*aval - F;
-            if(syn_a<0.0f) syn_a=0.0f; if(syn_a>syn_a_max) syn_a=syn_a_max;
-            if(syn_b<0.0f) syn_b=0.0f; if(syn_b>syn_b_max) syn_b=syn_b_max;
-
-            // compute the new rate of change of a and b
-            da[i][j] = R*( syn_a - D*aval ) + D_a * dda;
-            db[i][j] = R*( syn_b - G*bval ) + D_b * ddb;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++)
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
diff --git a/MeinhardtSpots/CMakeLists.txt b/MeinhardtSpots/CMakeLists.txt
deleted file mode 100644
index ba5d928d6..000000000
--- a/MeinhardtSpots/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(MeinhardtSpots)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(MeinhardtSpots 
-  meinhardt_spots.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(MeinhardtSpots ${OpenCV_LIBS} )
diff --git a/MeinhardtSpots/meinhardt_spots.cpp b/MeinhardtSpots/meinhardt_spots.cpp
deleted file mode 100644
index a3ece6566..000000000
--- a/MeinhardtSpots/meinhardt_spots.cpp
+++ /dev/null
@@ -1,141 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-NB. Turk's thesis has errors in the formulae for this system, the correct ones are here:
-http://www1.cse.wustl.edu/~faanly/materials/Sketching_RD_Texture.pdf
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y],float beta[X][Y],
-          float p1,float p2);
-
-void compute(float a[X][Y],float b[X][Y],float beta[X][Y],
-             float da[X][Y],float db[X][Y],
-             float diff1,float diff2,float p1,float p2,float p3,float s,
-             float speed);
-
-int main()
-{
-    // -- parameters --
-    float p1 = 0.03f;
-    float p2 = 0.04f;
-    float p3 = 0.0f;
-    float diff1 = 0.01f;
-    float diff2 = 0.2f;
-    float s = 0.2f;
-    float speed = 1.0f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    float beta[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b,beta,p1,p2);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true)
-    {
-		start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,beta,da,db,diff1,diff2,p1,p2,p3,s,speed);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Meinhardt spots - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,b,iteration,true,30.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y],float beta[X][Y],
-          float p1,float p2)
-{
-    srand((unsigned int)time(NULL));
-    
-    float ainit,binit;
-    ainit = p2 / p1;
-    binit = 0.01f * ainit * ainit / p2;
-
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] = ainit;
-            b[i][j] = binit;
-            if(i<X/2)
-                beta[i][j] = frand(0.9f,1.1f);
-            else
-                beta[i][j] = 1;
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],float beta[X][Y],
-             float da[X][Y],float db[X][Y],
-             float diff1,float diff2,float p1,float p2,float p3,float s,
-             float speed)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) {
-
-        int iprev = (i + X - 1) % X;
-        int inext = (i + 1) % X;
-
-        for(int j = 0; j < Y; j++) {
-
-            int jprev = (j + Y - 1) % Y;
-            int jnext = (j + 1) % Y;
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4 * aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4 * bval;
-
-            // compute the new rates of change
-            da[i][j] = s * (0.01f * aval * aval * beta[i][j] / bval - aval * p1 + p3) + diff1 * dda;
-            db[i][j] = s * (0.01f * aval * aval * beta[i][j] - bval * p2 + p3) + diff2 * ddb;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
-
diff --git a/MeinhardtStripes/CMakeLists.txt b/MeinhardtStripes/CMakeLists.txt
deleted file mode 100644
index f88858f15..000000000
--- a/MeinhardtStripes/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(MeinhardtStripes)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(MeinhardtStripes 
-  meinhardt_stripes.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(MeinhardtStripes ${OpenCV_LIBS} )
diff --git a/MeinhardtStripes/meinhardt_stripes.cpp b/MeinhardtStripes/meinhardt_stripes.cpp
deleted file mode 100644
index 954c427bf..000000000
--- a/MeinhardtStripes/meinhardt_stripes.cpp
+++ /dev/null
@@ -1,169 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-See also:
-http://www1.cse.wustl.edu/~faanly/materials/Sketching_RD_Texture.pdf
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y],float c[X][Y],float d[X][Y],float e[X][Y],
-          float beta[X][Y],
-          float k_ab,float k_c,float arand);
-
-void compute(float a[X][Y],float b[X][Y],float c[X][Y],float d[X][Y],float e[X][Y],
-             float beta[X][Y],
-             float da[X][Y],float db[X][Y],float dc[X][Y],float dd[X][Y],float de[X][Y],
-             float diff1,float diff2,float k_ab,float k_c,float k_de,float speed);
-
-int main()
-{
-    // -- parameters --
-    float k_ab = 0.04f; // p1
-    float k_c = 0.06f; // p2
-    float k_de = 0.04f; // p3
-    float diff1 = 0.009f;
-    float diff2 = 0.2f;
-    float arand = 0.01f;
-    float speed = 1.0f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y], c[X][Y], d[X][Y], e[X][Y];
-    float beta[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y], dc[X][Y], dd[X][Y], de[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b,c,d,e,beta,k_ab,k_c,arand);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true)
-    {
-		start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,c,d,e,beta,da,db,dc,dd,de,diff1,diff2,k_ab,k_c,k_de,speed);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Meinhardt stripes - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,b,iteration,true,30.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y],float c[X][Y],float d[X][Y],float e[X][Y],
-          float beta[X][Y],
-          float k_ab,float k_c,float arand)
-{
-    srand((unsigned int)time(NULL));
-    
-    float ainit,binit;
-    float cinit,dinit,einit;
-
-    ainit = binit = cinit = dinit = einit = 0;
-
-    // figure the values
-    for(int i = 0; i < X; i++) 
-    {
-        ainit = k_c / (2 * k_ab);
-        binit = ainit;
-        cinit = 0.02f * ainit * ainit * ainit / k_c;
-        dinit = ainit;
-        einit = ainit;
-
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] = ainit;
-            b[i][j] = binit;
-            c[i][j] = cinit;
-            d[i][j] = dinit;
-            e[i][j] = einit;
-            if(i<X/4) // only initialise part of the grid, to allow regular stripes to form elsewhere
-                beta[i][j] = 1 + frand (-0.5f * arand, 0.5f * arand);
-            else
-                beta[i][j] = 1;
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],float c[X][Y],float d[X][Y],float e[X][Y],
-             float beta[X][Y],
-             float da[X][Y],float db[X][Y],float dc[X][Y],float dd[X][Y],float de[X][Y],
-             float diff1,float diff2,float k_ab,float k_c,float k_de,float speed)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        int iprev = (i + X - 1) % X;
-        int inext = (i + 1) % X;
-
-        for(int j = 0; j < Y; j++) 
-        {
-            int jprev = (j + Y - 1) % Y;
-            int jnext = (j + 1) % Y;
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-            float cval = c[i][j];
-            float dval = d[i][j];
-            float eval = e[i][j];
-
-            // compute the Laplacians of a, b, d and e (we don't need c)
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4 * aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4 * bval;
-            float ddd = d[i][jprev] + d[i][jnext] + d[iprev][j] + d[inext][j] - 4 * dval;
-            float dde = e[i][jprev] + e[i][jnext] + e[iprev][j] + e[inext][j] - 4 * eval;
-
-            // compute the rates of change of each chemical
-            da[i][j] = 0.01f * aval * aval * eval * beta[i][j] / cval - aval * k_ab + diff1 * dda;
-            db[i][j] = 0.01f * bval * bval * dval / cval - bval * k_ab + diff1 * ddb;
-            dc[i][j] = 0.01f * aval * aval * eval * beta[i][j] + 0.01f * bval * bval * dval - cval * k_c;
-            dd[i][j] = (aval - dval) * k_de + diff2 * ddd;
-            de[i][j] = (bval - eval) * k_de + diff2 * dde;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++)
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-            c[i][j] += (speed * dc[i][j]);
-            d[i][j] += (speed * dd[i][j]);
-            e[i][j] += (speed * de[i][j]);
-        }
-    }
-}
-
diff --git a/Oregonator/CMakeLists.txt b/Oregonator/CMakeLists.txt
deleted file mode 100644
index 57785babc..000000000
--- a/Oregonator/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(Oregonator)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(Oregonator
-  oregonator.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(Oregonator ${OpenCV_LIBS} )
diff --git a/Oregonator/oregonator.cpp b/Oregonator/oregonator.cpp
deleted file mode 100644
index c66eabbd7..000000000
--- a/Oregonator/oregonator.cpp
+++ /dev/null
@@ -1,157 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y],float c[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],float c[X][Y],
-             float da[X][Y],float db[X][Y],float dc[X][Y],
-             float Da,float Db,float Dc,float q,float epsilon,float f,
-             float speed,
-             bool parameter_space);
-
-int main()
-{
-    // Following:
-    // http://hopf.chem.brandeis.edu/yanglingfa/pattern/oreg/index.html
-
-    // But I can't get this to work at the moment.
-
-    // -- parameters --
-    float q = 0.01f;
-    float epsilon = 0.5f;
-    float f = 0.80f; // labyrinthine
-    float Da = 3.0f; // (Dx,Dz,Dr at http://hopf.chem.brandeis.edu/yanglingfa/pattern/oreg/index.html )
-    float Db = 100.0f;
-    float Dc = 0.1f;
-    float speed = 0.00001f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y], c[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y], dc[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b,c);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,c,da,db,dc,Da,Db,Dc,q,epsilon,f,speed,false);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Oregonator - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,b,c,iteration,true,20.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y],float c[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    for(int i = 0; i < X; i++) 
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] = frand(-0.01f,0.01f);
-            b[i][j] = frand(-0.01f,0.01f);
-            c[i][j] = frand(-0.01f,0.01f);
-        }
-    }
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void compute(float a[X][Y],float b[X][Y],float c[X][Y],
-             float da[X][Y],float db[X][Y],float dc[X][Y],
-             float Da,float Db,float Dc,float q,float epsilon,float f,
-             float speed,
-             bool parameter_space)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        //int iprev = (i + X - 1) % X;
-        //int inext = (i + 1) % X;
-        int iprev = max(0,i-1);
-        int inext = min(X-1,i+1);
-
-        for(int j = 0; j < Y; j++) 
-        {
-            //int jprev = (j + Y - 1) % Y;
-            //int jnext = (j + 1) % Y;
-            int jprev = max(0,j-1);
-            int jnext = min(Y-1,j+1);
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-            float cval = c[i][j];
-
-            if(parameter_space)
-            {
-                /*const float A1=0.0f,A2=4.0f,B1=0.0f,B2=15.0f;
-                A = A1+(A2-A1)*i/X;
-                B = B2+(B1-B2)*j/Y;*/
-            }
-
-            // compute the Laplacians of a, b and c
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-            float ddc = c[i][jprev] + c[i][jnext] + c[iprev][j] + c[inext][j] - 4*cval;
-
-            // compute the new rate of change of a, b and c
-            da[i][j] = Da*dda + ( aval - aval*aval - f*bval*(aval-q)/(aval+q) - (aval-cval)/2.0f ) / epsilon;
-            db[i][j] = Db*ddb + aval - bval;
-            dc[i][j] = Dc*ddc + ( aval - cval ) / ( 2.0f * epsilon );
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-            c[i][j] += (speed * dc[i][j]);
-        }
-    }
-}
-
diff --git a/Ready/Patterns/Brusselator.vti b/Patterns/Brusselator.vti
similarity index 100%
rename from Ready/Patterns/Brusselator.vti
rename to Patterns/Brusselator.vti
diff --git a/Ready/Patterns/CPU-only/grayscott_1D.vti b/Patterns/CPU-only/grayscott_1D.vti
similarity index 100%
rename from Ready/Patterns/CPU-only/grayscott_1D.vti
rename to Patterns/CPU-only/grayscott_1D.vti
diff --git a/Ready/Patterns/CPU-only/grayscott_2D.vti b/Patterns/CPU-only/grayscott_2D.vti
similarity index 100%
rename from Ready/Patterns/CPU-only/grayscott_2D.vti
rename to Patterns/CPU-only/grayscott_2D.vti
diff --git a/Ready/Patterns/CPU-only/grayscott_3D.vti b/Patterns/CPU-only/grayscott_3D.vti
similarity index 100%
rename from Ready/Patterns/CPU-only/grayscott_3D.vti
rename to Patterns/CPU-only/grayscott_3D.vti
diff --git a/Ready/Patterns/CellularAutomata/Bays_3D.vti b/Patterns/CellularAutomata/Bays_3D.vti
similarity index 100%
rename from Ready/Patterns/CellularAutomata/Bays_3D.vti
rename to Patterns/CellularAutomata/Bays_3D.vti
diff --git a/Ready/Patterns/CellularAutomata/Buss_hex.vtu b/Patterns/CellularAutomata/Buss_hex.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/Buss_hex.vtu
rename to Patterns/CellularAutomata/Buss_hex.vtu
diff --git a/Ready/Patterns/CellularAutomata/Conway_life.vti b/Patterns/CellularAutomata/Conway_life.vti
similarity index 100%
rename from Ready/Patterns/CellularAutomata/Conway_life.vti
rename to Patterns/CellularAutomata/Conway_life.vti
diff --git a/Ready/Patterns/CellularAutomata/PenroseTilings/Goucher_glider.vtu b/Patterns/CellularAutomata/PenroseTilings/Goucher_glider.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/PenroseTilings/Goucher_glider.vtu
rename to Patterns/CellularAutomata/PenroseTilings/Goucher_glider.vtu
diff --git a/Ready/Patterns/CellularAutomata/PenroseTilings/Goucher_loops.vtu b/Patterns/CellularAutomata/PenroseTilings/Goucher_loops.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/PenroseTilings/Goucher_loops.vtu
rename to Patterns/CellularAutomata/PenroseTilings/Goucher_loops.vtu
diff --git a/Ready/Patterns/CellularAutomata/PenroseTilings/Imai_glider_B2SC4.vtu b/Patterns/CellularAutomata/PenroseTilings/Imai_glider_B2SC4.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/PenroseTilings/Imai_glider_B2SC4.vtu
rename to Patterns/CellularAutomata/PenroseTilings/Imai_glider_B2SC4.vtu
diff --git a/Ready/Patterns/CellularAutomata/PenroseTilings/life.vtu b/Patterns/CellularAutomata/PenroseTilings/life.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/PenroseTilings/life.vtu
rename to Patterns/CellularAutomata/PenroseTilings/life.vtu
diff --git a/Ready/Patterns/CellularAutomata/PenroseTilings/life_oscillators.vtu b/Patterns/CellularAutomata/PenroseTilings/life_oscillators.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/PenroseTilings/life_oscillators.vtu
rename to Patterns/CellularAutomata/PenroseTilings/life_oscillators.vtu
diff --git a/Ready/Patterns/CellularAutomata/Salt/salt2D_demo.vti b/Patterns/CellularAutomata/Salt/salt2D_demo.vti
similarity index 100%
rename from Ready/Patterns/CellularAutomata/Salt/salt2D_demo.vti
rename to Patterns/CellularAutomata/Salt/salt2D_demo.vti
diff --git a/Ready/Patterns/CellularAutomata/Salt/salt3D_circular330.vti b/Patterns/CellularAutomata/Salt/salt3D_circular330.vti
similarity index 100%
rename from Ready/Patterns/CellularAutomata/Salt/salt3D_circular330.vti
rename to Patterns/CellularAutomata/Salt/salt3D_circular330.vti
diff --git a/Ready/Patterns/CellularAutomata/hex_B2oS2m34_gliders.vtu b/Patterns/CellularAutomata/hex_B2oS2m34_gliders.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/hex_B2oS2m34_gliders.vtu
rename to Patterns/CellularAutomata/hex_B2oS2m34_gliders.vtu
diff --git a/Ready/Patterns/CellularAutomata/larger-than-life.vti b/Patterns/CellularAutomata/larger-than-life.vti
similarity index 100%
rename from Ready/Patterns/CellularAutomata/larger-than-life.vti
rename to Patterns/CellularAutomata/larger-than-life.vti
diff --git a/Ready/Patterns/CellularAutomata/life_torus.vtu b/Patterns/CellularAutomata/life_torus.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/life_torus.vtu
rename to Patterns/CellularAutomata/life_torus.vtu
diff --git a/Ready/Patterns/CellularAutomata/tri_life.vtu b/Patterns/CellularAutomata/tri_life.vtu
similarity index 100%
rename from Ready/Patterns/CellularAutomata/tri_life.vtu
rename to Patterns/CellularAutomata/tri_life.vtu
diff --git a/Ready/Patterns/Experiments/cglrd_ramps_example_djw.vti b/Patterns/Experiments/cglrd_ramps_example_djw.vti
similarity index 100%
rename from Ready/Patterns/Experiments/cglrd_ramps_example_djw.vti
rename to Patterns/Experiments/cglrd_ramps_example_djw.vti
diff --git a/Ready/Patterns/Experiments/gladman_vermiformSolitons.vti b/Patterns/Experiments/gladman_vermiformSolitons.vti
similarity index 100%
rename from Ready/Patterns/Experiments/gladman_vermiformSolitons.vti
rename to Patterns/Experiments/gladman_vermiformSolitons.vti
diff --git a/Ready/Patterns/Experiments/grayscott-historyWaveDC_solitonsAndWorms_init.vti b/Patterns/Experiments/grayscott-historyWaveDC_solitonsAndWorms_init.vti
similarity index 100%
rename from Ready/Patterns/Experiments/grayscott-historyWaveDC_solitonsAndWorms_init.vti
rename to Patterns/Experiments/grayscott-historyWaveDC_solitonsAndWorms_init.vti
diff --git a/Ready/Patterns/Experiments/grayscott-historyWave_coralGrow_djw.vti b/Patterns/Experiments/grayscott-historyWave_coralGrow_djw.vti
similarity index 100%
rename from Ready/Patterns/Experiments/grayscott-historyWave_coralGrow_djw.vti
rename to Patterns/Experiments/grayscott-historyWave_coralGrow_djw.vti
diff --git a/Ready/Patterns/Experiments/grayscott-historyWave_fuseWorms.vti b/Patterns/Experiments/grayscott-historyWave_fuseWorms.vti
similarity index 100%
rename from Ready/Patterns/Experiments/grayscott-historyWave_fuseWorms.vti
rename to Patterns/Experiments/grayscott-historyWave_fuseWorms.vti
diff --git a/Ready/Patterns/Experiments/grayscott-historyWave_moreLifelike.vti b/Patterns/Experiments/grayscott-historyWave_moreLifelike.vti
similarity index 100%
rename from Ready/Patterns/Experiments/grayscott-historyWave_moreLifelike.vti
rename to Patterns/Experiments/grayscott-historyWave_moreLifelike.vti
diff --git a/Ready/Patterns/Experiments/mutually-catalytic_spots.vti b/Patterns/Experiments/mutually-catalytic_spots.vti
similarity index 100%
rename from Ready/Patterns/Experiments/mutually-catalytic_spots.vti
rename to Patterns/Experiments/mutually-catalytic_spots.vti
diff --git a/Ready/Patterns/Experiments/orbits_explodey_init.djw.vti b/Patterns/Experiments/orbits_explodey_init.djw.vti
similarity index 100%
rename from Ready/Patterns/Experiments/orbits_explodey_init.djw.vti
rename to Patterns/Experiments/orbits_explodey_init.djw.vti
diff --git a/Ready/Patterns/Experiments/orbits_sharpWaves-init_djw.vti b/Patterns/Experiments/orbits_sharpWaves-init_djw.vti
similarity index 100%
rename from Ready/Patterns/Experiments/orbits_sharpWaves-init_djw.vti
rename to Patterns/Experiments/orbits_sharpWaves-init_djw.vti
diff --git a/Ready/Patterns/FitzHugh-Nagumo/Ising_regime.vti b/Patterns/FitzHugh-Nagumo/Ising_regime.vti
similarity index 100%
rename from Ready/Patterns/FitzHugh-Nagumo/Ising_regime.vti
rename to Patterns/FitzHugh-Nagumo/Ising_regime.vti
diff --git a/Ready/Patterns/FitzHugh-Nagumo/pulsate.vti b/Patterns/FitzHugh-Nagumo/pulsate.vti
similarity index 100%
rename from Ready/Patterns/FitzHugh-Nagumo/pulsate.vti
rename to Patterns/FitzHugh-Nagumo/pulsate.vti
diff --git a/Ready/Patterns/FitzHugh-Nagumo/spiral_turbulence.vti b/Patterns/FitzHugh-Nagumo/spiral_turbulence.vti
similarity index 100%
rename from Ready/Patterns/FitzHugh-Nagumo/spiral_turbulence.vti
rename to Patterns/FitzHugh-Nagumo/spiral_turbulence.vti
diff --git a/Ready/Patterns/FitzHugh-Nagumo/squid_axon.vti b/Patterns/FitzHugh-Nagumo/squid_axon.vti
similarity index 100%
rename from Ready/Patterns/FitzHugh-Nagumo/squid_axon.vti
rename to Patterns/FitzHugh-Nagumo/squid_axon.vti
diff --git a/Ready/Patterns/FitzHugh-Nagumo/tip-splitting.vti b/Patterns/FitzHugh-Nagumo/tip-splitting.vti
similarity index 100%
rename from Ready/Patterns/FitzHugh-Nagumo/tip-splitting.vti
rename to Patterns/FitzHugh-Nagumo/tip-splitting.vti
diff --git a/Ready/Patterns/Ginzburg-Landau/complex_Ginzburg-Landau.vti b/Patterns/Ginzburg-Landau/complex_Ginzburg-Landau.vti
similarity index 100%
rename from Ready/Patterns/Ginzburg-Landau/complex_Ginzburg-Landau.vti
rename to Patterns/Ginzburg-Landau/complex_Ginzburg-Landau.vti
diff --git a/Ready/Patterns/Ginzburg-Landau/complex_Ginzburg-Landau_magnitude.vti b/Patterns/Ginzburg-Landau/complex_Ginzburg-Landau_magnitude.vti
similarity index 100%
rename from Ready/Patterns/Ginzburg-Landau/complex_Ginzburg-Landau_magnitude.vti
rename to Patterns/Ginzburg-Landau/complex_Ginzburg-Landau_magnitude.vti
diff --git a/Ready/Patterns/Gray-Scott/Lesmes_noisy.vti b/Patterns/Gray-Scott/Lesmes_noisy.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/Lesmes_noisy.vti
rename to Patterns/Gray-Scott/Lesmes_noisy.vti
diff --git a/Ready/Patterns/Gray-Scott/Pearson1993.vti b/Patterns/Gray-Scott/Pearson1993.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/Pearson1993.vti
rename to Patterns/Gray-Scott/Pearson1993.vti
diff --git a/Ready/Patterns/Gray-Scott/U-Skate/Hutton-and-helix-gliders.vti b/Patterns/Gray-Scott/U-Skate/Hutton-and-helix-gliders.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/U-Skate/Hutton-and-helix-gliders.vti
rename to Patterns/Gray-Scott/U-Skate/Hutton-and-helix-gliders.vti
diff --git a/Ready/Patterns/Gray-Scott/U-Skate/Munafo_glider.vti b/Patterns/Gray-Scott/U-Skate/Munafo_glider.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/U-Skate/Munafo_glider.vti
rename to Patterns/Gray-Scott/U-Skate/Munafo_glider.vti
diff --git a/Ready/Patterns/Gray-Scott/U-Skate/o-ring_2D.vti b/Patterns/Gray-Scott/U-Skate/o-ring_2D.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/U-Skate/o-ring_2D.vti
rename to Patterns/Gray-Scott/U-Skate/o-ring_2D.vti
diff --git a/Ready/Patterns/Gray-Scott/noisy_solitons_mitosis.vti b/Patterns/Gray-Scott/noisy_solitons_mitosis.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/noisy_solitons_mitosis.vti
rename to Patterns/Gray-Scott/noisy_solitons_mitosis.vti
diff --git a/Ready/Patterns/Gray-Scott/parameter-map.vti b/Patterns/Gray-Scott/parameter-map.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/parameter-map.vti
rename to Patterns/Gray-Scott/parameter-map.vti
diff --git a/Ready/Patterns/Gray-Scott/self-replicating_spots.vti b/Patterns/Gray-Scott/self-replicating_spots.vti
similarity index 100%
rename from Ready/Patterns/Gray-Scott/self-replicating_spots.vti
rename to Patterns/Gray-Scott/self-replicating_spots.vti
diff --git a/Ready/Patterns/Kytta2007/Fig5.7a.vti b/Patterns/Kytta2007/Fig5.7a.vti
similarity index 100%
rename from Ready/Patterns/Kytta2007/Fig5.7a.vti
rename to Patterns/Kytta2007/Fig5.7a.vti
diff --git a/Ready/Patterns/Kytta2007/Fig5.7c.vti b/Patterns/Kytta2007/Fig5.7c.vti
similarity index 100%
rename from Ready/Patterns/Kytta2007/Fig5.7c.vti
rename to Patterns/Kytta2007/Fig5.7c.vti
diff --git a/Ready/Patterns/Kytta2007/Fig5.8c.vti b/Patterns/Kytta2007/Fig5.8c.vti
similarity index 100%
rename from Ready/Patterns/Kytta2007/Fig5.8c.vti
rename to Patterns/Kytta2007/Fig5.8c.vti
diff --git a/Ready/Patterns/Kytta2007/Fig5.8d.vti b/Patterns/Kytta2007/Fig5.8d.vti
similarity index 100%
rename from Ready/Patterns/Kytta2007/Fig5.8d.vti
rename to Patterns/Kytta2007/Fig5.8d.vti
diff --git a/Ready/Patterns/Kytta2007/Fig5.8e.vti b/Patterns/Kytta2007/Fig5.8e.vti
similarity index 100%
rename from Ready/Patterns/Kytta2007/Fig5.8e.vti
rename to Patterns/Kytta2007/Fig5.8e.vti
diff --git a/Ready/Patterns/Kytta2007/Fig5.8f.vti b/Patterns/Kytta2007/Fig5.8f.vti
similarity index 100%
rename from Ready/Patterns/Kytta2007/Fig5.8f.vti
rename to Patterns/Kytta2007/Fig5.8f.vti
diff --git a/Ready/Patterns/Kytta2007/Fig5.8g.vti b/Patterns/Kytta2007/Fig5.8g.vti
similarity index 100%
rename from Ready/Patterns/Kytta2007/Fig5.8g.vti
rename to Patterns/Kytta2007/Fig5.8g.vti
diff --git a/Ready/Patterns/McCabe/McCabe.vti b/Patterns/McCabe/McCabe.vti
similarity index 100%
rename from Ready/Patterns/McCabe/McCabe.vti
rename to Patterns/McCabe/McCabe.vti
diff --git a/Ready/Patterns/McCabe/McCabe_additive2a.vti b/Patterns/McCabe/McCabe_additive2a.vti
similarity index 100%
rename from Ready/Patterns/McCabe/McCabe_additive2a.vti
rename to Patterns/McCabe/McCabe_additive2a.vti
diff --git a/Ready/Patterns/McCabe/McCabe_additive2b.vti b/Patterns/McCabe/McCabe_additive2b.vti
similarity index 100%
rename from Ready/Patterns/McCabe/McCabe_additive2b.vti
rename to Patterns/McCabe/McCabe_additive2b.vti
diff --git a/Ready/Patterns/McCabe/McCabe_simple.vti b/Patterns/McCabe/McCabe_simple.vti
similarity index 100%
rename from Ready/Patterns/McCabe/McCabe_simple.vti
rename to Patterns/McCabe/McCabe_simple.vti
diff --git a/Ready/Patterns/Meinhardt1982/stripes.vti b/Patterns/Meinhardt1982/stripes.vti
similarity index 100%
rename from Ready/Patterns/Meinhardt1982/stripes.vti
rename to Patterns/Meinhardt1982/stripes.vti
diff --git a/Ready/Patterns/Meinhardt1982/zebra.vtu b/Patterns/Meinhardt1982/zebra.vtu
similarity index 100%
rename from Ready/Patterns/Meinhardt1982/zebra.vtu
rename to Patterns/Meinhardt1982/zebra.vtu
diff --git a/Ready/Patterns/Purwins1999/glider.vti b/Patterns/Purwins1999/glider.vti
similarity index 100%
rename from Ready/Patterns/Purwins1999/glider.vti
rename to Patterns/Purwins1999/glider.vti
diff --git a/Ready/Patterns/Purwins1999/glider_3D.vti b/Patterns/Purwins1999/glider_3D.vti
similarity index 100%
rename from Ready/Patterns/Purwins1999/glider_3D.vti
rename to Patterns/Purwins1999/glider_3D.vti
diff --git a/Ready/Patterns/Purwins1999/multiGlider.vti b/Patterns/Purwins1999/multiGlider.vti
similarity index 100%
rename from Ready/Patterns/Purwins1999/multiGlider.vti
rename to Patterns/Purwins1999/multiGlider.vti
diff --git a/Ready/Patterns/Schlogl.vti b/Patterns/Schlogl.vti
similarity index 100%
rename from Ready/Patterns/Schlogl.vti
rename to Patterns/Schlogl.vti
diff --git a/Ready/Patterns/Schrodinger1926/packet.vti b/Patterns/Schrodinger1926/packet.vti
similarity index 100%
rename from Ready/Patterns/Schrodinger1926/packet.vti
rename to Patterns/Schrodinger1926/packet.vti
diff --git a/Ready/Patterns/Schrodinger1926/packet_pass.vti b/Patterns/Schrodinger1926/packet_pass.vti
similarity index 100%
rename from Ready/Patterns/Schrodinger1926/packet_pass.vti
rename to Patterns/Schrodinger1926/packet_pass.vti
diff --git a/Ready/Patterns/Schrodinger1926/packet_reflect.vti b/Patterns/Schrodinger1926/packet_reflect.vti
similarity index 100%
rename from Ready/Patterns/Schrodinger1926/packet_reflect.vti
rename to Patterns/Schrodinger1926/packet_reflect.vti
diff --git a/Ready/Patterns/Schrodinger1926/packet_reflect2D.vti b/Patterns/Schrodinger1926/packet_reflect2D.vti
similarity index 100%
rename from Ready/Patterns/Schrodinger1926/packet_reflect2D.vti
rename to Patterns/Schrodinger1926/packet_reflect2D.vti
diff --git a/Ready/Patterns/Schrodinger1926/quantum_tunnelling.vti b/Patterns/Schrodinger1926/quantum_tunnelling.vti
similarity index 100%
rename from Ready/Patterns/Schrodinger1926/quantum_tunnelling.vti
rename to Patterns/Schrodinger1926/quantum_tunnelling.vti
diff --git a/Ready/Patterns/SmoothLife/glider_3D.vti b/Patterns/SmoothLife/glider_3D.vti
similarity index 100%
rename from Ready/Patterns/SmoothLife/glider_3D.vti
rename to Patterns/SmoothLife/glider_3D.vti
diff --git a/Ready/Patterns/SmoothLife/smoothglider.vti b/Patterns/SmoothLife/smoothglider.vti
similarity index 100%
rename from Ready/Patterns/SmoothLife/smoothglider.vti
rename to Patterns/SmoothLife/smoothglider.vti
diff --git a/Ready/Patterns/SmoothLife/smoothlifeL.vti b/Patterns/SmoothLife/smoothlifeL.vti
similarity index 100%
rename from Ready/Patterns/SmoothLife/smoothlifeL.vti
rename to Patterns/SmoothLife/smoothlifeL.vti
diff --git a/Ready/Patterns/Turing1952/spots.vti b/Patterns/Turing1952/spots.vti
similarity index 100%
rename from Ready/Patterns/Turing1952/spots.vti
rename to Patterns/Turing1952/spots.vti
diff --git a/Ready/Patterns/Turing1952/spots_noisy.vti b/Patterns/Turing1952/spots_noisy.vti
similarity index 100%
rename from Ready/Patterns/Turing1952/spots_noisy.vti
rename to Patterns/Turing1952/spots_noisy.vti
diff --git a/Ready/Patterns/Yang2002/Yang_1.vti b/Patterns/Yang2002/Yang_1.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_1.vti
rename to Patterns/Yang2002/Yang_1.vti
diff --git a/Ready/Patterns/Yang2002/Yang_2b.vti b/Patterns/Yang2002/Yang_2b.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_2b.vti
rename to Patterns/Yang2002/Yang_2b.vti
diff --git a/Ready/Patterns/Yang2002/Yang_2c.vti b/Patterns/Yang2002/Yang_2c.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_2c.vti
rename to Patterns/Yang2002/Yang_2c.vti
diff --git a/Ready/Patterns/Yang2002/Yang_2d.vti b/Patterns/Yang2002/Yang_2d.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_2d.vti
rename to Patterns/Yang2002/Yang_2d.vti
diff --git a/Ready/Patterns/Yang2002/Yang_3a.vti b/Patterns/Yang2002/Yang_3a.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_3a.vti
rename to Patterns/Yang2002/Yang_3a.vti
diff --git a/Ready/Patterns/Yang2002/Yang_3b.vti b/Patterns/Yang2002/Yang_3b.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_3b.vti
rename to Patterns/Yang2002/Yang_3b.vti
diff --git a/Ready/Patterns/Yang2002/Yang_3c.vti b/Patterns/Yang2002/Yang_3c.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_3c.vti
rename to Patterns/Yang2002/Yang_3c.vti
diff --git a/Ready/Patterns/Yang2002/Yang_3d.vti b/Patterns/Yang2002/Yang_3d.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_3d.vti
rename to Patterns/Yang2002/Yang_3d.vti
diff --git a/Ready/Patterns/Yang2002/Yang_4.vti b/Patterns/Yang2002/Yang_4.vti
similarity index 100%
rename from Ready/Patterns/Yang2002/Yang_4.vti
rename to Patterns/Yang2002/Yang_4.vti
diff --git a/Ready/Patterns/Yang2003/Fig2.vti b/Patterns/Yang2003/Fig2.vti
similarity index 100%
rename from Ready/Patterns/Yang2003/Fig2.vti
rename to Patterns/Yang2003/Fig2.vti
diff --git a/Ready/Patterns/Yang2003/Fig3a.vti b/Patterns/Yang2003/Fig3a.vti
similarity index 100%
rename from Ready/Patterns/Yang2003/Fig3a.vti
rename to Patterns/Yang2003/Fig3a.vti
diff --git a/Ready/Patterns/Yang2003/Fig3b.vti b/Patterns/Yang2003/Fig3b.vti
similarity index 100%
rename from Ready/Patterns/Yang2003/Fig3b.vti
rename to Patterns/Yang2003/Fig3b.vti
diff --git a/Ready/Patterns/Yang2003/Fig3c.vti b/Patterns/Yang2003/Fig3c.vti
similarity index 100%
rename from Ready/Patterns/Yang2003/Fig3c.vti
rename to Patterns/Yang2003/Fig3c.vti
diff --git a/Ready/Patterns/Yang2006/jumping.vti b/Patterns/Yang2006/jumping.vti
similarity index 100%
rename from Ready/Patterns/Yang2006/jumping.vti
rename to Patterns/Yang2006/jumping.vti
diff --git a/Ready/Patterns/Yang2006/jumping_cGL.vti b/Patterns/Yang2006/jumping_cGL.vti
similarity index 100%
rename from Ready/Patterns/Yang2006/jumping_cGL.vti
rename to Patterns/Yang2006/jumping_cGL.vti
diff --git a/Ready/Patterns/bunny.vtu b/Patterns/bunny.vtu
similarity index 100%
rename from Ready/Patterns/bunny.vtu
rename to Patterns/bunny.vtu
diff --git a/Ready/Patterns/heat_equation.vti b/Patterns/heat_equation.vti
similarity index 100%
rename from Ready/Patterns/heat_equation.vti
rename to Patterns/heat_equation.vti
diff --git a/Ready/Patterns/heat_equation_interpolation.vti b/Patterns/heat_equation_interpolation.vti
similarity index 100%
rename from Ready/Patterns/heat_equation_interpolation.vti
rename to Patterns/heat_equation_interpolation.vti
diff --git a/Ready/Patterns/kernel_test.vti b/Patterns/kernel_test.vti
similarity index 100%
rename from Ready/Patterns/kernel_test.vti
rename to Patterns/kernel_test.vti
diff --git a/Ready/Patterns/lion.vtu b/Patterns/lion.vtu
similarity index 100%
rename from Ready/Patterns/lion.vtu
rename to Patterns/lion.vtu
diff --git a/Ready/Patterns/oregonator.vti b/Patterns/oregonator.vti
similarity index 100%
rename from Ready/Patterns/oregonator.vti
rename to Patterns/oregonator.vti
diff --git a/Ready/Patterns/parameter_modulation_demo.vti b/Patterns/parameter_modulation_demo.vti
similarity index 100%
rename from Ready/Patterns/parameter_modulation_demo.vti
rename to Patterns/parameter_modulation_demo.vti
diff --git a/Ready/Patterns/parameter_modulation_demo2.vti b/Patterns/parameter_modulation_demo2.vti
similarity index 100%
rename from Ready/Patterns/parameter_modulation_demo2.vti
rename to Patterns/parameter_modulation_demo2.vti
diff --git a/Ready/Patterns/wave_equation.vti b/Patterns/wave_equation.vti
similarity index 100%
rename from Ready/Patterns/wave_equation.vti
rename to Patterns/wave_equation.vti
diff --git a/README.txt b/README.txt
index 85fbae587..dd9508b2e 100644
--- a/README.txt
+++ b/README.txt
@@ -1,69 +1,35 @@
-    Copyright (C) 1991, 2010, Greg Turk, Tim Hutton
-    
-    This program is free software: you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation, either version 3 of the License, or
-    (at your option) any later version.
+---------------------------------------------------------------------------------
+                                    About
+---------------------------------------------------------------------------------
 
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
+For help, visit: http://code.google.com/p/reaction-diffusion/
+Or email: reaction-diffusion@googlegroups.com
 
-    You should have received a copy of the GNU General Public License
-    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+Ready is free software. Help us improve it!
 
-------------------------
+For build instructions, see BUILD.txt.
 
-This is a port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
+See Help/credits.html for a list of credits and acknowledgements.
 
-The following copyright notice appeared with the original source code. 
-Greg Turk has given permission for the code to be distributed under the GPL.
+---------------------------------------------------------------------------------
+                                   License
+---------------------------------------------------------------------------------
 
-==begin==
+Copyright 2011, 2012, 2013 The Ready Bunch
 
-Make spots and stripes with reaction-diffusion.
+The Ready Bunch is: Tim Hutton, Robert Munafo, Andrew Trevorrow, Tom Rokicki, Dan Wills
 
-The spot-formation system is described in the article:
+This file is part of Ready.
 
-  "A Model for Generating Aspects of Zebra and Other Mammailian
-   Coat Patterns"
-  Jonathan B. L. Bard
-  Journal of Theoretical Biology, Vol. 93, No. 2, pp. 363-385
-  (November 1981)
+Ready is free software: you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation, either version 3 of the License, or
+(at your option) any later version.
 
-The stripe-formation system is described in the book:
-
-  Models of Biological Pattern Formation
-  Hans Meinhardt
-  Academic Press, 1982
-
-
-Permission is granted to modify and/or distribute this program so long
-as the program is distributed free of charge and this header is retained
-as part of the program.
-
-Copyright (c) Greg Turk, 1991
-
-==end==
-
-------------------------
-
-Build instructions:
-
-1) install CMake and OpenCV
-2) run CMake to generate the build files for your chosen compiler
-3) build
-
-Tested on Linux and Windows XP. Should work on all operating systems.
-
-There's a SpeedComparisons folder which isn't built by default. If you're interested in OpenMP, OpenCL etc. then you can build it.
-
-------------------------
-
-Contact:
-
-Tim Hutton <tim.hutton@gmail.com>
-Greg Turk <turk.gatech@gmail.com>
+Ready is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
 
+See COPYING.txt for the full GNU General Public License.
+Or visit <http://www.gnu.org/licenses/>.
diff --git a/Ready/CMakeLists.txt b/Ready/CMakeLists.txt
deleted file mode 100644
index b5a025f0a..000000000
--- a/Ready/CMakeLists.txt
+++ /dev/null
@@ -1,459 +0,0 @@
-#  Copyright 2011, 2012, 2013 The Ready Bunch
-#
-#  This file is part of Ready.
-#
-#  Ready is free software: you can redistribute it and/or modify
-#  it under the terms of the GNU General Public License as published by
-#  the Free Software Foundation, either version 3 of the License, or
-#  (at your option) any later version.
-#
-#  Ready is distributed in the hope that it will be useful,
-#  but WITHOUT ANY WARRANTY; without even the implied warranty of
-#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-#  GNU General Public License for more details.
-#
-#  You should have received a copy of the GNU General Public License
-#  along with Ready. If not, see <http://www.gnu.org/licenses/>.
-#
-#--------------------------------------------------------------------------
-    
-cmake_minimum_required( VERSION 2.6 )
-cmake_policy( SET CMP0003 NEW )
-
-project( Ready )
-
-set( READY_VERSION 0.6 )   # check matches Help/about.html
-add_definitions( -D READY_VERSION=${READY_VERSION} )
-
-if( APPLE OR WIN32 )
-  # app names are usually capitalized on Mac OS X and Windows
-  set( APP_NAME Ready )
-else()
-  # Linux binaries are usually all lowercase
-  set( APP_NAME ready )
-endif()
-set( CMD_NAME rdy ) # command-line version
-
-#-------------------------------------------source files----------------------------------------------
-
-set( BASE_SOURCES      # low-level code used in all executables
-  src/readybase/AbstractRD.hpp                src/readybase/AbstractRD.cpp
-  src/readybase/ImageRD.hpp                   src/readybase/ImageRD.cpp
-  src/readybase/GrayScottImageRD.hpp          src/readybase/GrayScottImageRD.cpp
-  src/readybase/OpenCLImageRD.hpp             src/readybase/OpenCLImageRD.cpp
-  src/readybase/FormulaOpenCLImageRD.hpp      src/readybase/FormulaOpenCLImageRD.cpp
-  src/readybase/FullKernelOpenCLImageRD.hpp   src/readybase/FullKernelOpenCLImageRD.cpp
-  src/readybase/MeshRD.hpp                    src/readybase/MeshRD.cpp
-  src/readybase/GrayScottMeshRD.hpp           src/readybase/GrayScottMeshRD.cpp
-  src/readybase/OpenCLMeshRD.hpp              src/readybase/OpenCLMeshRD.cpp
-  src/readybase/FormulaOpenCLMeshRD.hpp       src/readybase/FormulaOpenCLMeshRD.cpp
-  src/readybase/FullKernelOpenCLMeshRD.hpp    src/readybase/FullKernelOpenCLMeshRD.cpp
-  src/readybase/OpenCL_MixIn.hpp              src/readybase/OpenCL_MixIn.cpp
-  src/readybase/OpenCL_utils.hpp              src/readybase/OpenCL_utils.cpp
-  src/readybase/IO_XML.hpp                    src/readybase/IO_XML.cpp
-  src/readybase/overlays.hpp                  src/readybase/overlays.cpp
-  src/readybase/Properties.hpp                src/readybase/Properties.cpp
-  src/readybase/utils.hpp                     src/readybase/utils.cpp
-  src/readybase/OpenCL_Dyn_Load.h             src/readybase/OpenCL_Dyn_Load.c 
-  src/readybase/MeshGenerators.hpp            src/readybase/MeshGenerators.cpp
-  src/readybase/SystemFactory.hpp             src/readybase/SystemFactory.cpp
-  src/readybase/scene_items.hpp               src/readybase/scene_items.cpp
-)
-include_directories( src/readybase )
-
-set( GUI_SOURCES      # high-level GUI code used only in Ready
-  src/gui/IDs.hpp
-  src/gui/wxutils.hpp                      src/gui/wxutils.cpp
-  src/gui/dialogs.hpp                      src/gui/dialogs.cpp
-  src/gui/prefs.hpp                        src/gui/prefs.cpp
-  src/gui/app.hpp                          src/gui/app.cpp 
-  src/gui/frame.hpp                        src/gui/frame.cpp 
-  src/gui/HelpPanel.hpp                    src/gui/HelpPanel.cpp
-  src/gui/InfoPanel.hpp                    src/gui/InfoPanel.cpp
-  src/gui/PatternsPanel.hpp                src/gui/PatternsPanel.cpp
-  src/gui/vtk_pipeline.hpp                 src/gui/vtk_pipeline.cpp
-  src/gui/InteractorStylePainter.hpp       src/gui/InteractorStylePainter.cpp
-  src/gui/wxVTKRenderWindowInteractor.h    src/gui/wxVTKRenderWindowInteractor.cxx
-  src/gui/RecordingDialog.hpp              src/gui/RecordingDialog.cpp
-)
-include_directories( src/gui )
-
-set( CMD_SOURCES      # code used for the command-line version
-  src/cmd/main.cpp
-)
-
-set( RESOURCES
-  resources/ready.rc
-  resources/appicon.ico 
-  resources/appicon16.ico 
-  resources/appicon32.ico 
-  resources/appicon48.ico 
-  resources/appicon.xpm
-  resources/Info.plist.in
-  resources/app.icns 
-  resources/file.icns
-)
-include_directories( resources )
-
-set( PATTERN_FILES
-  Patterns/Meinhardt1982/stripes.vti       Patterns/Meinhardt1982/zebra.vtu
-  Patterns/Schlogl.vti 
-  Patterns/heat_equation.vti 
-  Patterns/Turing1952/spots.vti            Patterns/Turing1952/spots_noisy.vti
-  Patterns/kernel_test.vti 
-  Patterns/parameter_modulation_demo.vti
-  Patterns/parameter_modulation_demo2.vti
-  Patterns/bunny.vtu
-  Patterns/lion.vtu
-  Patterns/heat_equation_interpolation.vti
-  Patterns/Ginzburg-Landau/complex_Ginzburg-Landau.vti
-  Patterns/Ginzburg-Landau/complex_Ginzburg-Landau_magnitude.vti
-  Patterns/wave_equation.vti
-  Patterns/oregonator.vti
-  Patterns/Brusselator.vti
-  Patterns/SmoothLife/smoothglider.vti     Patterns/SmoothLife/smoothlifeL.vti     Patterns/SmoothLife/glider_3D.vti
-  Patterns/Purwins1999/glider.vti          Patterns/Purwins1999/glider_3D.vti      Patterns/Purwins1999/multiGlider.vti
-  Patterns/CPU-only/grayscott_1D.vti 
-  Patterns/CPU-only/grayscott_2D.vti 
-  Patterns/CPU-only/grayscott_3D.vti
-  Patterns/FitzHugh-Nagumo/tip-splitting.vti 
-  Patterns/FitzHugh-Nagumo/spiral_turbulence.vti
-  Patterns/FitzHugh-Nagumo/pulsate.vti 
-  Patterns/FitzHugh-Nagumo/squid_axon.vti
-  Patterns/FitzHugh-Nagumo/Ising_regime.vti
-  Patterns/Gray-Scott/Lesmes_noisy.vti
-  Patterns/Gray-Scott/noisy_solitons_mitosis.vti
-  Patterns/Gray-Scott/parameter-map.vti
-  Patterns/Gray-Scott/Pearson1993.vti
-  Patterns/Gray-Scott/self-replicating_spots.vti 
-  Patterns/Gray-Scott/U-Skate/Hutton-and-helix-gliders.vti
-  Patterns/Gray-Scott/U-Skate/Munafo_glider.vti
-  Patterns/Gray-Scott/U-Skate/o-ring_2D.vti
-  Patterns/CellularAutomata/Bays_3D.vti
-  Patterns/CellularAutomata/Conway_life.vti 
-  Patterns/CellularAutomata/life_torus.vtu
-  Patterns/CellularAutomata/larger-than-life.vti
-  Patterns/CellularAutomata/Buss_hex.vtu
-  Patterns/CellularAutomata/tri_life.vtu
-  Patterns/CellularAutomata/hex_B2oS2m34_gliders.vtu
-  Patterns/CellularAutomata/PenroseTilings/life.vtu
-  Patterns/CellularAutomata/PenroseTilings/life_oscillators.vtu
-  Patterns/CellularAutomata/PenroseTilings/Goucher_glider.vtu
-  Patterns/CellularAutomata/PenroseTilings/Imai_glider_B2SC4.vtu
-  Patterns/CellularAutomata/PenroseTilings/Goucher_loops.vtu
-  Patterns/CellularAutomata/Salt/salt2D_demo.vti
-  Patterns/CellularAutomata/Salt/salt3D_circular330.vti
-  Patterns/Yang2002/Yang_1.vti    Patterns/Yang2002/Yang_2b.vti  Patterns/Yang2002/Yang_2c.vti
-  Patterns/Yang2002/Yang_2d.vti   Patterns/Yang2002/Yang_3a.vti  Patterns/Yang2002/Yang_3b.vti
-  Patterns/Yang2002/Yang_3c.vti   Patterns/Yang2002/Yang_3d.vti  Patterns/Yang2002/Yang_4.vti
-  Patterns/Yang2003/Fig2.vti
-  Patterns/Yang2003/Fig3a.vti     Patterns/Yang2003/Fig3b.vti    Patterns/Yang2003/Fig3c.vti
-  Patterns/McCabe/McCabe.vti               Patterns/McCabe/McCabe_simple.vti
-  Patterns/McCabe/McCabe_additive2b.vti    Patterns/McCabe/McCabe_additive2a.vti
-  Patterns/Kytta2007/Fig5.7a.vti  Patterns/Kytta2007/Fig5.7c.vti
-  Patterns/Kytta2007/Fig5.8c.vti  Patterns/Kytta2007/Fig5.8d.vti Patterns/Kytta2007/Fig5.8e.vti
-  Patterns/Kytta2007/Fig5.8f.vti  Patterns/Kytta2007/Fig5.8g.vti
-  Patterns/Yang2006/jumping.vti   Patterns/Yang2006/jumping_cGL.vti
-  Patterns/Schrodinger1926/packet.vti             Patterns/Schrodinger1926/packet_reflect.vti
-  Patterns/Schrodinger1926/packet_pass.vti        Patterns/Schrodinger1926/quantum_tunnelling.vti
-  Patterns/Schrodinger1926/packet_reflect2D.vti
-  Patterns/Experiments/mutually-catalytic_spots.vti
-  Patterns/Experiments/cglrd_ramps_example_djw.vti  
-  Patterns/Experiments/grayscott-historyWave_fuseWorms.vti
-  Patterns/Experiments/grayscott-historyWave_moreLifelike.vti
-  Patterns/Experiments/grayscott-historyWaveDC_solitonsAndWorms_init.vti
-  Patterns/Experiments/orbits_explodey_init.djw.vti
-  Patterns/Experiments/orbits_sharpWaves-init_djw.vti
-  Patterns/Experiments/grayscott-historyWave_coralGrow_djw.vti
-  Patterns/Experiments/gladman_vermiformSolitons.vti
-)
-
-set( HELP_FILES
-  Help/about.gif              Help/about.html
-  Help/action.html            Help/credits.html 
-  Help/file.html              Help/help.html 
-  Help/mouse.html             Help/quickstart.html 
-  Help/tips.html              Help/changes.html 
-  Help/edit.html              Help/formats.html 
-  Help/index.html             Help/problems.html 
-  Help/view.html              Help/introduction.html
-)
-
-set( OTHER_FILES
-  ./README.txt 
-  ./COPYING.txt 
-  ./TODO.txt 
-  ./BUILD.txt 
-  ./CMakeLists.txt 
-  src/FindOpenCL.cmake 
-  src/Doxyfile.in
-  resources/logo.png
-  resources/Icons/22px/icon-pointer.png
-  resources/Icons/22px/draw-freehand.png
-  resources/Icons/22px/draw-brush.png
-  resources/Icons/22px/color-picker.png
-  resources/Icons/22px/document-new.png
-  resources/Icons/22px/document-open.png
-  resources/Icons/22px/document-save.png
-  resources/Icons/22px/document-revert.png
-  resources/Icons/22px/media-playback-start_green.png
-  resources/Icons/22px/media-playback-pause_red.png
-  resources/Icons/22px/media-seek-forward.png
-  resources/Icons/22px/media-seek-backward.png
-  resources/Icons/22px/media-skip-backward_modified.png
-  resources/Icons/22px/media-record.png
-  resources/Icons/22px/system-run.png
-  resources/Icons/22px/list-add_gray.png
-  resources/Icons/22px/camera-photo.png
-  resources/Icons/32px/icon-pointer.png
-  resources/Icons/32px/draw-freehand.png
-  resources/Icons/32px/draw-brush.png
-  resources/Icons/32px/color-picker.png
-  resources/Icons/32px/document-new.png
-  resources/Icons/32px/document-open.png
-  resources/Icons/32px/document-save.png
-  resources/Icons/32px/document-revert.png
-  resources/Icons/32px/media-playback-start_green.png
-  resources/Icons/32px/media-playback-pause_red.png
-  resources/Icons/32px/media-seek-forward.png
-  resources/Icons/32px/media-seek-backward.png
-  resources/Icons/32px/media-skip-backward_modified.png
-  resources/Icons/32px/media-record.png
-  resources/Icons/32px/system-run.png
-  resources/Icons/32px/list-add_gray.png
-  resources/Icons/32px/camera-photo.png
-  resources/Cursors/pencil-cursor.png
-  resources/Cursors/brush-cursor.png
-  resources/Cursors/picker-cursor.png
-)
-
-#-------------------------------------------VTK----------------------------------------------
-
-find_package( VTK )
-if( VTK_FOUND )
-  include( ${VTK_USE_FILE} )
-else()
-  message(FATAL_ERROR "Cannot build the executable without VTK. Please set the VTK variables.")
-endif()
-
-#-------------------------------------------wxVTK----------------------------------------------
-
-# The following allows you to access wxGLCanvas for GTK
-IF(WIN32)
-  SET(GUI_EXECUTABLE WIN32)
-ELSE(WIN32)
-  IF(APPLE)
-    SET(GUI_EXECUTABLE MACOSX_BUNDLE)
-    IF(VTK_USE_COCOA)
-      SET_SOURCE_FILES_PROPERTIES(
-        src/gui/wxVTKRenderWindowInteractor.cxx
-        PROPERTIES COMPILE_FLAGS "-ObjC++")
-    ENDIF(VTK_USE_COCOA)
-  ELSE(APPLE)
-    # Ok X11 for sure, but just check:
-    IF(NOT VTK_USE_X)
-      MESSAGE(FATAL_ERROR "You need to have VTK_USE_X")
-    ENDIF(NOT VTK_USE_X)
-    # CMake 2.6:
-    # technically those packages are not required since one can still use the Motif/X11 version and not the gtk one:
-    FIND_PACKAGE(PkgConfig)
-    pkg_check_modules (GTK2 gtk+-2.0)
-    #MESSAGE("${GTK2_INCLUDE_DIRS}")
-    INCLUDE_DIRECTORIES(${GTK2_INCLUDE_DIRS})
-    LINK_LIBRARIES(${GTK2_LIBRARIES})
-    # Can I require all my user to have the gl lib on linux, even if they do not really need it...
-    SET(WXGLCANVASLIBS "gl")
-  ENDIF(APPLE)
-ENDIF(WIN32)
-
-#-------------------------------------------wxWidgets----------------------------------------------
-
-if( APPLE )
-  # on Mac OS X it's better to use locally installed wxWidgets headers and libs
-  # (the pre-installed stuff tends to be out of date; eg. 10.6 has wxMac 2.8.8 and it's a 32-bit debug build)
-  set( wxWidgets_CONFIG_EXECUTABLE /usr/local/bin/wx-config )
-  set( wxWidgets_wxrc_EXECUTABLE /usr/local/bin/wxrc )         # not used, but no harm leaving it in
-elseif(UNIX)
-  # remove -rdynamic from link options on Linux to reduce size by about 1.2MB
-  set( CMAKE_SHARED_LIBRARY_LINK_CXX_FLAGS "" )
-endif()
-
-# wxWidgets is required to build the project
-FIND_PACKAGE( wxWidgets COMPONENTS html aui ${WXGLCANVASLIBS} core adv base )
-
-IF( wxWidgets_FOUND )
-  INCLUDE( ${wxWidgets_USE_FILE} )
-ELSE()
-  MESSAGE( FATAL_ERROR "Cannot build the executable without wxWidgets. Please set the wxWidgets variables." )
-ENDIF()
-
-if( WIN32 )
-  # prevent link errors with wxMSW 2.9.x
-  add_definitions( -DwxDEBUG_LEVEL=0 )
-endif()
-
-#-------------------------------------------OpenCL----------------------------------------------
-
-set( CMAKE_MODULE_PATH ${Ready_SOURCE_DIR}/src )
-# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
-
-# we need to build against OpenCL
-find_package( OpenCL REQUIRED )
-include_directories( ${OPENCL_INCLUDE_DIRS} )
-if( APPLE )
-  link_libraries( ${OPENCL_LIBRARIES} ) # on MacOSX we assume that OpenCL is available (might need to rethink for versions before 10.6)
-endif()
-  
-#---------------copy installation files to build folder (helps with testing)--------------------
-
-foreach( file ${PATTERN_FILES} ${HELP_FILES} ${RESOURCES} ${OTHER_FILES} )
-  add_custom_command( 
-    OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/${file}"
-    COMMAND cmake -E copy "${CMAKE_CURRENT_SOURCE_DIR}/${file}" "${CMAKE_CURRENT_BINARY_DIR}/${file}"
-    DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/${file}"
-  )
-  list( APPEND files_dest "${CMAKE_CURRENT_BINARY_DIR}/${file}" )
-endforeach() 
-
-add_custom_target( CopyFiles ALL DEPENDS ${files_dest} )
-
-#-------------------------------- build ------------------------------------------------------
-
-# ensure we link the C runtime statically (N.B. still appears as /MD in the CMake gui but ignore this)
-# see: http://www.cmake.org/Wiki/CMake_FAQ#Dynamic_Replace
-foreach( var CMAKE_CXX_FLAGS CMAKE_CXX_FLAGS_DEBUG CMAKE_CXX_FLAGS_RELEASE CMAKE_CXX_FLAGS_MINSIZEREL CMAKE_CXX_FLAGS_RELWITHDEBINFO )
-  string( REGEX REPLACE "/MD" "/MT" ${var} "${${var}}" )
-endforeach()
-
-set( USE_SSE "YES" )
-if( USE_SSE )
-  # enable SSE to allow us to set flags to avoid denormals, and relax the floating-point accuracy for speed
-  if( MSVC )
-    if( CMAKE_CL_64 )
-      set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} /fp:fast" )
-      set( CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /fp:fast" )
-    else() # avoid warning on x64, which always comes with SSE2
-      set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} /arch:SSE2 /fp:fast" )
-      set( CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:SSE2 /fp:fast" )
-    endif()
-  else()
-    add_definitions( -msse2 -ffast-math )
-  endif()
-  add_definitions( -DUSE_SSE )
-endif()
-
-if( APPLE )
-  # support Mac OS 10.5 or later
-  add_definitions( -mmacosx-version-min=10.5 )
-endif()
-if( APPLE OR UNIX )
-  # use same settings as in makefiles
-  add_definitions( -D_LARGE_FILES )
-endif()
-
-# create base library used by all executables
-add_library( readybase STATIC ${BASE_SOURCES} )
-if( ${VTK_MAJOR_VERSION} GREATER 5 )
-  target_link_libraries( 
-    readybase 
-    vtkCommonCore 
-    vtkFiltersModeling
-    vtkFiltersTexture
-    vtkInteractionStyle 
-    vtkIOXML 
-    vtkRenderingCore 
-    vtkRenderingAnnotation 
-    vtkRenderingFreeType 
-    vtkRenderingFreeTypeOpenGL 
-  )
-else()
-  target_link_libraries( readybase vtkCommon vtkGraphics vtkIO vtkRendering vtkHybrid )
-endif()
-
-# create command-line utility
-add_executable( ${CMD_NAME} ${CMD_SOURCES} )
-target_link_libraries( ${CMD_NAME} readybase )
-
-# create GUI application
-add_executable( ${APP_NAME} ${GUI_EXECUTABLE} ${GUI_SOURCES} ${RESOURCES} )
-target_link_libraries( ${APP_NAME} readybase ${wxWidgets_LIBRARIES} )
-
-if( APPLE )
-  # create Info.plist (using Info.plist.in) and PkgInfo files inside .app bundle
-  add_custom_target( app_bundle
-    COMMAND sed -e "s/VERSION/${READY_VERSION}/" ${CMAKE_SOURCE_DIR}/resources/Info.plist.in >Ready.app/Contents/Info.plist
-    COMMAND echo -n "APPLReDy" >Ready.app/Contents/PkgInfo
-  )
-  add_dependencies( ${APP_NAME} app_bundle )
-   
-  # copy *.icns files into Resources directory inside .app bundle
-  set_source_files_properties( ${CMAKE_SOURCE_DIR}/resources/app.icns PROPERTIES MACOSX_PACKAGE_LOCATION Resources )
-  set_source_files_properties( ${CMAKE_SOURCE_DIR}/resources/file.icns PROPERTIES MACOSX_PACKAGE_LOCATION Resources )
-   
-  # remove unreachable functions and data, and don't add debug info (reduces app size by about 12MB)
-  target_link_libraries( ${APP_NAME} -Wl,-dead_strip -Wl,-S )
-endif()
-
-# (Visual Studio only) put the executable in the root binary folder, not in "Debug" or "Release"
-# http://stackoverflow.com/questions/543203/cmake-runtime-output-directory-on-windows
-if( MSVC_IDE )
-  set_target_properties( ${APP_NAME} PROPERTIES PREFIX "../" )
-  set_target_properties( ${CMD_NAME} PROPERTIES PREFIX "../" )
-endif()
-
-# avoid security warnings
-if( MSVC )
-  add_definitions( /D_CRT_SECURE_NO_WARNINGS /D_CRT_NONSTDC_NO_WARNINGS )
-endif()
-
-if( CMAKE_COMPILER_IS_GNUCXX )
-  list( APPEND CMAKE_EXE_LINKER_FLAGS_RELEASE "-s" )  # strip release binary, for smaller file size
-endif()
-
-#----------------------------------------doxygen------------------------------------------------
-
-find_package( Doxygen )
-if(DOXYGEN_FOUND)
-  configure_file(${CMAKE_CURRENT_SOURCE_DIR}/src/Doxyfile.in ${CMAKE_CURRENT_BINARY_DIR}/Doxyfile @ONLY)
-  add_custom_target(doc
-    ${DOXYGEN_EXECUTABLE} ${CMAKE_CURRENT_BINARY_DIR}/Doxyfile
-    WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
-  )
-endif()
-# (if doxygen is found then 'make doc' should produce html documentation of the source code)
-
-#----------------------------------------install------------------------------------------------
-
-# put Ready in the root of the installation folder instead of in "bin"
-install( TARGETS ${APP_NAME} DESTINATION "." )      # (add ${CMD_NAME} if we want to distribute the command-line version too)
-
-# install our source files, resource files, pattern files, help files and text files
-foreach( source_file ${BASE_SOURCES} ${GUI_SOURCES} ${CMD_SOURCES} ${RESOURCES} ${PATTERN_FILES} ${HELP_FILES} ${OTHER_FILES} )
-  get_filename_component( path_name "${source_file}" PATH )
-  install( FILES "${source_file}" DESTINATION ${path_name} )
-endforeach()
-
-#----------------------------------------package----------------------------------------------
-
-if( APPLE )
-  set( CPACK_SYSTEM_NAME "Mac" )  # nicer than "Darwin"
-elseif( UNIX )
-  if( CMAKE_SIZEOF_VOID_P EQUAL 8 )
-    set( CPACK_SYSTEM_NAME "Linux-64bit" )  
-  else()
-    set( CPACK_SYSTEM_NAME "Linux-32bit" )  
-  endif()
-elseif( WIN32)
-  if( CMAKE_SIZEOF_VOID_P EQUAL 8 )
-    set( CPACK_SYSTEM_NAME "Windows-64bit" )  
-  else()
-    set( CPACK_SYSTEM_NAME "Windows-32bit" )  
-  endif()
-endif()
-if( NOT USE_SSE )
-  set( CPACK_SYSTEM_NAME "${CPACK_SYSTEM_NAME}-noSSE" )
-endif()
-set( CPACK_GENERATOR "ZIP" )
-set( CPACK_PACKAGE_VERSION "${READY_VERSION}" )
-set( CPACK_SOURCE_GENERATOR "ZIP" )
-include( CPack )
diff --git a/Ready/COPYING.txt b/Ready/COPYING.txt
deleted file mode 100644
index 94a9ed024..000000000
--- a/Ready/COPYING.txt
+++ /dev/null
@@ -1,674 +0,0 @@
-                    GNU GENERAL PUBLIC LICENSE
-                       Version 3, 29 June 2007
-
- Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
- Everyone is permitted to copy and distribute verbatim copies
- of this license document, but changing it is not allowed.
-
-                            Preamble
-
-  The GNU General Public License is a free, copyleft license for
-software and other kinds of works.
-
-  The licenses for most software and other practical works are designed
-to take away your freedom to share and change the works.  By contrast,
-the GNU General Public License is intended to guarantee your freedom to
-share and change all versions of a program--to make sure it remains free
-software for all its users.  We, the Free Software Foundation, use the
-GNU General Public License for most of our software; it applies also to
-any other work released this way by its authors.  You can apply it to
-your programs, too.
-
-  When we speak of free software, we are referring to freedom, not
-price.  Our General Public Licenses are designed to make sure that you
-have the freedom to distribute copies of free software (and charge for
-them if you wish), that you receive source code or can get it if you
-want it, that you can change the software or use pieces of it in new
-free programs, and that you know you can do these things.
-
-  To protect your rights, we need to prevent others from denying you
-these rights or asking you to surrender the rights.  Therefore, you have
-certain responsibilities if you distribute copies of the software, or if
-you modify it: responsibilities to respect the freedom of others.
-
-  For example, if you distribute copies of such a program, whether
-gratis or for a fee, you must pass on to the recipients the same
-freedoms that you received.  You must make sure that they, too, receive
-or can get the source code.  And you must show them these terms so they
-know their rights.
-
-  Developers that use the GNU GPL protect your rights with two steps:
-(1) assert copyright on the software, and (2) offer you this License
-giving you legal permission to copy, distribute and/or modify it.
-
-  For the developers' and authors' protection, the GPL clearly explains
-that there is no warranty for this free software.  For both users' and
-authors' sake, the GPL requires that modified versions be marked as
-changed, so that their problems will not be attributed erroneously to
-authors of previous versions.
-
-  Some devices are designed to deny users access to install or run
-modified versions of the software inside them, although the manufacturer
-can do so.  This is fundamentally incompatible with the aim of
-protecting users' freedom to change the software.  The systematic
-pattern of such abuse occurs in the area of products for individuals to
-use, which is precisely where it is most unacceptable.  Therefore, we
-have designed this version of the GPL to prohibit the practice for those
-products.  If such problems arise substantially in other domains, we
-stand ready to extend this provision to those domains in future versions
-of the GPL, as needed to protect the freedom of users.
-
-  Finally, every program is threatened constantly by software patents.
-States should not allow patents to restrict development and use of
-software on general-purpose computers, but in those that do, we wish to
-avoid the special danger that patents applied to a free program could
-make it effectively proprietary.  To prevent this, the GPL assures that
-patents cannot be used to render the program non-free.
-
-  The precise terms and conditions for copying, distribution and
-modification follow.
-
-                       TERMS AND CONDITIONS
-
-  0. Definitions.
-
-  "This License" refers to version 3 of the GNU General Public License.
-
-  "Copyright" also means copyright-like laws that apply to other kinds of
-works, such as semiconductor masks.
-
-  "The Program" refers to any copyrightable work licensed under this
-License.  Each licensee is addressed as "you".  "Licensees" and
-"recipients" may be individuals or organizations.
-
-  To "modify" a work means to copy from or adapt all or part of the work
-in a fashion requiring copyright permission, other than the making of an
-exact copy.  The resulting work is called a "modified version" of the
-earlier work or a work "based on" the earlier work.
-
-  A "covered work" means either the unmodified Program or a work based
-on the Program.
-
-  To "propagate" a work means to do anything with it that, without
-permission, would make you directly or secondarily liable for
-infringement under applicable copyright law, except executing it on a
-computer or modifying a private copy.  Propagation includes copying,
-distribution (with or without modification), making available to the
-public, and in some countries other activities as well.
-
-  To "convey" a work means any kind of propagation that enables other
-parties to make or receive copies.  Mere interaction with a user through
-a computer network, with no transfer of a copy, is not conveying.
-
-  An interactive user interface displays "Appropriate Legal Notices"
-to the extent that it includes a convenient and prominently visible
-feature that (1) displays an appropriate copyright notice, and (2)
-tells the user that there is no warranty for the work (except to the
-extent that warranties are provided), that licensees may convey the
-work under this License, and how to view a copy of this License.  If
-the interface presents a list of user commands or options, such as a
-menu, a prominent item in the list meets this criterion.
-
-  1. Source Code.
-
-  The "source code" for a work means the preferred form of the work
-for making modifications to it.  "Object code" means any non-source
-form of a work.
-
-  A "Standard Interface" means an interface that either is an official
-standard defined by a recognized standards body, or, in the case of
-interfaces specified for a particular programming language, one that
-is widely used among developers working in that language.
-
-  The "System Libraries" of an executable work include anything, other
-than the work as a whole, that (a) is included in the normal form of
-packaging a Major Component, but which is not part of that Major
-Component, and (b) serves only to enable use of the work with that
-Major Component, or to implement a Standard Interface for which an
-implementation is available to the public in source code form.  A
-"Major Component", in this context, means a major essential component
-(kernel, window system, and so on) of the specific operating system
-(if any) on which the executable work runs, or a compiler used to
-produce the work, or an object code interpreter used to run it.
-
-  The "Corresponding Source" for a work in object code form means all
-the source code needed to generate, install, and (for an executable
-work) run the object code and to modify the work, including scripts to
-control those activities.  However, it does not include the work's
-System Libraries, or general-purpose tools or generally available free
-programs which are used unmodified in performing those activities but
-which are not part of the work.  For example, Corresponding Source
-includes interface definition files associated with source files for
-the work, and the source code for shared libraries and dynamically
-linked subprograms that the work is specifically designed to require,
-such as by intimate data communication or control flow between those
-subprograms and other parts of the work.
-
-  The Corresponding Source need not include anything that users
-can regenerate automatically from other parts of the Corresponding
-Source.
-
-  The Corresponding Source for a work in source code form is that
-same work.
-
-  2. Basic Permissions.
-
-  All rights granted under this License are granted for the term of
-copyright on the Program, and are irrevocable provided the stated
-conditions are met.  This License explicitly affirms your unlimited
-permission to run the unmodified Program.  The output from running a
-covered work is covered by this License only if the output, given its
-content, constitutes a covered work.  This License acknowledges your
-rights of fair use or other equivalent, as provided by copyright law.
-
-  You may make, run and propagate covered works that you do not
-convey, without conditions so long as your license otherwise remains
-in force.  You may convey covered works to others for the sole purpose
-of having them make modifications exclusively for you, or provide you
-with facilities for running those works, provided that you comply with
-the terms of this License in conveying all material for which you do
-not control copyright.  Those thus making or running the covered works
-for you must do so exclusively on your behalf, under your direction
-and control, on terms that prohibit them from making any copies of
-your copyrighted material outside their relationship with you.
-
-  Conveying under any other circumstances is permitted solely under
-the conditions stated below.  Sublicensing is not allowed; section 10
-makes it unnecessary.
-
-  3. Protecting Users' Legal Rights From Anti-Circumvention Law.
-
-  No covered work shall be deemed part of an effective technological
-measure under any applicable law fulfilling obligations under article
-11 of the WIPO copyright treaty adopted on 20 December 1996, or
-similar laws prohibiting or restricting circumvention of such
-measures.
-
-  When you convey a covered work, you waive any legal power to forbid
-circumvention of technological measures to the extent such circumvention
-is effected by exercising rights under this License with respect to
-the covered work, and you disclaim any intention to limit operation or
-modification of the work as a means of enforcing, against the work's
-users, your or third parties' legal rights to forbid circumvention of
-technological measures.
-
-  4. Conveying Verbatim Copies.
-
-  You may convey verbatim copies of the Program's source code as you
-receive it, in any medium, provided that you conspicuously and
-appropriately publish on each copy an appropriate copyright notice;
-keep intact all notices stating that this License and any
-non-permissive terms added in accord with section 7 apply to the code;
-keep intact all notices of the absence of any warranty; and give all
-recipients a copy of this License along with the Program.
-
-  You may charge any price or no price for each copy that you convey,
-and you may offer support or warranty protection for a fee.
-
-  5. Conveying Modified Source Versions.
-
-  You may convey a work based on the Program, or the modifications to
-produce it from the Program, in the form of source code under the
-terms of section 4, provided that you also meet all of these conditions:
-
-    a) The work must carry prominent notices stating that you modified
-    it, and giving a relevant date.
-
-    b) The work must carry prominent notices stating that it is
-    released under this License and any conditions added under section
-    7.  This requirement modifies the requirement in section 4 to
-    "keep intact all notices".
-
-    c) You must license the entire work, as a whole, under this
-    License to anyone who comes into possession of a copy.  This
-    License will therefore apply, along with any applicable section 7
-    additional terms, to the whole of the work, and all its parts,
-    regardless of how they are packaged.  This License gives no
-    permission to license the work in any other way, but it does not
-    invalidate such permission if you have separately received it.
-
-    d) If the work has interactive user interfaces, each must display
-    Appropriate Legal Notices; however, if the Program has interactive
-    interfaces that do not display Appropriate Legal Notices, your
-    work need not make them do so.
-
-  A compilation of a covered work with other separate and independent
-works, which are not by their nature extensions of the covered work,
-and which are not combined with it such as to form a larger program,
-in or on a volume of a storage or distribution medium, is called an
-"aggregate" if the compilation and its resulting copyright are not
-used to limit the access or legal rights of the compilation's users
-beyond what the individual works permit.  Inclusion of a covered work
-in an aggregate does not cause this License to apply to the other
-parts of the aggregate.
-
-  6. Conveying Non-Source Forms.
-
-  You may convey a covered work in object code form under the terms
-of sections 4 and 5, provided that you also convey the
-machine-readable Corresponding Source under the terms of this License,
-in one of these ways:
-
-    a) Convey the object code in, or embodied in, a physical product
-    (including a physical distribution medium), accompanied by the
-    Corresponding Source fixed on a durable physical medium
-    customarily used for software interchange.
-
-    b) Convey the object code in, or embodied in, a physical product
-    (including a physical distribution medium), accompanied by a
-    written offer, valid for at least three years and valid for as
-    long as you offer spare parts or customer support for that product
-    model, to give anyone who possesses the object code either (1) a
-    copy of the Corresponding Source for all the software in the
-    product that is covered by this License, on a durable physical
-    medium customarily used for software interchange, for a price no
-    more than your reasonable cost of physically performing this
-    conveying of source, or (2) access to copy the
-    Corresponding Source from a network server at no charge.
-
-    c) Convey individual copies of the object code with a copy of the
-    written offer to provide the Corresponding Source.  This
-    alternative is allowed only occasionally and noncommercially, and
-    only if you received the object code with such an offer, in accord
-    with subsection 6b.
-
-    d) Convey the object code by offering access from a designated
-    place (gratis or for a charge), and offer equivalent access to the
-    Corresponding Source in the same way through the same place at no
-    further charge.  You need not require recipients to copy the
-    Corresponding Source along with the object code.  If the place to
-    copy the object code is a network server, the Corresponding Source
-    may be on a different server (operated by you or a third party)
-    that supports equivalent copying facilities, provided you maintain
-    clear directions next to the object code saying where to find the
-    Corresponding Source.  Regardless of what server hosts the
-    Corresponding Source, you remain obligated to ensure that it is
-    available for as long as needed to satisfy these requirements.
-
-    e) Convey the object code using peer-to-peer transmission, provided
-    you inform other peers where the object code and Corresponding
-    Source of the work are being offered to the general public at no
-    charge under subsection 6d.
-
-  A separable portion of the object code, whose source code is excluded
-from the Corresponding Source as a System Library, need not be
-included in conveying the object code work.
-
-  A "User Product" is either (1) a "consumer product", which means any
-tangible personal property which is normally used for personal, family,
-or household purposes, or (2) anything designed or sold for incorporation
-into a dwelling.  In determining whether a product is a consumer product,
-doubtful cases shall be resolved in favor of coverage.  For a particular
-product received by a particular user, "normally used" refers to a
-typical or common use of that class of product, regardless of the status
-of the particular user or of the way in which the particular user
-actually uses, or expects or is expected to use, the product.  A product
-is a consumer product regardless of whether the product has substantial
-commercial, industrial or non-consumer uses, unless such uses represent
-the only significant mode of use of the product.
-
-  "Installation Information" for a User Product means any methods,
-procedures, authorization keys, or other information required to install
-and execute modified versions of a covered work in that User Product from
-a modified version of its Corresponding Source.  The information must
-suffice to ensure that the continued functioning of the modified object
-code is in no case prevented or interfered with solely because
-modification has been made.
-
-  If you convey an object code work under this section in, or with, or
-specifically for use in, a User Product, and the conveying occurs as
-part of a transaction in which the right of possession and use of the
-User Product is transferred to the recipient in perpetuity or for a
-fixed term (regardless of how the transaction is characterized), the
-Corresponding Source conveyed under this section must be accompanied
-by the Installation Information.  But this requirement does not apply
-if neither you nor any third party retains the ability to install
-modified object code on the User Product (for example, the work has
-been installed in ROM).
-
-  The requirement to provide Installation Information does not include a
-requirement to continue to provide support service, warranty, or updates
-for a work that has been modified or installed by the recipient, or for
-the User Product in which it has been modified or installed.  Access to a
-network may be denied when the modification itself materially and
-adversely affects the operation of the network or violates the rules and
-protocols for communication across the network.
-
-  Corresponding Source conveyed, and Installation Information provided,
-in accord with this section must be in a format that is publicly
-documented (and with an implementation available to the public in
-source code form), and must require no special password or key for
-unpacking, reading or copying.
-
-  7. Additional Terms.
-
-  "Additional permissions" are terms that supplement the terms of this
-License by making exceptions from one or more of its conditions.
-Additional permissions that are applicable to the entire Program shall
-be treated as though they were included in this License, to the extent
-that they are valid under applicable law.  If additional permissions
-apply only to part of the Program, that part may be used separately
-under those permissions, but the entire Program remains governed by
-this License without regard to the additional permissions.
-
-  When you convey a copy of a covered work, you may at your option
-remove any additional permissions from that copy, or from any part of
-it.  (Additional permissions may be written to require their own
-removal in certain cases when you modify the work.)  You may place
-additional permissions on material, added by you to a covered work,
-for which you have or can give appropriate copyright permission.
-
-  Notwithstanding any other provision of this License, for material you
-add to a covered work, you may (if authorized by the copyright holders of
-that material) supplement the terms of this License with terms:
-
-    a) Disclaiming warranty or limiting liability differently from the
-    terms of sections 15 and 16 of this License; or
-
-    b) Requiring preservation of specified reasonable legal notices or
-    author attributions in that material or in the Appropriate Legal
-    Notices displayed by works containing it; or
-
-    c) Prohibiting misrepresentation of the origin of that material, or
-    requiring that modified versions of such material be marked in
-    reasonable ways as different from the original version; or
-
-    d) Limiting the use for publicity purposes of names of licensors or
-    authors of the material; or
-
-    e) Declining to grant rights under trademark law for use of some
-    trade names, trademarks, or service marks; or
-
-    f) Requiring indemnification of licensors and authors of that
-    material by anyone who conveys the material (or modified versions of
-    it) with contractual assumptions of liability to the recipient, for
-    any liability that these contractual assumptions directly impose on
-    those licensors and authors.
-
-  All other non-permissive additional terms are considered "further
-restrictions" within the meaning of section 10.  If the Program as you
-received it, or any part of it, contains a notice stating that it is
-governed by this License along with a term that is a further
-restriction, you may remove that term.  If a license document contains
-a further restriction but permits relicensing or conveying under this
-License, you may add to a covered work material governed by the terms
-of that license document, provided that the further restriction does
-not survive such relicensing or conveying.
-
-  If you add terms to a covered work in accord with this section, you
-must place, in the relevant source files, a statement of the
-additional terms that apply to those files, or a notice indicating
-where to find the applicable terms.
-
-  Additional terms, permissive or non-permissive, may be stated in the
-form of a separately written license, or stated as exceptions;
-the above requirements apply either way.
-
-  8. Termination.
-
-  You may not propagate or modify a covered work except as expressly
-provided under this License.  Any attempt otherwise to propagate or
-modify it is void, and will automatically terminate your rights under
-this License (including any patent licenses granted under the third
-paragraph of section 11).
-
-  However, if you cease all violation of this License, then your
-license from a particular copyright holder is reinstated (a)
-provisionally, unless and until the copyright holder explicitly and
-finally terminates your license, and (b) permanently, if the copyright
-holder fails to notify you of the violation by some reasonable means
-prior to 60 days after the cessation.
-
-  Moreover, your license from a particular copyright holder is
-reinstated permanently if the copyright holder notifies you of the
-violation by some reasonable means, this is the first time you have
-received notice of violation of this License (for any work) from that
-copyright holder, and you cure the violation prior to 30 days after
-your receipt of the notice.
-
-  Termination of your rights under this section does not terminate the
-licenses of parties who have received copies or rights from you under
-this License.  If your rights have been terminated and not permanently
-reinstated, you do not qualify to receive new licenses for the same
-material under section 10.
-
-  9. Acceptance Not Required for Having Copies.
-
-  You are not required to accept this License in order to receive or
-run a copy of the Program.  Ancillary propagation of a covered work
-occurring solely as a consequence of using peer-to-peer transmission
-to receive a copy likewise does not require acceptance.  However,
-nothing other than this License grants you permission to propagate or
-modify any covered work.  These actions infringe copyright if you do
-not accept this License.  Therefore, by modifying or propagating a
-covered work, you indicate your acceptance of this License to do so.
-
-  10. Automatic Licensing of Downstream Recipients.
-
-  Each time you convey a covered work, the recipient automatically
-receives a license from the original licensors, to run, modify and
-propagate that work, subject to this License.  You are not responsible
-for enforcing compliance by third parties with this License.
-
-  An "entity transaction" is a transaction transferring control of an
-organization, or substantially all assets of one, or subdividing an
-organization, or merging organizations.  If propagation of a covered
-work results from an entity transaction, each party to that
-transaction who receives a copy of the work also receives whatever
-licenses to the work the party's predecessor in interest had or could
-give under the previous paragraph, plus a right to possession of the
-Corresponding Source of the work from the predecessor in interest, if
-the predecessor has it or can get it with reasonable efforts.
-
-  You may not impose any further restrictions on the exercise of the
-rights granted or affirmed under this License.  For example, you may
-not impose a license fee, royalty, or other charge for exercise of
-rights granted under this License, and you may not initiate litigation
-(including a cross-claim or counterclaim in a lawsuit) alleging that
-any patent claim is infringed by making, using, selling, offering for
-sale, or importing the Program or any portion of it.
-
-  11. Patents.
-
-  A "contributor" is a copyright holder who authorizes use under this
-License of the Program or a work on which the Program is based.  The
-work thus licensed is called the contributor's "contributor version".
-
-  A contributor's "essential patent claims" are all patent claims
-owned or controlled by the contributor, whether already acquired or
-hereafter acquired, that would be infringed by some manner, permitted
-by this License, of making, using, or selling its contributor version,
-but do not include claims that would be infringed only as a
-consequence of further modification of the contributor version.  For
-purposes of this definition, "control" includes the right to grant
-patent sublicenses in a manner consistent with the requirements of
-this License.
-
-  Each contributor grants you a non-exclusive, worldwide, royalty-free
-patent license under the contributor's essential patent claims, to
-make, use, sell, offer for sale, import and otherwise run, modify and
-propagate the contents of its contributor version.
-
-  In the following three paragraphs, a "patent license" is any express
-agreement or commitment, however denominated, not to enforce a patent
-(such as an express permission to practice a patent or covenant not to
-sue for patent infringement).  To "grant" such a patent license to a
-party means to make such an agreement or commitment not to enforce a
-patent against the party.
-
-  If you convey a covered work, knowingly relying on a patent license,
-and the Corresponding Source of the work is not available for anyone
-to copy, free of charge and under the terms of this License, through a
-publicly available network server or other readily accessible means,
-then you must either (1) cause the Corresponding Source to be so
-available, or (2) arrange to deprive yourself of the benefit of the
-patent license for this particular work, or (3) arrange, in a manner
-consistent with the requirements of this License, to extend the patent
-license to downstream recipients.  "Knowingly relying" means you have
-actual knowledge that, but for the patent license, your conveying the
-covered work in a country, or your recipient's use of the covered work
-in a country, would infringe one or more identifiable patents in that
-country that you have reason to believe are valid.
-
-  If, pursuant to or in connection with a single transaction or
-arrangement, you convey, or propagate by procuring conveyance of, a
-covered work, and grant a patent license to some of the parties
-receiving the covered work authorizing them to use, propagate, modify
-or convey a specific copy of the covered work, then the patent license
-you grant is automatically extended to all recipients of the covered
-work and works based on it.
-
-  A patent license is "discriminatory" if it does not include within
-the scope of its coverage, prohibits the exercise of, or is
-conditioned on the non-exercise of one or more of the rights that are
-specifically granted under this License.  You may not convey a covered
-work if you are a party to an arrangement with a third party that is
-in the business of distributing software, under which you make payment
-to the third party based on the extent of your activity of conveying
-the work, and under which the third party grants, to any of the
-parties who would receive the covered work from you, a discriminatory
-patent license (a) in connection with copies of the covered work
-conveyed by you (or copies made from those copies), or (b) primarily
-for and in connection with specific products or compilations that
-contain the covered work, unless you entered into that arrangement,
-or that patent license was granted, prior to 28 March 2007.
-
-  Nothing in this License shall be construed as excluding or limiting
-any implied license or other defenses to infringement that may
-otherwise be available to you under applicable patent law.
-
-  12. No Surrender of Others' Freedom.
-
-  If conditions are imposed on you (whether by court order, agreement or
-otherwise) that contradict the conditions of this License, they do not
-excuse you from the conditions of this License.  If you cannot convey a
-covered work so as to satisfy simultaneously your obligations under this
-License and any other pertinent obligations, then as a consequence you may
-not convey it at all.  For example, if you agree to terms that obligate you
-to collect a royalty for further conveying from those to whom you convey
-the Program, the only way you could satisfy both those terms and this
-License would be to refrain entirely from conveying the Program.
-
-  13. Use with the GNU Affero General Public License.
-
-  Notwithstanding any other provision of this License, you have
-permission to link or combine any covered work with a work licensed
-under version 3 of the GNU Affero General Public License into a single
-combined work, and to convey the resulting work.  The terms of this
-License will continue to apply to the part which is the covered work,
-but the special requirements of the GNU Affero General Public License,
-section 13, concerning interaction through a network will apply to the
-combination as such.
-
-  14. Revised Versions of this License.
-
-  The Free Software Foundation may publish revised and/or new versions of
-the GNU General Public License from time to time.  Such new versions will
-be similar in spirit to the present version, but may differ in detail to
-address new problems or concerns.
-
-  Each version is given a distinguishing version number.  If the
-Program specifies that a certain numbered version of the GNU General
-Public License "or any later version" applies to it, you have the
-option of following the terms and conditions either of that numbered
-version or of any later version published by the Free Software
-Foundation.  If the Program does not specify a version number of the
-GNU General Public License, you may choose any version ever published
-by the Free Software Foundation.
-
-  If the Program specifies that a proxy can decide which future
-versions of the GNU General Public License can be used, that proxy's
-public statement of acceptance of a version permanently authorizes you
-to choose that version for the Program.
-
-  Later license versions may give you additional or different
-permissions.  However, no additional obligations are imposed on any
-author or copyright holder as a result of your choosing to follow a
-later version.
-
-  15. Disclaimer of Warranty.
-
-  THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
-APPLICABLE LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
-HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
-OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
-THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
-IS WITH YOU.  SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
-ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
-
-  16. Limitation of Liability.
-
-  IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
-WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
-THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
-GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
-USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
-DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
-PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
-EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
-SUCH DAMAGES.
-
-  17. Interpretation of Sections 15 and 16.
-
-  If the disclaimer of warranty and limitation of liability provided
-above cannot be given local legal effect according to their terms,
-reviewing courts shall apply local law that most closely approximates
-an absolute waiver of all civil liability in connection with the
-Program, unless a warranty or assumption of liability accompanies a
-copy of the Program in return for a fee.
-
-                     END OF TERMS AND CONDITIONS
-
-            How to Apply These Terms to Your New Programs
-
-  If you develop a new program, and you want it to be of the greatest
-possible use to the public, the best way to achieve this is to make it
-free software which everyone can redistribute and change under these terms.
-
-  To do so, attach the following notices to the program.  It is safest
-to attach them to the start of each source file to most effectively
-state the exclusion of warranty; and each file should have at least
-the "copyright" line and a pointer to where the full notice is found.
-
-    <one line to give the program's name and a brief idea of what it does.>
-    Copyright (C) <year>  <name of author>
-
-    This program is free software: you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation, either version 3 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program.  If not, see <http://www.gnu.org/licenses/>.
-
-Also add information on how to contact you by electronic and paper mail.
-
-  If the program does terminal interaction, make it output a short
-notice like this when it starts in an interactive mode:
-
-    <program>  Copyright (C) <year>  <name of author>
-    This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
-    This is free software, and you are welcome to redistribute it
-    under certain conditions; type `show c' for details.
-
-The hypothetical commands `show w' and `show c' should show the appropriate
-parts of the General Public License.  Of course, your program's commands
-might be different; for a GUI interface, you would use an "about box".
-
-  You should also get your employer (if you work as a programmer) or school,
-if any, to sign a "copyright disclaimer" for the program, if necessary.
-For more information on this, and how to apply and follow the GNU GPL, see
-<http://www.gnu.org/licenses/>.
-
-  The GNU General Public License does not permit incorporating your program
-into proprietary programs.  If your program is a subroutine library, you
-may consider it more useful to permit linking proprietary applications with
-the library.  If this is what you want to do, use the GNU Lesser General
-Public License instead of this License.  But first, please read
-<http://www.gnu.org/philosophy/why-not-lgpl.html>.
diff --git a/Ready/README.txt b/Ready/README.txt
deleted file mode 100644
index dd9508b2e..000000000
--- a/Ready/README.txt
+++ /dev/null
@@ -1,35 +0,0 @@
----------------------------------------------------------------------------------
-                                    About
----------------------------------------------------------------------------------
-
-For help, visit: http://code.google.com/p/reaction-diffusion/
-Or email: reaction-diffusion@googlegroups.com
-
-Ready is free software. Help us improve it!
-
-For build instructions, see BUILD.txt.
-
-See Help/credits.html for a list of credits and acknowledgements.
-
----------------------------------------------------------------------------------
-                                   License
----------------------------------------------------------------------------------
-
-Copyright 2011, 2012, 2013 The Ready Bunch
-
-The Ready Bunch is: Tim Hutton, Robert Munafo, Andrew Trevorrow, Tom Rokicki, Dan Wills
-
-This file is part of Ready.
-
-Ready is free software: you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation, either version 3 of the License, or
-(at your option) any later version.
-
-Ready is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-GNU General Public License for more details.
-
-See COPYING.txt for the full GNU General Public License.
-Or visit <http://www.gnu.org/licenses/>.
diff --git a/Ready_old_CLI/CMakeLists.txt b/Ready_old_CLI/CMakeLists.txt
deleted file mode 100644
index 857219474..000000000
--- a/Ready_old_CLI/CMakeLists.txt
+++ /dev/null
@@ -1,39 +0,0 @@
-cmake_minimum_required(VERSION 2.6)
-
-project(ReaDy)
-
-if (NOT CMAKE_BUILD_TYPE)
- message(STATUS "No build type selected, default to Release")
- set(CMAKE_BUILD_TYPE "Release")
-endif()
-
-if(MSVC)
-  add_definitions(/arch:SSE2)
-else()
-  add_definitions(-msse2)
-endif()
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "calc_Brusselator" )
-INCLUDE_DIRECTORIES( "calc_Gray_Scott" )
-INCLUDE_DIRECTORIES( "ready_display" )
-
-add_executable(ReaDy 
-  main-ready.cpp
-  util.h
-  util.cpp
-  calc_Brusselator/brusselator.cpp
-  calc_Brusselator/brusselator.h
-  calc_Gray_Scott/gray_scott_scalar.cpp
-  calc_Gray_Scott/gray_scott_scalar.h
-  calc_Gray_Scott/gray_scott_hwivector.cpp
-  calc_Gray_Scott/gray_scott_hwivector.h
-  calc_Gray_Scott/hwi_vector.h
-  calc_Gray_Scott/dicek.h
-  ready_display/ready_display.cpp
-  ready_display/ready_display.h
-)
-
-TARGET_LINK_LIBRARIES(ReaDy ${OpenCV_LIBS} )
diff --git a/Ready_old_CLI/calc_Brusselator/brusselator.cpp b/Ready_old_CLI/calc_Brusselator/brusselator.cpp
deleted file mode 100644
index 2afa0ce2c..000000000
--- a/Ready_old_CLI/calc_Brusselator/brusselator.cpp
+++ /dev/null
@@ -1,94 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-#include <time.h>
-#include <stdlib.h>
-
-#include "brusselator.h"
-#include "../util.h"
-
-static int g_width;
-static int g_height;
-static bool g_wrap;
-static bool g_paramspace;
-
-
-void bruss_init(float *a, float *b, int width, int height, float A, float B)
-{
-  srand((unsigned int)time(NULL));
-    
-  // figure the values
-  for(int i = 0; i < width*height; i++) 
-  {
-    a[i] = A + ut_frand(-0.01f,0.01f);
-    b[i] = B/A + ut_frand(-0.01f,0.01f);
-  }
-}
-
-void bruss_compute_setup(int width, int height, bool wrap, bool paramspace)
-{
-  g_width = width;
-  g_height = height;
-  g_wrap = wrap;
-  g_paramspace = paramspace;
-}
-
-void compute_bruss(float *a, float *b, float *da, float *db,
-             float A, float B, float D1, float D2, float speed)
-{
-    // compute change in each cell
-    for(int i = 0; i < g_height; i++) 
-    {
-        int iprev, inext;
-        if (g_wrap) {
-          iprev = (i + g_height - 1) % g_height;
-          inext = (i + 1) % g_height;
-        } else {
-          iprev = max(0,i-1);
-          inext = min(g_height-1,i+1);
-        }
-
-        for(int j = 0; j < g_width; j++) 
-        {
-            int jprev, jnext;
-            if (g_wrap) {
-              jprev = (j + g_width - 1) % g_width;
-              jnext = (j + 1) % g_width;
-            } else {
-              jprev = max(0,j-1);
-              jnext = min(g_width-1,j+1);
-            }
-
-            float aval = a[i*g_width+j];
-            float bval = b[i*g_width+j];
-
-            if(g_paramspace)
-            {
-                const float A1=0.0, A2=4.0, B1=0.0, B2=15.0;
-                A = A1+(A2-A1)*((float)i)/((float)g_height);
-                B = B1+(B2-B1)*((float)j)/((float)g_width);
-            }
-
-            // compute the Laplacians of a and b
-            float dda = a[i*g_width+jprev] + a[i*g_width+jnext] + a[iprev*g_width+j] + a[inext*g_width+j] - 4*aval;
-            float ddb = b[i*g_width+jprev] + b[i*g_width+jnext] + b[iprev*g_width+j] + b[inext*g_width+j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i*g_width+j] = A-(B+1)*aval + aval*aval*bval + D1*dda;
-            db[i*g_width+j] = B*aval - aval*aval*bval + D2*ddb;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < g_width*g_height; i++) {
-        a[i] += (speed * da[i]);
-        b[i] += (speed * db[i]);
-    }
-}
-
diff --git a/Ready_old_CLI/calc_Brusselator/brusselator.h b/Ready_old_CLI/calc_Brusselator/brusselator.h
deleted file mode 100644
index ce73bd8ce..000000000
--- a/Ready_old_CLI/calc_Brusselator/brusselator.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-void bruss_init(float *a, float *b, int width, int height, float A, float B);
-void bruss_compute_setup(int width, int height, bool wrap, bool paramspace);
-void compute_bruss(float *a, float *b, float *da, float *db,
-             float A, float B, float D1, float D2, float speed);
diff --git a/Ready_old_CLI/calc_Gray_Scott/dicek.h b/Ready_old_CLI/calc_Gray_Scott/dicek.h
deleted file mode 100644
index c68f87dfc..000000000
--- a/Ready_old_CLI/calc_Gray_Scott/dicek.h
+++ /dev/null
@@ -1,506 +0,0 @@
-/*
-  dicek.h                                                                                                ^u 120 ^x f
-
-This provides macros for multi-threaded programming. The intention is to allow a single program source to be compiled
-for Windows, Linux or Mac OS, and accomplish multi-threading without having to have a lot of #ifdefs around the
-OS-specific code.
-
-REVISION HISTORY
- 20110929 First version of simple DICEK_SPLIT_MERGE function. This is a blocking, one-time N-way parallel subroutine
-call (no provision for threads to continue through multiple synchronization barriers). Works in DICEK_USE_THREADS and
-DICEK_EMULATE modes (tested with math3000.cxx A094358() routine).
- 20110930 Add a first (extremely speculative and untested) shot at the Windows implementation, currently protected by a
-fall-back #ifdef block at the beginning that checks for Windows and reverts to EMULATE mode, pending testing by a real
-Windows programmer.
- 20111001 Add the thread interlock macros and get them working in EMULATE and POSIX modes.
- 20111009 Initialize mutexes in Win32 version of DICEK_SPLIT_1
- 20111011 Use semaphores instead of mutexes in Win32 version.
-
-*/
-
-/* First we test for each of the known operating system environments */
-
-#if (defined(__linux__) || defined(__APPLE__))
-# ifndef DICEK_USE_POSIX
-#   ifndef DICEK_EMULATE
-#     define DICEK_USE_POSIX
-#   endif
-# endif
-#endif
-
-
-#ifdef _WIN32
-# ifndef DICEK_EMULATE
-#   define DICEK_USE_PROCESS_H
-# endif
-#endif
-
-
-#ifndef DICEK_USE_PROCESS_H
-# ifndef DICEK_USE_POSIX
-#   ifndef DICEK_EMULATE
-#     define DICEK_EMULATE
-#   endif
-# endif
-#endif
-
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//  #include<stdio.h>                                                                                                //
-//  /*                                                   xx*/                    /*                             */   //
-//  /*                                                     */                    /*                             */   //
-//  int                 main()  {char      a[]     ={     'T',    'h','i',     's',32,     'p','r'        ,'o','g'   //
-//  ,'r','a','m',      32,   'j',   'u',   's',    't'    ,32           ,'d',    'o',   'e'     ,'s'    ,32    ,'i'  //
-//  ,'t'              ,32    ,'t'    ,'h'  ,'e'   ,32,    'h'    ,97,'r','d'   ,32,    'w',97,121,33,   32,    40,   //
-//  68,               'o',   'n',    39,   't'    ,32     ,'y'   ,'o',   117,    32    ,'t'             ,'h'   ,'i'  //
-//  ,'n'               /*     Xy     =a     +3     +n      ++     ;a=     b-     (*     x/z              );     if   //
-//  (Xy-++n<(z+*x))z  =b;a   +b,     z+=     x*/,107 ,    63,63    ,63,41,'\n'   ,00};    puts(a);}      /*.RPM.*/   //
-//                                                                                                                   //
-//        Emulated versions of the DICEK macros. (These also serve as documentation for what each macro does)        //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-#ifdef DICEK_EMULATE
-
-/*
- We cannot do inter-thread communication because we execute each "spawned thread" to completion before starting the next
-one. Therefore, the first time any child thread tries to DICEK_INTERLOCK, the program will deadlock because there is no
-way for the parent to get to the DICEK_CH_BEGIN until the child has exited.
- */
-#define DICEK_SUPPORTS_BLOCKING 0
-
-/*
-DICEK_THREAD_VARS defines variables which need to be included in the parameter block of any subroutine called by
-DICEK_SPLIT_MERGE
-  Since this is the emulated version, there are no master_wkg and child_wkg semaphores.
- */
-#define DICEK_THREAD_VARS \
-  long DICEK_tnum; \
-  void * DICEK_thread; \
-  void * DICEK_return;
-
-/*
- DICEK_INIT_NTHR declares an integer-type variable named "nth" and initializes it to the number of hardware threads
-supported by this system. Place it in your main() or a function from which threads will be launched.
-  If you want your threads to have access to the value of nth, you should do "glob = nth;" right after the
-DICEK_INIT_NTHR(), where "glob" is a suitably scoped global variable.
-
-In emulated mode we return 3 as the number of "hardware threads", partly because 3 is a fairly rare value in real
-computers (the Athlon X3 being about the only one anyone will have) and therefore if you find that DICEK is running 3
-"threads" then you know it probably failed to auto-detect the environment.
-  Also, as of this writing (2011) 3 threads is about the average number of hardware threads across all portable and
-desktop computers that are out there. (Anything by Intel with the "Core i3" or higher brand has 4 threads, and most
-desktop machines have at least 2 cores and 4 threads as well)
-   However, since this is the emulated-mode version of the macros, the "threads" are actually going to just run one
-after the next.
- */
-#define DICEK_INIT_NTHR(nth) int nth = 3;
-
-/*
- Place DICEK_DATA in the variable declarations area of the function containing an DICEK_FORK directive. dtype should be
-a struct which contains the DICEK_THREAD_VARS macro, and as many other fields as you want. nth is the number of threads
-that you will be creating with DICEK_SPLIT_MERGE.
-  It will declare an array aname (dynamically allocated) which will be an array of structs of type dtype, and fill the
-DICEK-specific fields with a thread number and a pointer to the pthreads data for each thread that will be used by
-DICEK_SPLIT_MERGE.
- */
-#define DICEK_DATA(dtype, arrayname, nth) \
-    dtype * arrayname; \
-    arrayname = (dtype *)malloc(nth * sizeof(dtype)); \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      arrayname[_DICEK_i].DICEK_tnum = _DICEK_i; \
-      arrayname[_DICEK_i].DICEK_thread = 0; \
-    }
-
-/*
- DICEK_SPLIT_1 starts a thread with the given index; the array "arrayname" should be allocated by DICEK_DATA and should
-contain an element at the given index. The thread will call a function called funcname, which should be type
-
-  void(* funcname)(void *)
-
-The instance of funcname will be passed the index'th element of arrayname.
-  After the thread has exited, execution proceeds with the instruction following DICEK_SPLIT_1.
- */
-#define DICEK_SPLIT_1(funcname, arrayname, index) \
-    arrayname[index].DICEK_return = 0; \
-    arrayname[index].DICEK_thread = (void *) index; \
-    (funcname)((void *) &(arrayname[index]));
-
-#define DICEK_MERGE_1(arrayname, index)  /* nop */
-
-/*
- DICEK_SPLIT_MERGE starts nth threads vis DICEK_SPLIT_1. After the last of the threads has exited, execution proceeds
-with the instruction following DICEK_SPLIT_MERGE.
- */
-#define DICEK_SPLIT_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_1((funcname), (arrayname), (_DICEK_i)) \
-    }
-
-#define DICEK_SPLIT_M(funcname, arrayname, index) DICEK_SPLIT_1(funcname, arrayname, index)
-
-/* Just start the threads, without a merge. This is used for applications that do inter-thread synchronization */
-#define DICEK_SPLIT(funcname, arrayname, nth) DICEK_SPLIT_MERGE(funcname, arrayname, nth)
-
-/*
- Place a DICEK_SUB at the beginning of any routine called by a DICEK_SPLIT_MERGE. dtype is the type of the struct which
-is the array elements of arrayname, and argname is the name of the subroutine's argument. For example:
-
-  typedef struct f_vars {
-    DICEK_THREAD_VARS
-    int foo;
-    char * bar;
-  } f_vars;
-
-  (void *)myfunc(void * params)
-  {
-    DICEK_SUB(f_vars, params);
-    // other local variable declarations
-    ...
-  }
-
-For an example of a thread that performs inter-thread synchronization, see the example below (DICEK_CH_BEGIN macro)
-
-In emulated mode, we need a pointer to the struct that was passed in. This pointer is used by DICEK_RETURN
- */
-#define DICEK_SUB(dtype, argname) \
-  dtype * _DICEK_params = (dtype *) argname;
-
-/*
- The child thread should start with DICEK_CH_BEGIN, which grabs the semaphore indicating that the child is working.
-Then it should do its work, followed DICEK_CH_SYNC to tell the other threads in its group that it has finished
-its section of work.
- The master thread will eventually release the semaphore used by DICEK_CH_BEGIN allowing the child to continue
-with the next piece of work.
-
-  (void *)myfunc(void * params)
-  {
-    DICEK_SUB(f_vars, params);
-    // other local variable declarations
-    while(working) {
-      DICEK_CH_BEGIN
-      // Do computation here
-      DICEK_CH_SYNC
-    }
-    // We're done performing useful work
-    DICEK_CH_END
-    // Here we deallocate any temporary private storage
-    DICEK_RETURN(result)
-  }
-
-*/
-#define DICEK_CH_BEGIN \
-  fprintf(stderr, "Runtime error: DICEK does not support inter-thread communication in EMULATE mode.\n"); \
-  exit(-1);
-
-/*
- DICEK_CH_SYNC releases the "master working" semaphore, which tells the master that this thread is done with its
-work and the master can proceed to use the results that were produced by the child.
- */
-#define DICEK_CH_SYNC /* nop */
-
-/*
-  DICEK_INTERLOCK is called by the master thread to wait for the children to be done with their bits of work.
-It attempts to take all of the master_working semaphores, which are initially locked and don't get unlocked
-until the child threads reach their DICEK_CH_SYNC.
- */
-#define DICEK_INTERLOCK(arrayname, nth) DICEK_CH_BEGIN
-
-/*
-  DICEK_RESUME is called by the master thread to tell the child threads that they cal now continue with another
-piece of work. It unlocks the child_working semaphore so that the children can execute their DICEK_CH_BEGIN
-statement a second time.
- */
-#define DICEK_RESUME(arrayname, nth) /* nop */
-
-/*
- Put DICEK_CH_END at the point where a child thread has finished performing any actions that require
-cooperation or synchronization, and before the process ends. This macro releases the child_working semaphore
-so that the semaphore can be deallocated by the master process.
- The typical sequence is: DICEK_CH_END; deallocate private storage; DICEK_RETURN; exit. See the DICEK_CH_BEGIN
-block comment for a pseudo-code example.
- */
-#define DICEK_CH_END /* nop */
-
-/*
- Place a DICEK_RETURN at the very end of any routine called by a DICEK_SPLIT_MERGE. It passes returnv back
-to the calling (master) thread via a field routine's parameter block.
- WARNING: On some operating systems, this return value is passed through the kernel, while on others it is actually
-written directly into the parameter block. On some systems, this routine causes the thread to exit immediately, while on
-others it is possible to execute statements after the DICEK_RETURN. Users of this package should not rely on any of
-these specifics to be true in runtime.
- */
-#define DICEK_RETURN(returnv) _DICEK_params->DICEK_return = (void*)(returnv);
-
-#define DICEK_MERGE_M(arrayname, index)  /* nop */
-
-#define DICEK_MERGE(funcname, arrayname, nth) /* nop */
-
-#endif
-
-/* - - - - - - - - - - - - - - - - - - - - - End of the EMULATED section  - - - - - - - - - - - - - - - - - - - - - - */
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                ,        ,         //
-//                                                                                               /(        )`        //
-//                                                                                               \ \___   / |        //
-//         _nnnn_                                                                                /- _  `-/  '        //
-//        dGGGGMMb                                                                              (/\/ \ \   /\        //
-//       @p~qp~~qMb                                                                             / /   | `    \       //
-//       M|@||@) M|       ooooooooo.     .oooooo.    .oooooo..o ooooo ooooooo  ooooo            O O   ) /    |       //
-//       @,----.JM|       `888   `Y88.  d8P'  `Y8b  d8P'    `Y8 `888'  `8888    d8'             `-^--'`<     '       //
-//      JS^\__/  qKL       888   .d88' 888      888 Y88bo.       888     Y888..8P              (_.)  _  )   /        //
-//     dZP        qKRb     888ooo88P'  888      888  `"Y8888o.   888      `8888'                `.___/`    /         //
-//    dZP          qKKb    888         888      888      `"Y88b  888     .8PY888.                 `-----' /          //
-//   fZP            SMMb   888         `88b    d88' oo     .d8P  888    d8'  `888b   <----.     __ / __   \          //
-//   HZM            MMMM  o888o         `Y8bood8P'  8""88888P'  o888o o888o  o88888o <----|====O)))==) \) /====      //
-//   FqM            MMMM                                                             <----'    `--' `.__,' \         //
-// __| ".        |\dS"qML                                                                         |        |         //
-// |    `.       | `' \Zq                                                                          \       /       /\//
-//_)      \.___.,|     .'                                                                     ______( (_  / \______/ //
-//\____   )MMMMMP|   .'           Versions of the DICEK macros for Linux, Mac OS,           ,'  ,-----'   |          //
-//     `-'       `--' hjm             and other POSIX-compliant environments                `--{__________)          //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-#ifdef DICEK_USE_POSIX
-
-#define DICEK_SUPPORTS_BLOCKING 1
-
-#include <pthread.h>
-
-// Mach-based system (most notably Apple) use sysctlbyname to query the number of CPUs/cores/threads
-# ifdef __MACH__
-#   include <sys/types.h>
-#   include <sys/sysctl.h>
-# endif
-
-// TODO: Some Linux systems might need an #include to access the mechanism for finding out the number of threads.
-
-
-#define DICEK_THREAD_VARS \
-  long DICEK_tnum; \
-  void * DICEK_thread; \
-  pthread_mutex_t DICEK_master_wkg; \
-  pthread_mutex_t DICEK_child_wkg; \
-  void * DICEK_return;
-
-
-#ifdef __APPLE__
-# define DICEK_INIT_NTHR(nth) int nth; { \
-    size_t _DICEK_sz_in = sizeof(nth); \
-    long _DICEK_rv2 = sysctlbyname("hw.ncpu", (void *) (&nth), &_DICEK_sz_in, 0, 0); \
-    if (nth <= 0) { nth = 1; } \
-    if (nth > 64) { nth = 64; } }
-#else
-// TODO: How to query number of threads in Linux (I suspect reading /proc/cpuinfo will work on most,
-// but not all, Linuces)
-# define DICEK_INIT_NTHR(nth) int nth = 3;
-#endif
-
-#define DICEK_DATA(dtype, arrayname, nth) \
-    dtype * arrayname; \
-    arrayname = (dtype *)malloc(nth * sizeof(dtype)); \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      arrayname[_DICEK_i].DICEK_tnum = _DICEK_i; \
-      arrayname[_DICEK_i].DICEK_thread = 0; \
-      arrayname[_DICEK_i].DICEK_return = 0; \
-    }
-
-/* Just start a single thread. */
-#define DICEK_SPLIT_1(funcname, arrayname, index) \
-    arrayname[index].DICEK_return = 0; \
-    pthread_create((pthread_t *) (&(arrayname[index].DICEK_thread)), \
-        0, (void* (*)(void*))funcname, (void *) &(arrayname[index]));
-
-/* End a single thread. */
-#define DICEK_MERGE_1(arrayname, index) \
-    pthread_join((pthread_t) (arrayname[index].DICEK_thread), \
-        &(arrayname[index].DICEK_return));
-
-/* In the POSIX implementation, semaphore communication is supported, so the SPLIT_MERGE function
-(which by definition has no communication between master and spawned threads) does not need to
-initialize the semaphores, whereas DICEK_SPLIT_1 does. */
-#define DICEK_SPLIT_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_1(funcname, arrayname, _DICEK_i) \
-    } \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_MERGE_1(arrayname, _DICEK_i) \
-    }
-
-/* DICEK_SPLIT_M also sets up the semaphores in case the client is going to want to do that. */
-#define DICEK_SPLIT_M(funcname, arrayname, index) \
-    pthread_mutex_init(&(arrayname[index].DICEK_child_wkg), 0); \
-    pthread_mutex_init(&(arrayname[index].DICEK_master_wkg), 0); \
-    pthread_mutex_trylock(&(arrayname[index].DICEK_master_wkg)); \
-    DICEK_SPLIT_1(funcname, arrayname, index)
-
-/* Just start the threads, without a merge. This is used for applications that do inter-thread synchronization */
-#define DICEK_SPLIT(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_M((void* (*)(void*))funcname, (arrayname), _DICEK_i) \
-    }
-
-#define DICEK_SUB(dtype, argname) dtype * _DICEK_params = (dtype *) argname;
-
-#define DICEK_CH_BEGIN \
-  pthread_mutex_lock(&(_DICEK_params->DICEK_child_wkg));
-
-#define DICEK_CH_SYNC \
-  pthread_mutex_unlock(&(_DICEK_params->DICEK_master_wkg));
-
-#define DICEK_INTERLOCK(arrayname, nth) \
-  for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-    pthread_mutex_lock(&(arrayname[_DICEK_i].DICEK_master_wkg)); \
-  }
-
-#define DICEK_RESUME(arrayname, nth) \
-  for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-    pthread_mutex_unlock(&(arrayname[_DICEK_i].DICEK_child_wkg)); \
-  }
-
-#define DICEK_CH_END \
-  pthread_mutex_unlock(&(_DICEK_params->DICEK_child_wkg));
-
-#define DICEK_RETURN(returnv) pthread_exit((void*)(returnv));
-
-#define DICEK_MERGE_M(arrayname, index) \
-    DICEK_MERGE_1(arrayname, index) \
-    pthread_mutex_destroy(&(arrayname[index].DICEK_child_wkg)); \
-    pthread_mutex_unlock(&(arrayname[index].DICEK_master_wkg)); \
-    pthread_mutex_destroy(&(arrayname[index].DICEK_master_wkg));
-
-#define DICEK_MERGE(funcname, arrayname, nth) \
-  for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-    DICEK_MERGE_M((arrayname), _DICEK_i) \
-  }
-
-#endif
-
-/* - - - - - - - - - - - - - - - - - - - - - - End of the POSIX section - - - - - - - - - - - - - - - - - - - - - - - */
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//       _-^^^^-_                                  ,,                    ,,                                          //
-//      /      './^-____-^/  `7MMF'     A     `7MF'db                  `7MM                                          //
-//     /       ./.'      /     `MA     ,MA     ,V                        MM                                          //
-//     /      './.'      /      VM:   ,VVM:   ,V `7MM  `7MMpMMMb.   ,M""bMM  ,pW"Wq.`7M'    ,A    `MF',pP"Ybd        //
-//    /_-^^^^-_/.'      /        MM.  M' MM.  M'   MM    MM    MM ,AP    MM 6W'   `Wb VA   ,VAA   ,V  8I   `"        //
-//    /      './^-____-^/        `MM A'  `MM A'    MM    MM    MM 8MI    MM 8M     M8  VA ,V  VA ,V   `YMMMa.        //
-//   /       '/.'      /          :MM;    :MM;     MM    MM    MM `Mb    MM YA.   ,A9   VVV    VVV    L.   I8        //
-//   /      .'/.'      /           VF      VF    .JMML..JMML  JMML.`Wbmd"MML.`Ybmd9'     W      W     M9mmmP'        //
-//  /_-^^^^-_/.'      /                                                                                              //
-//            ^-___rpm           Versions of the DICEK macros for Microsoft Windows                                  //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-#ifdef DICEK_USE_PROCESS_H
-
-// This is based on:
-//   http://msdn.microsoft.com/en-US/library/kdzttdcb(v=VS.80).aspx  (_beginthread)
-//   http://msdn.microsoft.com/en-us/library/ms687032(v=VS.85).aspx  (WaitForSingleObject)
-//   http://msdn.microsoft.com/en-us/library/ms685071(v=VS.85).aspx  (ReleaseSemaphore)
-//   http://msdn.microsoft.com/en-us/library/ms682438(v=VS.85).aspx  (CreateSemaphore)
-//
-//   http://msdn.microsoft.com/en-us/library/ms682411(v=VS.85).aspx  (CreateMutex, no longer used)
-//   http://msdn.microsoft.com/en-us/library/ms685066(v=VS.85).aspx  (ReleaseMutex, no longer used)   
-
-#define DICEK_SUPPORTS_BLOCKING 1
-
-#include <process.h>
-#include <windows.h>
-
-#define DICEK_THREAD_VARS \
-  long DICEK_tnum; \
-  HANDLE DICEK_thread; \
-  HANDLE DICEK_master_wkg; \
-  HANDLE DICEK_child_wkg; \
-  void * DICEK_return;
-
-#define DICEK_INIT_NTHR(nth) \
-    int nth; \
-    { SYSTEM_INFO si; GetSystemInfo(&si); nth = si.dwNumberOfProcessors; }
-
-#define DICEK_DATA(dtype, arrayname, nth) \
-    dtype * arrayname; \
-    arrayname = (dtype *)malloc(nth * sizeof(dtype)); \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      arrayname[_DICEK_i].DICEK_tnum = _DICEK_i; \
-      arrayname[_DICEK_i].DICEK_thread = (HANDLE) NULL; \
-      arrayname[_DICEK_i].DICEK_return = 0; \
-    }
-
-#define DICEK_SPLIT_1(funcname, arrayname, index) \
-    arrayname[index].DICEK_return = 0; \
-    arrayname[index].DICEK_thread = (HANDLE) _beginthread(funcname, 0, \
-        (void *) &(arrayname[index]));
-
-#define DICEK_MERGE_1(arrayname, index) \
-    WaitForSingleObject(arrayname[index].DICEK_thread, INFINITE);
-
-/* We use _beginthread, there are other options including _beginthreadex and the more native CreateThread */
-#define DICEK_SPLIT_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_1(funcname, arrayname, _DICEK_i) \
-    } \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_MERGE_1(arrayname, _DICEK_i) \
-    }
-
-// DICEK_SPLIT_M also sets up the semaphores in case the client is going to want to do that.
-// child semaphore is initially free (value 1); master semaphore is initially taken
-#define DICEK_SPLIT_M(funcname, arrayname, index) \
-    arrayname[index].DICEK_child_wkg = CreateSemaphore(NULL, 1, 1, NULL); \
-    arrayname[index].DICEK_master_wkg = CreateSemaphore(NULL, 0, 1, NULL); \
-    DICEK_SPLIT_1(funcname, arrayname, index)
-
-/* Just start the threads, without a merge. This is used for applications that do inter-thread synchronization */
-#define DICEK_SPLIT(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_M((funcname), (arrayname), (_DICEK_i)) \
-    }
-
-/* Save a pointer to the param block for use by DICEK_RETURN */
-#define DICEK_SUB(dtype, argname) dtype * _DICEK_params = (dtype *) argname;
-
-#define DICEK_CH_BEGIN \
-    WaitForSingleObject(_DICEK_params->DICEK_child_wkg,INFINITE);
-
-#define DICEK_CH_SYNC \
-    ReleaseSemaphore(_DICEK_params->DICEK_master_wkg, 1, NULL);
-
-#define DICEK_INTERLOCK(arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-        WaitForSingleObject(arrayname[_DICEK_i].DICEK_master_wkg,INFINITE); \
-    }
-
-#define DICEK_RESUME(arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      ReleaseSemaphore(arrayname[_DICEK_i].DICEK_child_wkg, 1, NULL); \
-    }
-
-#define DICEK_CH_END \
-    ReleaseSemaphore(_DICEK_params->DICEK_child_wkg, 1, NULL);
-
-#define DICEK_RETURN(returnv) \
-    _DICEK_params->DICEK_return = (void*)(returnv); \
-    _endthread();
-
-#define DICEK_MERGE_M(arrayname, index) \
-    DICEK_MERGE_1(arrayname, index) \
-    CloseHandle(arrayname[index].DICEK_thread); \
-    CloseHandle(arrayname[index].DICEK_child_wkg); \
-    ReleaseSemaphore(arrayname[index].DICEK_master_wkg, 1, NULL); \
-    CloseHandle(arrayname[index].DICEK_master_wkg);
-
-#define DICEK_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_MERGE_M((arrayname), _DICEK_i) \
-    }
-
-#endif
-
-/* - - - - - - - - - - - - - - - - - - - - - End of the PROCESS_H section - - - - - - - - - - - - - - - - - - - - - - */
diff --git a/Ready_old_CLI/calc_Gray_Scott/gray_scott_hwivector.cpp b/Ready_old_CLI/calc_Gray_Scott/gray_scott_hwivector.cpp
deleted file mode 100644
index 3bc0fef9e..000000000
--- a/Ready_old_CLI/calc_Gray_Scott/gray_scott_hwivector.cpp
+++ /dev/null
@@ -1,679 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See ../../README.txt for more details.
-
-*/
-
-// hardware
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-# include <xmmintrin.h>
-#endif
-
-// To convince yourself that the macro library works on any hardware,
-// un-comment this "#define HWIV_EMULATE", and you'll get the macros inside
-// the #ifdef HWIV_V4F4_EMULATED block in hwi_vector.h.  The emulated
-// macros do everything with normal floats and arrays, and run about 3-4 times
-// slower.
-//#define HWIV_EMULATE
-#define HWIV_WANT_V4F4
-#include "hwi_vector.h";
-
-// stdlib:
-#include <stdio.h>
-
-#include "gray_scott_hwivector.h"
-#include "../util.h"
-
-static int g_width;
-static int g_height;
-static bool g_wrap;
-static bool g_paramspace;
-
-void gs_hwi_compute_setup(int width, int height, bool wrap, bool paramspace)
-{
-  g_width = width;
-  g_height = height;
-  g_wrap = wrap;
-  g_paramspace = paramspace;
-}
-
-/* Spawn threads and dispatch to compute() routine inside threads */
-void compute_dispatch(float *u, float *v, float *du, float *dv,
-  float D_u, float D_v, float F, float k, float speed,
-  int num_its, int nthreads)
-{
-  if (nthreads <= 0) {
-    nthreads = 1;
-  }
-  DICEK_DATA(compute_params, cp, nthreads);
-
-  int i;
-  int a_row = 0;
-
-  /* Set up all the parameter blocks */
-  for(i=0; i<nthreads; i++) {
-    cp[i].u = u; cp[i].v = v; cp[i].du = du; cp[i].dv = dv;
-    cp[i].D_u = D_u; cp[i].D_v = D_v;
-    cp[i].F = F; cp[i].k = k; cp[i].speed = speed;
-    cp[i].num_its = num_its; cp[i].start_row = a_row;
-    a_row = (g_height * (i+1)) / nthreads;
-    cp[i].end_row = a_row;
-    cp[i].interlock_type = (nthreads>1) ? 1 : 0;
-  }
-
-  if (nthreads > 1) {
-    /* Start N threads, each will immediately begin the first part of its computation */
-    DICEK_SPLIT(compute_gs_hwiv, cp, nthreads);
-
-    /* For each iteration we sync the threads once. */
-    for(i=0; i<num_its; i++) {
-      DICEK_INTERLOCK(cp, nthreads);
-      DICEK_RESUME(cp, nthreads);
-    }
-
-    DICEK_MERGE(compute_gs_hwiv, cp, nthreads);
-  } else {
-    /* With 1 thread it's more efficient to just call the compute routine directly */
-    compute_gs_hwiv((void *)&(cp[0]));
-  }
-}
-
-#ifdef HWIV_V4F4_SSE2
-
-#define VECSIZE 4
-
-/* The parameter space code, specifically the "if(parameter_space)" test itself, causes a significant slowdown
-   even when the parameter_space flag is false. This #define is here in case we want to get the speed boost
-   by eliminating the -paramspace user option. */
-#define SUPPORT_PARAM_SPACE
-
-void * compute_gs_hwiv(void * gpb)
-{
-  DICEK_SUB(compute_params, gpb);
-  const int full_width = g_width + 2*VECSIZE;
-  const int wid_v = full_width / VECSIZE;
-
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-  /* On Intel we disable accurate handling of denorms and zeros. This is an
-     important speed optimization. */
-  int oldMXCSR = _mm_getcsr(); //read the old MXCSR setting
-  int newMXCSR = oldMXCSR | 0x8040; // set DAZ and FZ bits
-  _mm_setcsr( newMXCSR ); //write the new MXCSR setting to the MXCSR
-#endif
-
-  compute_params * param_block;
-  param_block = (compute_params *) gpb;
-  float *u = param_block->u;
-  float *v = param_block->v;
-  float *du = param_block->du;
-  float *dv = param_block->dv;
-  float D_u = param_block->D_u;
-  float D_v = param_block->D_v;
-  float F = param_block->F;
-  float k = param_block->k;
-  float speed = param_block->speed;
-  int num_its = param_block->num_its;
-  int start_row = param_block->start_row;
-  int end_row = param_block->end_row;
-  int interlock = param_block->interlock_type;
-
-
-  int iter;
-
-#ifdef SUPPORT_PARAM_SPACE
-  const float k_min=0.045, k_max=0.07, F_min=0.01, F_max=0.09;
-  float k_diff;
-  V4F4 v4_kdiff;
-  k_diff = (k_max-k_min)/g_height;
-  v4_kdiff = v4SET(0, -k_diff, -2*k_diff, -3*k_diff);
-#endif
-
-  if (interlock) { DICEK_CH_BEGIN }
-
-  // Scan per iteration
-  for(iter = 0; iter < num_its; iter++) {
-
-  // Scan per row
-  for(int i = start_row; i < end_row; i++) {
-#ifdef SUPPORT_PARAM_SPACE
-    V4F4 v4_F = v4SPLAT(F);
-    V4F4 v4_k = v4SPLAT(k);
-#else
-    const V4F4 v4_F = v4SPLAT(F);
-    const V4F4 v4_k = v4SPLAT(k);
-#endif
-    const V4F4 v4_Du = v4SPLAT(D_u);
-    const V4F4 v4_Dv = v4SPLAT(D_v);
-    int iprev,inext;
-    if (g_wrap) {
-      /* Periodic boundary condition */
-      iprev = (i+g_height-1) % g_height;
-      inext = (i+1) % g_height;
-    } else {
-      /* The edges are their own neighbors. This amounts to a Neumann boundary condition. */
-      iprev = max(i-1, 0);
-      inext = min(i+1, g_height-1);
-    }
-
-#ifdef SUPPORT_PARAM_SPACE
-    if (g_paramspace) {
-      // set F for this row (ignore the provided value)
-      F = F_min + (g_height-i-1) * (F_max-F_min)/g_height;
-      v4_F = v4SPLAT(F);
-    }
-#endif
-
-    // Scan per column in steps of vector width
-    for(int j = 1; j < wid_v-1; j++) {
-      V4F4 * v_ubase = ((V4F4 *)u)+i*wid_v+j;
-      V4F4 * v_vbase = ((V4F4 *)v)+i*wid_v+j;
-      V4F4 * v_dubase = ((V4F4 *)du)+i*wid_v+j;
-      V4F4 * v_dvbase = ((V4F4 *)dv)+i*wid_v+j;
-      V4F4 u_left = _mm_loadu_ps(((float*)v_ubase)-1);
-      V4F4 u_right = _mm_loadu_ps(((float*)v_ubase)+1);
-      V4F4 v_left = _mm_loadu_ps(((float*)v_vbase)-1);
-      V4F4 v_right = _mm_loadu_ps(((float*)v_vbase)+1);
-      V4F4 * v_ub_prev = ((V4F4 *)u)+iprev*wid_v+j;
-      V4F4 * v_ub_next = ((V4F4 *)u)+inext*wid_v+j;
-      V4F4 * v_vb_prev = ((V4F4 *)v)+iprev*wid_v+j;
-      V4F4 * v_vb_next = ((V4F4 *)v)+inext*wid_v+j;
-
-#ifdef SUPPORT_PARAM_SPACE
-      if (g_paramspace) {
-        // set k for this column (ignore the provided value)
-        k = k_min + (g_width-(j*4)-5)*k_diff;
-        // k decreases by k_diff each time j increases by 1, so this vector
-        // needs to contain 4 different k values. We use v4_kdiff, pre-computed
-        // above, to accomplish this.
-        v4_k = v4ADD(v4SPLAT(k),v4_kdiff);
-      }
-#endif
-
-      // To compute the Laplacians of u and v, we use the 5-point neighbourhood for the Euler discrete method:
-      //    nabla(x) = x[i][j-1]+x[i][j+1]+x[i-1][j]+x[i+1][j] - 4*x[i][j];
-      // ("nabla" is the name of the "upside down delta" symbol used for the Laplacian in equations)
-#     define NABLA_5PT(ctr,left,right,up,down) \
-         v4SUB(v4ADD(v4ADD(v4ADD(left,right),up),down),v4MUL(ctr,v4SPLAT(4.0f)))
-
-      // compute the new rate of change of u and v
-      V4F4 v4_uvv = v4MUL(*v_ubase,v4MUL(*v_vbase,*v_vbase)); // u*v^2 is used twice
-
-      /* Scalar code is:     du[i][j] = D_u * nabla_u - u*v^2 + F*(1-u);
-         We treat it as:                D_u * nabla_u - (u*v^2 - F*(1-u)) */
-      *v_dubase = v4SUB(v4MUL(v4_Du,
-         NABLA_5PT(*v_ubase, u_left, u_right, *v_ub_prev, *v_ub_next)),
-                           v4SUB(v4_uvv,v4MUL(v4_F,v4SUB(v4SPLAT(1.0f),*v_ubase))));
-
-      /* dv formula is similar:  dv[i][j] = D_v * nabla_v + u*v^2 - (F+k)*v; */
-      *v_dvbase = v4ADD(v4MUL(v4_Dv,
-         NABLA_5PT(*v_vbase, v_left, v_right, *v_vb_prev, *v_vb_next)),
-                           v4SUB(v4_uvv,v4MUL(v4ADD(v4_F,v4_k),*v_vbase)));
-    }
-
-  } // End of scan per row
-
-  if (interlock) { DICEK_CH_SYNC }
-  if (interlock) { DICEK_CH_BEGIN }
-
-  {
-    int right_b, left_b;
-    if (g_wrap) {
-      right_b = wid_v-2;
-      left_b = 1;
-    } else {
-      right_b = 1;
-      left_b = wid_v-2;
-    }
-  // effect change
-    for(int i = start_row; i < end_row; i++) {
-      for(int j = 1; j < wid_v-1; j++) {
-        const V4F4 v4_speed = v4SPLAT(speed);
-        V4F4 * v_ubase = ((V4F4 *)u)+i*wid_v+j;
-        V4F4 * v_vbase = ((V4F4 *)v)+i*wid_v+j;
-        V4F4 * v_dubase = ((V4F4 *)du)+i*wid_v+j;
-        V4F4 * v_dvbase = ((V4F4 *)dv)+i*wid_v+j;
-        // u[i][j] = u[i][j] + speed * du[i][j];
-        *v_ubase = v4ADD(v4MUL(v4_speed, *v_dubase), *v_ubase);
-        // v[i][j] = v[i][j] + speed * dv[i][j];
-        *v_vbase = v4ADD(v4MUL(v4_speed, *v_dvbase), *v_vbase);
-      }
-      // Update cells on boundary from one row inland
-      *(((V4F4 *)u)+i*wid_v) = *(((V4F4 *)u)+i*wid_v+right_b);
-      *(((V4F4 *)v)+i*wid_v) = *(((V4F4 *)v)+i*wid_v+right_b);
-      *(((V4F4 *)u)+i*wid_v+wid_v-1) = *(((V4F4 *)u)+i*wid_v+left_b);
-      *(((V4F4 *)v)+i*wid_v+wid_v-1) = *(((V4F4 *)v)+i*wid_v+left_b);
-    }
-  }
-
-  } // End of scan per iteration
-
-  if (interlock) { DICEK_CH_SYNC }
-
-  DICEK_CH_END
-}
-#endif
-
-
-#ifdef HWIV_V4F4_PDE4
-
-/*
- This SSE code is based on the method in Munafo's PDE4 program, which tries to reduce memory accesses
-by using aligned loads and shifting data internally to vector registers. The more obvious approach above
-turned out to be faster.
- */
-
-#define VECSIZE 4
-
-/* The parameter space code, specifically the "if(parameter_space)" test itself, causes a 2.5% slowdown even when the
-   parameter_space flag is false */
-#define SUPPORT_PARAM_SPACE
-
-void * compute_gs_hwiv(void * gpb)
-{
-  DICEK_SUB(compute_params, gpb);
-
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-  /* On Intel we disable accurate handling of denorms and zeros. This is an
-     important speed optimization. */
-  int oldMXCSR = _mm_getcsr(); //read the old MXCSR setting
-  int newMXCSR = oldMXCSR | 0x8040; // set DAZ and FZ bits
-  _mm_setcsr( newMXCSR ); //write the new MXCSR setting to the MXCSR
-#endif
-
-  compute_params * param_block;
-  param_block = (compute_params *) gpb;
-  float *u = param_block->u;
-  float *v = param_block->v;
-  float *du = param_block->du;
-  float *dv = param_block->dv;
-  float D_u = param_block->D_u;
-  float D_v = param_block->D_v;
-  float F = param_block->F;
-  float k = param_block->k;
-  float speed = param_block->speed;
-  int parameter_space = param_block->parameter_space;
-  int num_its = param_block->num_its;
-  int width = param_block->width;
-  int height = param_block->height;
-  bool wrap = param_block->wrap;
-  int start_row = param_block->start_row;
-  int end_row = param_block->end_row;
-  int interlock = param_block->interlock_type;
-
-  int iter;
-#ifndef HWIV_HAVE_V4F4
-  fprintf(stdout, "Did not get vector macros from HWIV\n");
-  exit(-1);
-#endif
-  // Vector "constants": speed, F, k, D_u, D_v
-  V4F4 v4_speed, v4_F, v4_k, v4_Du, v4_Dv;
-  // Pointers used to load data from rows of the grid
-  V4F4 *v_ub_prev, *v_ubase, *v_ub_next;
-  V4F4 *v_vb_prev, *v_vbase, *v_vb_next;
-  // Actual grid data
-  V4F4 v4_u_l, v4_u, v4_u_r;
-  V4F4 v4_v_l, v4_v, v4_v_r;
-  V4F4 v4_uvv;
-  // Pointers to second grid where we write the results of the main computation
-  V4F4 *v_dubase, *v_dvbase;
-
-#ifdef SUPPORT_PARAM_SPACE
-  const float k_min=0.045f, k_max=0.07f, F_min=0.01f, F_max=0.09f;
-  float k_diff;
-  V4F4 v4_kdiff;
-  k_diff = (k_max-k_min)/height;
-  v4_kdiff = v4SET(0, -k_diff, -2*k_diff, -3*k_diff);
-#endif
-
-  // Initialize our vectorized scalars
-  v4_speed = v4SPLAT(speed);
-  v4_F = v4SPLAT(F);
-  v4_k = v4SPLAT(k);
-  v4_Du = v4SPLAT(D_u);
-  v4_Dv = v4SPLAT(D_v);
-
-  if (interlock) { DICEK_CH_BEGIN }
-
-  // Scan per iteration
-  for(iter = 0; iter < num_its; iter++) {
-
-//printf("iter %d rows [%ld,%ld)\n",iter,start_row,end_row);
-
-  // Scan per row
-  for(int i = start_row; i < end_row; i++) {
-    int iprev,inext;
-    int v_j2;
-    if (wrap) {
-      /* Periodic boundary condition */
-      iprev = (i+height-1) % height;
-      inext = (i+1) % height;
-    } else {
-      /* The edges are their own neighbors. This amounts to a Neumann boundary condition. */
-      iprev = max(i-1, 0);
-      inext = min(i+1, height-1);
-    }
-
-    /* Get pointers to beginning of rows for each of the grids. We access
-       3 rows each for u and v, and 1 row each for du and dv. */
-    v_ubase = (V4F4 *)&(u[i*width]);
-    v_ub_prev = (V4F4 *)&(u[iprev*width]);
-    v_ub_next = (V4F4 *)&(u[inext*width]);
-    v_vbase = (V4F4 *)&(v[i*width]);
-    v_vb_prev = (V4F4 *)&(v[iprev*width]);
-    v_vb_next = (V4F4 *)&(v[inext*width]);
-    v_dubase = (V4F4 *)&(du[i*width]);
-    v_dvbase = (V4F4 *)&(dv[i*width]);
-
-#ifdef SUPPORT_PARAM_SPACE
-    if (parameter_space) {
-      // set F for this row (ignore the provided value)
-      F = F_min + (height-i-1) * (F_max-F_min)/width;
-      v4_F = v4SPLAT(F);
-    }
-#endif
-
-    /* Pre-load the first two blocks of data we need, which are the "center"
-       and "right" blocks from the end of the row (as if we have just wrapped
-       around from the end of the row back to the beginning) */
-    v_j2 = wrap ? ((width-4)/VECSIZE) : 0;
-
-    v4_u = *(v_ubase+v_j2);
-    v4_v = *(v_vbase+v_j2);
-    v4_u_r = *v_ubase++;
-    v4_v_r = *v_vbase++;
-
-    // Scan per column in steps of vector width
-    for(int j = 0; j < width-VECSIZE; j+=VECSIZE) {
-      // Get a new 4 pixels from the current row and shift the other 8 pixels over
-      v4_u_l = v4_u; v4_u = v4_u_r; v4_u_r = *v_ubase;
-      v4_v_l = v4_v; v4_v = v4_v_r; v4_v_r = *v_vbase;
-
-#ifdef SUPPORT_PARAM_SPACE
-      if (parameter_space) {
-        // set k for this column (ignore the provided value)
-        k = k_min + (width-j-1)*k_diff;
-        // k decreases by k_diff each time j increases by 1, so this vector
-        // needs to contain 4 different k values. We use v4_kdiff, pre-computed
-        // above, to accomplish this.
-        v4_k = v4ADD(v4SPLAT(k),v4_kdiff);
-      }
-#endif
-
-      // To compute the Laplacians of u and v, we use the 5-point neighbourhood for the Euler discrete method:
-      //    nabla(x) = x[i][j-1]+x[i][j+1]+x[i-1][j]+x[i+1][j] - 4*x[i][j];
-      // ("nabla" is the name of the "upside down delta" symbol used for the Laplacian in equations)
-#     define NABLA_5PT(ctr,left,right,up,down) \
-         v4SUB(v4ADD(v4ADD(v4ADD(left,right),up),down),v4MUL(ctr,v4SPLAT(4.0f)))
-
-      // compute the new rate of change of u and v
-      v4_uvv = v4MUL(v4_u,v4MUL(v4_v,v4_v)); // u*v^2 is used twice
-
-      /* Scalar code is:     du[i][j] = D_u * nabla_u - u*v^2 + F*(1-u);
-         We treat it as:                D_u * nabla_u - (u*v^2 - F*(1-u)) */
-      *v_dubase = v4SUB(v4MUL(v4_Du,
-         NABLA_5PT(v4_u, v4RAISE(v4_u,v4_u_l), v4LOWER(v4_u,v4_u_r), *v_ub_prev, *v_ub_next)),
-                           v4SUB(v4_uvv,v4MUL(v4_F,v4SUB(v4SPLAT(1.0f),v4_u))));
-
-      /* dv formula is similar:  dv[i][j] = D_v * nabla_v + u*v^2 - (F+k)*v; */
-      *v_dvbase = v4ADD(v4MUL(v4_Dv,
-         NABLA_5PT(v4_v, v4RAISE(v4_v,v4_v_l), v4LOWER(v4_v,v4_v_r), *v_vb_prev, *v_vb_next)),
-                           v4SUB(v4_uvv,v4MUL(v4ADD(v4_F,v4_k),v4_v)));
-
-      v_ub_prev++; v_ub_next++;
-      v_ubase++;   v_vbase++;
-      v_vb_prev++; v_vb_next++;
-      v_dubase++;  v_dvbase++;
-    }
-
-    /* Now we do the last 4 pixels. This is unrolled out of the main loop just to avoid having to do the wrap
-       test every time in the j loop. */
-    v4_u_l = v4_u; v4_u = v4_u_r;
-    v4_v_l = v4_v; v4_v = v4_v_r;
-    if (wrap) {
-      /* The 4 cells to the "right" are the first 4 in this row */
-      v4_u_r = *((V4F4 *)&(u[i*width]));
-      v4_v_r = *((V4F4 *)&(v[i*width]));
-    } else {
-      /* just leave them alone, retaining the rightmost 4 values in this row, which were loaded on the last iteration
-         through the loop */
-    }
-    v4_uvv = v4MUL(v4_u,v4MUL(v4_v,v4_v));
-    *v_dubase = v4SUB(v4MUL(v4_Du,
-         NABLA_5PT(v4_u, v4RAISE(v4_u,v4_u_l), v4LOWER(v4_u,v4_u_r), *v_ub_prev, *v_ub_next)),
-                           v4SUB(v4_uvv,v4MUL(v4_F,v4SUB(v4SPLAT(1.0f),v4_u))));
-    *v_dvbase = v4ADD(v4MUL(v4_Dv,
-         NABLA_5PT(v4_v, v4RAISE(v4_v,v4_v_l), v4LOWER(v4_v,v4_v_r), *v_vb_prev, *v_vb_next)),
-                           v4SUB(v4_uvv,v4MUL(v4ADD(v4_F,v4_k),v4_v)));
-  } // End of scan per row
-
-  // First thread interlock goes here
-  if (interlock) { DICEK_CH_SYNC }
-  if (interlock) { DICEK_CH_BEGIN }
-
-  {
-  // effect change
-    for(int i = start_row; i < end_row; i++) {
-      v_ubase = ((V4F4 *) (&(u[i*width])));
-      v_vbase = ((V4F4 *) (&(v[i*width])));
-      v_dubase = ((V4F4 *) (&(du[i*width])));
-      v_dvbase = ((V4F4 *) (&(dv[i*width])));
-      for(int j = 0; j < width; j+=VECSIZE) {
-        // u[i][j] = u[i][j] + speed * du[i][j];
-        *v_ubase = v4ADD(v4MUL(v4_speed, *v_dubase), *v_ubase); v_ubase++; v_dubase++;
-        // v[i][j] = v[i][j] + speed * dv[i][j];
-        *v_vbase = v4ADD(v4MUL(v4_speed, *v_dvbase), *v_vbase); v_vbase++; v_dvbase++;
-      }
-    }
-  }
-
-  } // End of scan per iteration
-
-  // second thread interlock goes here
-  if (interlock) { DICEK_CH_SYNC }
-
-  DICEK_CH_END
-}
-#endif
-
-
-#ifdef HWIV_EMULATE
-
-/* This is the old version of compute() that used all "assembly-language" syntax */
-#define INDEX(a,x,y) ((a)[(x)*g_width+(y)])
-
-void compute(float *u, float *v, float *du, float *dv,
-             float D_u,float D_v,float F,float k,float speed,
-             int parameter_space)
-{
-#ifndef HWIV_HAVE_V4F4
-  fprintf(stdout, "Did not get vector macros from HWIV\n");
-  exit(-1);
-#endif
-  V4F4 v4_speed; // vectorized version of speed scalar
-  V4F4 v4_F; // vectorized version of F scalar
-  V4F4 v4_k; // vectorized version of k scalar
-  HWIV_4F4_ALIGNED talign; // used by FILL_4F4
-  V4F4 v4_u; V4F4 v4_du;
-  V4F4 v4_v; V4F4 v4_dv;
-  HWIV_INIT_MUL0_4F4; // used by MUL (on targets that need it)
-  HWIV_INIT_MTMP_4F4; // used by MADD (on targets that need it)
-  HWIV_INIT_FILL;     // used by FILL
-  HWIV_INIT_RLTMP_4F4; // used by RAISE and LOWER
-  V4F4 v4_tmp;
-  V4F4 v4_Du;
-  V4F4 v4_Dv;
-  V4F4 v4_nabla_u;
-  V4F4 v4_nabla_v;
-  V4F4 v4_1;
-  V4F4 v4_4;
-  const float k_min=0.045f, k_max=0.07f, F_min=0.01f, F_max=0.09f;
-  float k_diff;
-  V4F4 v4_kdiff;
-  float * ubase;
-  float * ub_prev;
-  float * ub_next;
-  float * vbase; float * vb_prev; float * vb_next;
-  float * dubase; float * dvbase;
-
-  V4F4 v4_u_l;
-  V4F4 v4_u_r;
-  V4F4 v4_v_l;
-  V4F4 v4_v_r;
-
-  //F_diff = (F_max-F_min)/width;
-  k_diff = (k_max-k_min)/height;
-
-  // Initialize our vectorized scalars
-  HWIV_SPLAT_4F4(v4_speed, speed);
-  HWIV_SPLAT_4F4(v4_F, F);
-  HWIV_SPLAT_4F4(v4_k, k);
-  HWIV_SPLAT_4F4(v4_Du, D_u);
-  HWIV_SPLAT_4F4(v4_Dv, D_v);
-  HWIV_SPLAT_4F4(v4_1, 1.0);
-  HWIV_SPLAT_4F4(v4_4, 4.0);
-  HWIV_FILL_4F4(v4_kdiff, 0, -k_diff, -2*k_diff, -3*k_diff);
-
-  // Scan per row
-  for(int i = 0; i < height; i++) {
-    int iprev,inext;
-    int j2;
-
-    if (g_wrap) {
-      iprev = (i+height-1) % height;
-      inext = (i+1) % height;
-    } else {
-      iprev = max(i-1, 0);
-      inext = min(i+1, height-1);
-    }
-    /* Get pointers to beginning of rows for each of the grids. We access
-       3 rows each for u and v, and 1 row each for du and dv. */
-    ubase = &INDEX(u,i,0);
-    ub_prev = &INDEX(u,iprev,0);
-    ub_next = &INDEX(u,inext,0);
-    vbase = &INDEX(v,i,0);
-    vb_prev = &INDEX(v,iprev,0);
-    vb_next = &INDEX(v,inext,0);
-    dubase = &INDEX(du,i,0);
-    dvbase = &INDEX(dv,i,0);
-
-    if (parameter_space) {
-      // set F for this row (ignore the provided value)
-      F = F_min + (height-i-1) * (F_max-F_min)/width;
-      HWIV_SPLAT_4F4(v4_F, F);
-    }
-
-    /* Pre-load the first two blocks of data we need, which are the "center"
-       and "right" blocks from the end of the row (as if we have just wrapped
-       around from the end of the row back to the beginning) */
-    j2 = g_wrap ? (width-4) : 0;
-    HWIV_LOAD_4F4(v4_u, ubase+j2);
-    HWIV_LOAD_4F4(v4_u_r, ubase);
-    HWIV_LOAD_4F4(v4_v, vbase+j2);
-    HWIV_LOAD_4F4(v4_v_r, vbase);
-
-    // Scan per column in steps of vector width
-    for(int j = 0; j < width; j+=4) {
-      if (g_wrap) {
-        j2 = (j+4) % width;
-      } else {
-        j2 = min(j+4, height-4);
-      }
-
-      HWIV_COPY_4F4(v4_u_l, v4_u);
-      HWIV_COPY_4F4(v4_v_l, v4_v);
-      HWIV_COPY_4F4(v4_u, v4_u_r);
-      HWIV_COPY_4F4(v4_v, v4_v_r);
-      HWIV_LOAD_4F4(v4_u_r, ubase+j2);
-      HWIV_LOAD_4F4(v4_v_r, vbase+j2);
-
-      if (parameter_space) {
-        // set k for this column (ignore the provided value)
-        k = k_min + (width-j-1)*k_diff;
-        // k decreases by k_diff each time j increases by 1, so this vector
-        // needs to contain 4 different k values.
-        HWIV_SPLAT_4F4(v4_tmp, k);
-        HWIV_ADD_4F4(v4_k, v4_tmp, v4_kdiff);
-      }
-
-      // compute the Laplacians of u and v. "nabla" is the name of the
-      // "upside down delta" symbol used for the Laplacian in equations
-
-      /* Scalar code is:
-         nabla_u = u[i][jprev]+u[i][jnext]+u[iprev][j]+u[inext][j] - 4*uval; */
-      HWIV_RAISE_4F4(v4_nabla_u, v4_u, v4_u_l);
-
-      HWIV_LOWER_4F4(v4_tmp, v4_u, v4_u_r);
-      HWIV_ADD_4F4(v4_nabla_u, v4_nabla_u, v4_tmp);
-
-      // Now we add in the "up" and "down" neighbors
-      HWIV_LOAD_4F4(v4_tmp, ub_prev+j);
-      HWIV_ADD_4F4(v4_nabla_u, v4_nabla_u, v4_tmp);
-      HWIV_LOAD_4F4(v4_tmp, ub_next+j);
-      HWIV_ADD_4F4(v4_nabla_u, v4_nabla_u, v4_tmp);
-
-      // Now we compute -(4*u-neighbors)  = neighbors - 4*u
-      HWIV_NMSUB_4F4(v4_nabla_u, v4_4, v4_u, v4_nabla_u);
-
-      // Same thing all over again for the v's
-      HWIV_RAISE_4F4(v4_nabla_v, v4_v, v4_v_l);
-      HWIV_LOWER_4F4(v4_tmp, v4_v, v4_v_r);
-      HWIV_ADD_4F4(v4_nabla_v, v4_nabla_v, v4_tmp);
-      HWIV_LOAD_4F4(v4_tmp, vb_prev+j);
-      HWIV_ADD_4F4(v4_nabla_v, v4_nabla_v, v4_tmp);
-      HWIV_LOAD_4F4(v4_tmp, vb_next+j);
-      HWIV_ADD_4F4(v4_nabla_v, v4_nabla_v, v4_tmp);
-      HWIV_NMSUB_4F4(v4_nabla_v, v4_4, v4_v, v4_nabla_v);
-
-      // compute the new rate of change of u and v
-
-      /* Scalar code is:
-           du[i][j] = D_u * nabla_u - uval*vval*vval + F*(1-uval);
-         We treat it as:
-                      D_u * nabla_u - (uval*vval*vval - (-(F*uval-F)) ) */
-
-      HWIV_NMSUB_4F4(v4_tmp, v4_F, v4_u, v4_F);        // -(F*u-F) = F-F*u = F(1-u)
-      HWIV_MUL_4F4(v4_dv, v4_v, v4_v);                 // v^2
-      HWIV_MSUB_4F4(v4_tmp, v4_u, v4_dv, v4_tmp);      // u*v^2 - F(1-u)
-      HWIV_MSUB_4F4(v4_du, v4_Du, v4_nabla_u, v4_tmp); // D_u*nabla_u - (u*v^2 - F(1-u))
-                                                       // = D_u*nabla_u - u*v^2 + F(1-u)
-      HWIV_SAVE_4F4(dubase+j, v4_du);
-
-      /* dv formula is similar:
-           dv[i][j] = D_v * nabla_v + uval*vval*vval - (F+k)*vval;
-         We treat it as:
-                      D_v * nabla_v + uval*vval*vval - (F*vval + k*vval); */
-      HWIV_MUL_4F4(v4_tmp, v4_k, v4_v);                // k*v
-      HWIV_MADD_4F4(v4_tmp, v4_F, v4_v, v4_tmp);       // F*v+k*v = (F+k)v
-                                                       // v^2 is still in v4_dv
-      HWIV_MSUB_4F4(v4_tmp, v4_u, v4_dv, v4_tmp);      // u*v^2 - (F+k)v
-      HWIV_MADD_4F4(v4_dv, v4_Dv, v4_nabla_v, v4_tmp); // D_v*nabla_v + u*v^2 - (F+k)v
-      HWIV_SAVE_4F4(dvbase+j, v4_dv);
-    }
-  }
-
-  // effect change
-    for(int i = 0; i < height; i++) {
-      ubase = &INDEX(u,i,0);
-      vbase = &INDEX(v,i,0);
-      dubase = &INDEX(du,i,0);
-      dvbase = &INDEX(dv,i,0);
-      for(int j = 0; j < width; j+=4) {
-        // u[i][j] = u[i][j] + speed * du[i][j];
-        HWIV_LOAD_4F4(v4_u, ubase+j);               // get u
-        HWIV_LOAD_4F4(v4_du, dubase+j);             // get du
-        HWIV_MADD_4F4(v4_u, v4_speed, v4_du, v4_u); // speed*du + u
-        HWIV_SAVE_4F4(ubase+j, v4_u);               // write it back
-
-        // v[i][j] = v[i][j] + speed * dv[i][j];
-        HWIV_LOAD_4F4(v4_v, vbase+j);
-        HWIV_LOAD_4F4(v4_dv, dvbase+j);
-        HWIV_MADD_4F4(v4_v, v4_speed, v4_dv, v4_v);
-        HWIV_SAVE_4F4(vbase+j, v4_v);
-      }
-    }
-}
-
-#endif
diff --git a/Ready_old_CLI/calc_Gray_Scott/gray_scott_hwivector.h b/Ready_old_CLI/calc_Gray_Scott/gray_scott_hwivector.h
deleted file mode 100644
index 1d4dbd8e9..000000000
--- a/Ready_old_CLI/calc_Gray_Scott/gray_scott_hwivector.h
+++ /dev/null
@@ -1,36 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-//#define DICEK_EMULATE
-#include "dicek.h"
-
-typedef struct compute_params {
-  DICEK_THREAD_VARS;
-  float *u;
-  float *v;
-  float *du;
-  float *dv;
-  float D_u;
-  float D_v;
-  float F;
-  float k;
-  float speed;
-  int num_its;
-  int start_row;
-  int end_row;
-  int interlock_type;
-} compute_params;
-
-void gs_hwi_compute_setup(int width, int height, bool wrap, bool paramspace);
-
-void compute_dispatch(float *u, float *v, float *du, float *dv,
-  float D_u, float D_v, float F, float k, float speed,
-  int num_its, int num_threads);
-
-void * compute_gs_hwiv(void * param_block);  // Arg is really "compute_params * param_block"
diff --git a/Ready_old_CLI/calc_Gray_Scott/gray_scott_scalar.cpp b/Ready_old_CLI/calc_Gray_Scott/gray_scott_scalar.cpp
deleted file mode 100644
index a2ce7c127..000000000
--- a/Ready_old_CLI/calc_Gray_Scott/gray_scott_scalar.cpp
+++ /dev/null
@@ -1,75 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See ../../README.txt for more details.
-
-*/
-
-#include "gray_scott_scalar.h"
-#include "../util.h"
-
-static int g_width;
-static int g_height;
-static bool g_wrap;
-static bool g_paramspace;
-
-void gs_scl_compute_setup(int width, int height, bool wrap, bool paramspace)
-{
-  g_width = width;
-  g_height = height;
-  g_wrap = wrap;
-  g_paramspace = paramspace;
-}
-
-void compute_gs_scalar(float *a, float *b, float *da, float *db,
-             float r_a, float r_b, float f, float k, float speed)
-{
-    // compute change in each cell
-    for(int i = 0; i < g_height; i++) {
-        int iprev,inext;
-        if (g_wrap) {
-            iprev = (i + g_height - 1) % g_height;
-            inext = (i + 1) % g_height;
-        } else {
-            iprev = max(0,i-1);
-            inext = min(g_height-1,i+1);
-        }
-
-        for(int j = 0; j < g_width; j++) {
-            int jprev,jnext;
-            if (g_wrap) {
-                jprev = (j + g_width - 1) % g_width;
-                jnext = (j + 1) % g_width;
-            } else {
-                jprev = max(0,j-1);
-                jnext = min(g_width-1,j+1);
-            }
-
-            float aval = a[i*g_width+j];
-            float bval = b[i*g_width+j];
-
-            if(g_paramspace)	{
-                const float kmin=0.045, kmax=0.07, fmin=0.01, fmax=0.09;
-                // set f and k for this location (ignore the provided values of f and k)
-                k = kmin + (g_width-j-1)*(kmax-kmin)/g_width;
-                f = fmin + (g_height-i-1)*(fmax-fmin)/g_height;
-            }
-
-            // compute the Laplacians of a and b
-            float dda = a[i*g_width+jprev] + a[i*g_width+jnext] + a[iprev*g_width+j] + a[inext*g_width+j] - 4*aval;
-            float ddb = b[i*g_width+jprev] + b[i*g_width+jnext] + b[iprev*g_width+j] + b[inext*g_width+j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i*g_width+j] = r_a * dda - aval*bval*bval + f*(1-aval);
-            db[i*g_width+j] = r_b * ddb + aval*bval*bval - (f+k)*bval;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < g_width*g_height; i++) {
-      a[i] += speed * da[i];
-      b[i] += speed * db[i];
-    }
-}
diff --git a/Ready_old_CLI/calc_Gray_Scott/gray_scott_scalar.h b/Ready_old_CLI/calc_Gray_Scott/gray_scott_scalar.h
deleted file mode 100644
index ba396affb..000000000
--- a/Ready_old_CLI/calc_Gray_Scott/gray_scott_scalar.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-void gs_scl_compute_setup(int width, int height, bool wrap, bool paramspace);
-
-void compute_gs_scalar(float *a, float *b, float *da, float *db,
-  float r_a, float r_b, float f, float k, float speed);
diff --git a/Ready_old_CLI/calc_Gray_Scott/hwi_vector.h b/Ready_old_CLI/calc_Gray_Scott/hwi_vector.h
deleted file mode 100644
index ea55acd25..000000000
--- a/Ready_old_CLI/calc_Gray_Scott/hwi_vector.h
+++ /dev/null
@@ -1,426 +0,0 @@
-/*
-  
-  hwi_vector.h                                                                                           ^u 140 ^x f
-
-This is a hardware-independent vector library. It enables vector SIMD code to be written, compiled and run on non-vector
-machines. Once the code is tested and working, a compile-time macro can be changed, and a recompile causes the program
-to use actual vector instructions.
-
-In the simplest case, that is all that is needed. In real applications there are usually several more steps:
-
-  * The application developer decides what base hardware platform to compile for. This might be the oldest CPU the
-    company will support, such as a Pentium 4 HT.
-
-  * The compiler defines certain flags, such as __i386__ and __SSE2__ that can be tested by an #ifdef
-
-  * The source code can set flags of its own, such as HWIV_EMULATE or HWIV_WANT_V4F4. This might be done in order to
-    create two versions of a calculation routine (one that uses vector instructions and one that does not)
-
-  * The program is built from one or more compilations of the same source. This might be done in order to create a
-    "universal binary" capable of being copied to and run on a variety of computer products. A typical example is a
-    program file that contains both a 32-bit and a 64-bit version, and the operating system loads whichever one is
-    appropriate when the program is launched.
-
-  * At run-time, the program tests for the presence of vector instructions using the CPUID instruction or its
-    equivalent on non Intel CPUs.
-
-  * At run-time, based on the CPUID test, the program transfers control to one or another of the calculation
-    subroutines depending on which vector instructions are actually available.
-
-*/
-
-// First, honor the user's request to use emulation
-#ifdef HWIV_EMULATE
-# define HWIV_V4F4_EMULATED
-#endif
-
-// Next, find out if the compiler will give us SSE2 intrinsics
-#ifndef HWIV_V4F4_EMULATED
-# if defined(__SSE2__)
-#   define HWIV_V4F4_SSE2
-# endif
-  // Workaround because Visual Studio doesn't seem to set its _M_IX86_FP flag, 
-  // or give us any indication what level of SSE support is available.
-  // So we just assume SSE2 is available
-# if defined(_M_X64) || defined(_M_IX86)
-#  define HWIV_V4F4_SSE2
-# endif
-#endif
-
-
-// Finally, fall back to emulated if no hardware option is available
-#ifndef HWIV_V4F4_SSE2
-# ifndef HWIV_V4F4_EMULATED
-#   define HWIV_V4F4_EMULATED
-# endif
-#endif
-
-
-// See is client wants V4F4
-#ifdef HWIV_WANT_V4F4
-
-// Okay, now see how we should create the macros for V4F4
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//  #include<stdio.h>                                                                                                //
-//  /*                                                     */                    /*                             */   //
-//  /*                                                     */                    /*                             */   //
-//  int                 main()  {char      a[]     ={     'T',    'h','i',     's',32,     'p','r'        ,'o','g'   //
-//  ,'r','a','m',      32,   'j',   'u',   's',    't'    ,32           ,'d',    'o',   'e'     ,'s'    ,32    ,'i'  //
-//  ,'t'              ,32    ,'t'    ,'h'  ,'e'   ,32,    'h'    ,97,'r','d'   ,32,    'w',97,121,33,   32,    40,   //
-//  68,               'o',   'n',    39,   't'    ,32     ,'y'   ,'o',   117,    32    ,'t'             ,'h'   ,'i'  //
-//  ,'n'               /*     Xy     =a     +3     +n      ++     ;a=     b-     (*     x/z              );     if   //
-//  (Xy-++n<(z+*x))z  =b;a   +b,     z+=     x*/,107 ,    63,63    ,63,41,'\n'   ,00};    puts(a);}      /*.RPM.*/   //
-//                                                                                                                   //
-//        Emulated versions of the V4F4 macros. (These also serve as documentation for what each macro does)         //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-# ifdef HWIV_V4F4_EMULATED
-
-/* Scalar code to emulate the V4F4 vector model */
-#define HWIV_HAVE_V4F4
-
-/* The vector register type */
-typedef float V4F4[4];
-
-/* Aligned memory suitable for load to / store from the vector register */
-typedef float HWIV_4F4_ALIGNED[4];
-
-/* LOAD_4F4: dst is a vector of 4 floats. src is a pointer to an array
-   of floats in memory. This opcode loads 4 consecutive floats from the
-   given location into the vector. The first float from memory will be loaded
-   into element 0 of the vector, the 2nd into element 1, and so on. */
-#define HWIV_LOAD_4F4(dst, src)  { (dst)[0]=(src)[0]; (dst)[1]=(src)[1]; \
-                                   (dst)[2]=(src)[2]; (dst)[3]=(src)[3]; }
-
-/* LOADO_4F4: dst is a vector of 4 floats. src is a pointer to an array
-   of floats in memory. offset is a byte offset, which must be a multiple
-   of sizeof(float).
-     This opcode loads 4 consecutive floats from the given location plus
-   offset into the vector. The first float from memory will be loaded into
-   element 0 of the vector, the 2nd into element 1, and so on. */
-#define HWIV_LOADO_4F4(dst, src, offset)  { (dst)[0]=(src)[(offset)/4]; \
-                                            (dst)[1]=(src)[(offset)/4+1]; \
-                                            (dst)[2]=(src)[(offset)/4+2]; \
-                                            (dst)[3]=(src)[(offset)/4+3]; }
-
-/* COPY_4F4: dst and src are each a vector of 4 floats. This opcode copies
-   the contents of the source vector into the destination vector. */
-#define HWIV_COPY_4F4(dst, src)  HWIV_LOAD_4F4((dst), (src))
-
-/* FILL_4F4: dst is a vector of 4 floats. sc0, sc1, sc2, and sc3 ("scalars")
-   are each a float. tmp is a (float *) pointing to memory of sufficient
-   size to hold four floats, and aligned to a 16-byte boundary. Use the
-   HWIV_4F4_ALIGNED typedef to declare a suitable float[4] which can
-   be passed as tmp to this and other similar macros.
-     This opcode loads the 4 scalar values into the vector. The sc0 will be
-   loaded into element 0 of the vector, sc1 into element 1, and so on. */
-#define HWIV_FILL_4F4(dst, sc0, sc1, sc2, sc3) { \
-            (dst)[0]=(sc0); (dst)[1]=(sc1); (dst)[2]=(sc2); (dst)[3]=(sc3); }
-#define HWIV_INIT_FILL /* nop */
-
-#define HWIV_SPLAT_4F4(dst, s) HWIV_FILL_4F4((dst), (s), (s), (s), (s))
-
-/* SAVE_4F4: dst is a pointer to an array of floats in memory. src is a
-   vector of 4 floats. 
-     This opcode stores the 4 floats in the source vector into 4 consecutive
-   float-sized blocks of memory (i.e. 16 consecutive bytes, 4 bytes per
-   float) beginning at the given destination location. Element 0 of the
-   vector will be stored into the first 4 bytes in memory, element 1 into
-   the next 4 bytes of memory, and so on. */
-#define HWIV_SAVE_4F4(dst, src)  memcpy((void *) (dst), (void *) (src), 16);
-
-/* SAVEO_4F4: dst is a pointer to an array of floats in memory. src is a
-   vector of 4 floats. offset is a byte offset, which must be a multiple
-   of sizeof(float). 
-     This opcode stores the 4 floats in the source vector into 4 consecutive
-   float-sized blocks of memory (i.e. 16 consecutive bytes, 4 bytes per
-   float) beginning at the given destination location plus offset. Element
-   0 of the vector will be stored into the first 4 bytes in memory, element
-   1 into the next 4 bytes of memory, and so on. */
-#define HWIV_SAVEO_4F4(dst, offset, src) { (dst)[(offset)/4]=(src)[0]; \
-                                         (dst)[(offset)/4+1]=(src)[1]; \
-                                         (dst)[(offset)/4+2]=(src)[2]; \
-                                         (dst)[(offset)/4+3]=(src)[3]; }
-
-/* ADD_4F4: dst, a, and b are each a vector of 4 floats.
-     This opcode adds each of the components of a to the corresponding
-   component of b and puts the result into dst. */
-#define HWIV_ADD_4F4(dst, a, b) { (dst)[0]=a[0]+b[0]; (dst)[1]=a[1]+b[1]; \
-                                (dst)[2]=a[2]+b[2]; (dst)[3]=a[3]+b[3]; }
-
-/* SUB_4F4: dst, a, and b are each a vector of 4 floats.
-     This opcode adds each of the components of a to the corresponding
-   component of b and puts the result into dst. */
-#define HWIV_SUB_4F4(dst, a, b) { (dst)[0]=a[0]-b[0]; (dst)[1]=a[1]-b[1]; \
-                                (dst)[2]=a[2]-b[2]; (dst)[3]=a[3]-b[3]; }
-
-#define HWIV_INIT_MUL0_4F4 /* nop */
-
-/* MUL_4F4: dst, a, and b are each a vector of 4 floats. v0 is a variable
-   declared with the HWIV_INIT_MUL0_4F4 macro.
-     This opcode multiplies each of the components of a to the corresponding
-   component of b and puts the result into dst. The varible v0 is used on
-   hardware that has no 2-argument multiply operation. */
-#define HWIV_MUL_4F4(dst, a, b) \
-                             { (dst)[0]=a[0]*b[0]; (dst)[1]=a[1]*b[1]; \
-                                (dst)[2]=a[2]*b[2]; (dst)[3]=a[3]*b[3]; }
-
-#define HWIV_INIT_MTMP_4F4 /* nop */
-
-/* MADD_4F4: dst, a, b, and c are each a vector of 4 floats. t is a variable
-   declared with the HWIV_INIT_MTMP_4F4 macro.
-     This opcode multiplies each of the components of a to the corresponding
-   component of b, then adds the corresponding component of c, and puts the
-   result into dst. The varible t is used on hardware that has no 3-argument
-   multiply-add operation. */
-#define HWIV_MADD_4F4(dst, a, b, c)  { (dst)[0]=a[0]*b[0] + c[0]; \
-                                        (dst)[1]=a[1]*b[1] + c[1]; \
-                                        (dst)[2]=a[2]*b[2] + c[2]; \
-                                        (dst)[3]=a[3]*b[3] + c[3]; }
-
-#define HWIV_MSUB_4F4(dst, a, b, c)  { (dst)[0]=a[0]*b[0] - c[0]; \
-                                           (dst)[1]=a[1]*b[1] - c[1]; \
-                                           (dst)[2]=a[2]*b[2] - c[2]; \
-                                           (dst)[3]=a[3]*b[3] - c[3]; }
-
-/* NMSUB_4F4: dst, a, b, and c are each a vector of 4 floats. t is a variable
-   declared with the HWIV_INIT_MTMP_4F4 macro.
-     This opcode multiplies each of the components of a to the corresponding
-   component of b, then subtracts that product from the corresponding
-   component of c, then puts the result into dst. The varible t is used on
-   hardware that has no 3-argument multiply-add operation. */
-#define HWIV_NMSUB_4F4(dst, a, b, c) { (dst)[0]=c[0] - a[0]*b[0]; \
-                                        (dst)[1]=c[1] - a[1]*b[1]; \
-                                        (dst)[2]=c[2] - a[2]*b[2]; \
-                                        (dst)[3]=c[3] - a[3]*b[3]; }
-
-// Declare this if you are doing any raise or lower operation
-#define HWIV_INIT_RLTMP_4F4 /* nop */
-
-/* RAISE_4F4: dst, src, extra, and tmp are each a vector of 4 floats.
-     This opcode "raises" three of the values from src to the next-higher
-   element of dst. Element 0 of dst is filled with the value from element
-   3 of "extra". The varible t is used on hardware that requires the
-   result to be computed in two pieces and then assembled via a blend
-   operation. ("VSHR_4F4" in old macros) */
-#define HWIV_RAISE_4F4(dst, src, extra) \
-                                      { dst[3]=src[2]; dst[2]=src[1]; \
-                                         dst[1]=src[0]; dst[0]=extra[3]; }
-
-/* LOWER_4F4: dst, src, extra, and tmp are each a vector of 4 floats.
-     This opcode "lowers" three of the values from src to the next-lower
-   element of dst. Element 3 of dst is filled with the value from element
-   0 of "extra". The varible t is used on hardware that requires the
-   result to be computed in two pieces and then assembled via a blend
-   operation. ("VSHL_4F4" in old macros) */
-#define HWIV_LOWER_4F4(dst, src, extra) \
-                                      { dst[0]=src[1]; dst[1]=src[2]; \
-                                         dst[2]=src[3]; dst[3]=extra[0]; }
-
-
-// We also define a small set of macros for FORTRAN-style code. Using these you can build up expressions like
-//
-//    a = v4ADD(v4MUL(b,c),d);  /*  a = b*c + d;  */
-//
-// These macros do not form a complete solution, you still need things like LOAD and SAVE to do any real work.
-
-#define v4ADD(a,b) {(a)[0]+(b)[0],(a)[1]+(b)[1],(a)[2]+(b)[2],(a)[3]+(b)[3]}
-#define v4SUB(a,b) {(a)[0]-(b)[0],(a)[1]-(b)[1],(a)[2]-(b)[2],(a)[3]-(b)[3]}
-#define v4MUL(a,b) {(a)[0]*(b)[0],(a)[1]*(b)[1],(a)[2]*(b)[2],(a)[3]*(b)[3]}
-#define v4SET(v0,v1,v2,v3) {(v0),(v1),(v2),(v3)}
-#define v4SPLAT(a) {(a),(a),(a),(a)}
-#define v4ROUP(a) {(a)[3],(a)[0],(a)[1],(a)[2]}
-#define v4RODN(a) {(a)[1],(a)[2],(a)[3],(a)[0]}
-#define v4RAISE(a, new) {(new)[3],(a)[0],(a)[1],(a)[2]}
-#define v4LOWER(a, new) {(a)[1],(a)[2],(a)[3],(new)[0]}
-
-# endif
-/* - - - - - - - - - - - - - - - - - - - - - End of the EMULATED section  - - - - - - - - - - - - - - - - - - - - - - */
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//                             @@@@                       @@@@                     @@@@                              //
-//                             @@@@                       @@@@                     @@@@                              //
-//                             """"                       @@@@                     @@@@                              //
-//                             eeee    eeee  ,e@@e..   eee@@@@eee                  @@@@                              //
-//                             @@@@    @@@@@@@@@@@@@@. @@@@@@@@@@                  @@@@                              //
-//                             @@@@    @@@@f'    `@@@@    @@@@                     @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@        ,e@@@e.      @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@     e@@@@@@@@@@@e   @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@   .@@@@'     `@@@@i @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@kee@@@@eeeeeeeee@@@@ @@@@                              //
-//                             @@@@    @@@@       @@@@    `@@@@@@@@@@@@@@@@@@@@@@@@@@@@  (R)                         //
-//                                                               @@@@.                                               //
-//                                                               `@@@@e.    .eeee-                                   //
-//                                                                 *@@@@@@@@@@@*                                     //
-//                                                                    "*@@@@*"                                       //
-//                                                                                                                   //
-//             Versions of the V4F4 macros for the Intel SSE2 (or later) 128-bit vector instruction set              //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-# ifdef HWIV_V4F4_SSE2
-
-/* SSE3 implementation of the V4F4 vector model */
-#define HWIV_HAVE_V4F4
-
-# if (defined(HWIV_USE_IMMINTRIN) || defined (__AVX__))
-#   include <immintrin.h>
-# else
-#   include <xmmintrin.h>
-#   include <emmintrin.h>
-# endif
-
-
-typedef __m128 V4F4;
-
-#ifdef _WIN32
-# define ALIGNED_16 __declspec( align( 16 ) )
-#else
-# define ALIGNED_16 __attribute__((aligned (16)))
-#endif
-
-typedef float ALIGNED_16 HWIV_4F4_ALIGNED[4];
-
-#define HWIV_LOAD_4F4(dst, src)        (dst) = _mm_load_ps(src)
-#define HWIV_LOADO_4F4(dst, src, offset) \
-                                       (dst) = _mm_load_ps((src)+(offset)/4)
-#define HWIV_COPY_4F4(dst, src)        (dst) = (src)
-
-#define HWIV_FILL_4F4(dst, sc0, sc1, sc2, sc3) \
-            { HWIV_fill_4F4[0]=(sc0); HWIV_fill_4F4[1]=(sc1); \
-               HWIV_fill_4F4[2]=(sc2); HWIV_fill_4F4[3]=(sc3); \
-               HWIV_LOAD_4F4(dst, HWIV_fill_4F4); }
-
-#define HWIV_INIT_FILL float ALIGNED_16 HWIV_fill_4F4[4];
-
-
-#define HWIV_SPLAT_4F4(dst, s)         (dst) = _mm_set1_ps(s)
-
-#define HWIV_SAVE_4F4(dst, src)        _mm_store_ps((dst), (src))
-
-#define HWIV_SAVEO_4F4(dst, offset, src) \
-                                        _mm_store_ps((dst)+(offset)/4, (src))
-
-#define HWIV_ADD_4F4(dst, a, b) (dst) = _mm_add_ps((a), (b))
-
-#define HWIV_SUB_4F4(dst, a, b) (dst) = _mm_sub_ps((a), (b))
-
-// For INIT_MUL0, on Intel we do nothing because Intel actually has a
-// 2-operand multiply operation.
-#define HWIV_INIT_MUL0_4F4 /* nop */
-
-// For INIT_MTMP, on Intel SSE2 we need to declare a variable because
-// Intel SSE2 has no FMA (fused multiply-add) operations (this is
-// expected to come wth AVX2 on Haskell in 2013)
-#define HWIV_INIT_MTMP_4F4     V4F4 HWIV_mtmp_4F4 = _mm_setzero_ps()
-
-#define HWIV_MUL_4F4(dst, a, b)    (dst) = _mm_mul_ps((a), (b))
-
-#define HWIV_MADD_4F4(dst, a, b, c) { HWIV_mtmp_4F4 = _mm_mul_ps((a), (b)); \
-                                    (dst) = _mm_add_ps(HWIV_mtmp_4F4, (c)); }
-#define HWIV_MSUB_4F4(dst, a, b, c) { HWIV_mtmp_4F4 = _mm_mul_ps((a), (b)); \
-                                    (dst) = _mm_sub_ps(HWIV_mtmp_4F4, (c)); }
-#define HWIV_NMSUB_4F4(dst, a, b, c) { HWIV_mtmp_4F4 = _mm_mul_ps((a), (b)); \
-                                    (dst) = _mm_sub_ps((c), HWIV_mtmp_4F4); }
-
-#define HWIV_INIT_RLTMP_4F4 /* nop */
-
-
-/*
-HWIV_RODN_4F4 (ROtate DOwn) does a "downwards rotate" of a 4-element vector using the Intel VSHUFPS instruction
-(intrinsic _mm_shuffle_ps). If the input is {a,b,c,d} (with a being element 0) the result of the downwards rotate is
-{b,c,d,a} (with each element moving down tot he next-lower slot, except for a which rotates into the top position).
-
-  SRC1  {  x3 ,  x2 ,  x1 ,  x0  }
-  SRC2  {  y3 ,  y2 ,  y1 ,  y0  }
-  DEST  {  y0 ,  y3 ,  x2 ,  x1  }
-  imm8:    00    11    10    01      = 0x39
-
- */
-#define HWIV_RODN_4F4(dest, src)  (dest) = _mm_shuffle_ps((src), (src), 0x39)
-
-/*
-HWIV_ROUP_4F4 is an "upwards rotate": if the input is {a,b,c,d} (with a being element 0) the result is {d,a,b,c}.
-
-  SRC1  {  x3 ,  x2 ,  x1 ,  x0  }
-  SRC2  {  y3 ,  y2 ,  y1 ,  y0  }
-  DEST  {  y2 ,  y1 ,  x0 ,  x3  }
-  imm8:    10    01    00    11      = 0x93
-
- */
-#define HWIV_ROUP_4F4(dest, src)  (dest) = _mm_shuffle_ps((src), (src), 0x93)
-
-/*
-HWIV_RAISE_4F4 is an "upwards shift": if the input is {a,b,c,d} (with a being element 0) the result is {X,a,b,c} with
-the new element X coming from element 3 of the "new" argument.
-
-  new   {  x3 ,  x2 ,  x1 ,  x0  }
-  src   {  y3 ,  y2 ,  y1 ,  y0  }
-  dest  {  y0 ,  y0 ,  x3 ,  x3  }
-  imm8:    00    00    11    11      = 0x0F
-          src0  src0  new3  new3
-
-  dest  {  x3 ,  x2 ,  x1 ,  x0  }
-   src  {  y3 ,  y2 ,  y1 ,  y0  }
-  dest  {  y2 ,  y1 ,  x2 ,  x0  }
-  imm8:    10    01    10    00      = 0x98
-          src2  src1  src0  new3
-*/
-#define HWIV_RAISE_4F4(dest, src, new) { (dest) = _mm_shuffle_ps((new), (src), 0x0f); \
-                                         (dest) = _mm_shuffle_ps((dest), (src), 0x98);  }
-
-/*
-HWIV_LOWER_4F4 is an "downwards shift": if the input is {a,b,c,d} (with a being element 0) the result is {b,c,d,X} with
-the new element X coming from element 0 of the "new" argument.
-
-To accomplish a downwards shift we can just use _mm_move_ss to move a single scalar into the bottom position and then do
-a RODN (downwards rotate)
- */
-#define HWIV_LOWER_4F4(dest, src, new) {  (dest) = _mm_move_ss((src), (new));  \
-                                           HWIV_RODN_4F4(dest, dest); }
-
-// Here is the subset for FORTRAN-style code
-#define v4ADD(a,b) _mm_add_ps((a), (b))
-#define v4SUB(a,b) _mm_sub_ps((a), (b))
-#define v4MUL(a,b) _mm_mul_ps((a), (b))
-// in v4SET, note the reversal of argument order
-#define v4SET(v0,v1,v2,v3) _mm_set_ps((v3),(v2),(v1),(v0))
-#define v4SPLAT(a) _mm_set1_ps(a)
-#define v4ROUP(src) _mm_shuffle_ps((src), (src), 0x93)
-#define v4RODN(src) _mm_shuffle_ps((src), (src), 0x39)
-#define v4RAISE(src, new) _mm_shuffle_ps(_mm_shuffle_ps((new), (src), 0x0f), (src), 0x98)
-#define v4LOWER(src, new) _mm_shuffle_ps(_mm_move_ss((src),(new)), _mm_move_ss((src),(new)), 0x39)
-
-# endif
-/* - - - - - - - - - - - - - - - - - - - - - - End of the INTEL section - - - - - - - - - - - - - - - - - - - - - - - */
-
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//                                                                                                                   //
-//                         /^^^^^\                                       /^^^^^\  _-^^^^^^/ TM                       //
-//                        /  ,-   )_-----_ --- ---- ----..----.  ---.---/  ,-   )/   ,---/                           //
-//                       /  /_)  //  __   )| |/   |/  _// .-   )/    _//  /_)  //   /                                //
-//                      /  //   //  / /  /|          / / (/___//   .^ /  //   /(   |                                 //
-//                     /  / ^^^'(   (/  / |   /|   _/ (  `----/   /  /  / ^^^' |   `---/                             //
-//                    /__/       \____-'  |__/ |__/    \____//___/  /__/        \_____/                              //
-//                                                                                                                   //
-//                                                                                                                   //
-//                Versions of the V4F4 macros for the PowerPC AltiVec 128-bit vector instruction set                 //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-/*
-
-The AltiVec instruction set first came with the "G4" (744x and 745x) processors from Motorola, then the "G5" (97x)
-series from IBM, the "Cell" 8-core CPU used in the Sony Playstation 3, and in IBM's POWER6 (and later) server CPUs.
-
-  Not yet implemented -- do we care about AltiVec? */
-
-#endif
diff --git a/Ready_old_CLI/main-ready.cpp b/Ready_old_CLI/main-ready.cpp
deleted file mode 100644
index 847d80ef1..000000000
--- a/Ready_old_CLI/main-ready.cpp
+++ /dev/null
@@ -1,930 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See ../README.txt for more details.
-
-*/
-
-// hardware
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-# include <xmmintrin.h>
-# define ALLOC_USE_MM
-#endif
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-
-	// TODO: Make sure this is the correct include file for the FindFirstFile function (used in FILE_LENGTH macro below)
-    #include <windows.h>
-#else
-    #include <sys/stat.h>
-    #include <sys/time.h>
-#endif
-
-#include "util.h"
-
-#include "ready_display.h"
-
-#include "brusselator.h"
-
-#include "gray_scott_scalar.h"
-#include "gray_scott_hwivector.h"
-
-static int g_width = 256;
-static int g_height = 256;
-
-float *allocate(int width, int height, const char * error_text, bool for_mm);
-float *allocate(int width, int height, const char * error_text, bool for_mm)
-{
-  long sz;
-  float * rv;
-
-  sz = ((long) width) * ((long) height) * sizeof(*rv);
-#ifdef ALLOC_USE_MM
-  if (for_mm) {
-    rv = (float *) _mm_malloc(sz, 16);
-  } else {
-#endif
-    rv = (float *) malloc(((size_t)sz));
-#ifdef ALLOC_USE_MM
-  }
-#endif
-
-  if (rv == NULL) {
-    fprintf(stderr, "allocate: Could get %ld bytes for %s\n", ((long) sz),
-      error_text);
-    exit(-1);
-  }
-  return (rv);
-}
-
-void init(float *a, float *b, int width, int height,
-          int density, int lowback, int BZrects);
-
-#define MAX_LOADS 100
-char * load_opts[MAX_LOADS];
-int num_loads = 0;
-
-void pearson_block(float *u, float *v, int width, int height,
-                 int vpos, int hpos, int h, int w, float U, float V);
-void ramp_block(float *u, float *v, int width, int height,
-                int vpos, int hpos, int h, int w, int fliph);
-void homogen_uv(float F, float k, float * hU, float * hV);
-void i5_bkg(float *u, float *v, int width, int height, int which);
-void load_option(float * u, float * v, int width, int height, const char * option);
-bool digit_p(char c);
-void pattern_load(float * u, float * v, int width, int height,
-  const char * pattern_filename, int x, int y, int orient);
-int next_patsize(int former);
-void byteswap_2_double(double * array);
-
-
-static const char * g_rd_name = "";
-static int g_color = 0;
-static int g_oldcolor = 0;
-static int g_pastel_mode = 0;
-static bool g_paramspace = false;
-static bool g_wrap = false;
-static float g_k, g_F;
-static int g_ramprects = 0;
-static int g_lowback = 0;
-static float g_scale = 1.0;
-static int g_density = 0;
-static bool g_video = false;
-static int g_threads;
-static bool g_autobright = false;
-
-#define READY_MODULE_GS_SCALAR 1
-#define READY_MODULE_GS_HWIV 2
-#define READY_MODULE_BRUSSELATOR 3
-
-static int g_module = READY_MODULE_GS_HWIV;
-
-int main(int argc, char * * argv)
-{
-  long frames_per_display = 200;
-  long i;
-  DICEK_INIT_NTHR(g_hw_threads)
-  g_threads = g_hw_threads;
-
-  float uv_range = 1.0;
-
-  float bruss_A = 3.0;
-  float bruss_B = 10.0;
-  float bruss_D1 = 5.0;
-  float bruss_D2 = 12.0;
-  float bruss_speed = 0.001;
-
-  // Here we implement the Gray-Scott model, as described here:
-  // http://arxiv.org/abs/patt-sol/9304003
-  //    (the seminal paper by Pearson)
-  // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-  //    (a present university course project, by Greg Turk at Georgia Tech)
-  // http://www.mrob.com/pub/comp/xmorphia/index.html
-  //    (a web exhibit with over 100 videos and 500 images)
-
-  // -- parameters --
-  float D_u = 0.082;
-  float D_v = 0.041;
-
-  // The default is equivalent to options:   -F 0.035 -k 0.064
-  g_k = 0.064;  g_F = 0.035;
-
-  // Other pattern-types to try (pass -F and -k as argument on command line, some require "-density 1" as well):
-  //
-  // -F 0.0118 -k 0.0475              Spiral waves
-  // -F 0.022  -k 0.059               Spots that multiply and keep killing each other off
-  // -F 0.035  -k 0.06                Stripes with branching (fingerprint)
-  // -F 0.04   -k 0.064               For spots that multiply, with stripes mixed in
-  // -F 0.056  -k 0.065               Long stripes ("-density 1" option helps here)
-  // -F 0.062  -k 0.0609 -density 2   "Uskate world", where I found all the Wolfram-class-4 behaviour
-  // -F 0.094  -k 0.059  -density 1   "soap bubbles"
-  // -F 0.094  -k 0.057 -lowback -density 1   Inverse soap bubbles
-
-  float speed = 1.0;
-
-  bool custom_Fk = false;
-
-  for (i = 1; i < argc; i++) {
-    if (0) {
-    } else if (strcmp(argv[i],"-autobright")==0) {
-      // force display()'s auto-brightness feature
-      g_autobright = true;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-calc-module")==0)) {
-      // select a calculation module
-      /* TODO: This will be replaced with two options, one to select the equations (Gray-Scott vs. Brusselator)
-         and another to select a type of calculation (OpenMP, OpenCL with Images, etc.) */
-      i++; g_module = atoi(argv[i]);
-    } else if (strcmp(argv[i],"-color")==0) {
-      // do output in wonderful technicolor
-      g_color = 1;
-      g_oldcolor = 17;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-density")==0)) {
-      // set density of spots for initial pattern
-      i++; g_density = atoi(argv[i]);
-    } else if ((i+1<argc) && 
-         ((strcmp(argv[i],"-F")==0) || (strcmp(argv[i],"-f")==0)) ) {
-      // set parameter F
-      i++; g_F = atof(argv[i]);
-      custom_Fk = true;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-height")==0)) {
-      // set height
-      i++; g_height = atoi(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-k")==0)) {
-      // set parameter k
-      i++; g_k = atof(argv[i]);
-      custom_Fk = true;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-oldcolor")==0)) {
-      // select one of the old PDE4 color schemes
-      g_color = 1; i++; g_oldcolor = atoi(argv[i]);
-    } else if (strcmp(argv[i],"-paramspace")==0) {
-      // do a parameter space plot, like in the Pearson paper
-      g_paramspace = true;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-pastel")==0)) {
-      // select a pastel mode
-      i++; g_pastel_mode = atoi(argv[i]);
-    } else if (strcmp(argv[i],"-ramprects")==0) {
-      // use ramped rectangles in the initial pattern
-      g_ramprects = 1;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-load")==0)) {
-      // load a pattern
-      i++; if (num_loads < MAX_LOADS) {
-        load_opts[num_loads++] = argv[i];
-      }
-    } else if (strcmp(argv[i],"-lowback")==0) {
-      // use "low" background value in init
-      g_lowback = 1;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-scale")==0)) {
-      // set scale
-      i++; g_scale = atof(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-size")==0)) {
-      // set both width and height
-      i++; g_height = g_width = atoi(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-threads")==0)) {
-      // specify number of threads to use
-      i++; g_threads = atoi(argv[i]);
-    } else if (strcmp(argv[i],"-video")==0) {
-      // Create a video media file
-      g_video = true;
-    } else if ((i+1<argc) && (strcmp(argv[i],"-width")==0)) {
-      // set width
-      i++; g_width = atoi(argv[i]);
-    } else if (strcmp(argv[i],"-wrap")==0) {
-      // patterns wrap around ("torus", also called "continuous boundary
-      // condition")
-      g_wrap = true;
-    } else {
-      fprintf(stderr, "Unrecognized argument, or parameter needed: '%s'\n", argv[i]);
-      exit(-1);
-    }
-  }
-
-  /* Set per-system parameters (there will eventually be a routine in each module that returns these
-     model-specific values) */
-  switch(g_module) {
-    default:
-    case READY_MODULE_GS_SCALAR:
-    case READY_MODULE_GS_HWIV:
-      uv_range = 1.0;
-      g_rd_name = "Gray-Scott";
-      break;
-
-    case READY_MODULE_BRUSSELATOR:
-      uv_range = 10.0;
-      g_rd_name = "Brusselator";
-      break;
-  }
-
-#define VECSIZE 4
-
-  if ((g_scale == 1.0) && (custom_Fk)) {
-    g_scale = 2.0;
-  }
-  D_u = D_u * g_scale;
-  D_v = D_v * g_scale;
-  speed = speed / g_scale;
-
-  if (g_width % VECSIZE) {
-    g_width = ((g_width/VECSIZE)+1)*VECSIZE;
-  }
-  int full_width, wid_x;
-
-  switch(g_module) {
-    case READY_MODULE_GS_HWIV:
-      wid_x = VECSIZE;
-      full_width = g_width + 2*wid_x;
-      break;
-
-    default:
-      wid_x = 0;
-      full_width = g_width;
-      break;
-  }
-
-  /* If user has selected a multi-threaded module, we need to check the threading library to
-     see if we have thread synchronization capability */
-  if (g_module == READY_MODULE_GS_HWIV) {
-    /* We cannot use multiple threads unless DiceK supports thread synchronization */
-    if (!(DICEK_SUPPORTS_BLOCKING)) {
-      g_threads = 1;
-    }
-    printf("DiceK reported %d threads.\n", (int) g_hw_threads);
-    printf("Will use %d threads.\n", g_threads);
-  }
-
-  // ----------------
-  
-  // these arrays store the chemical concentrations:
-  float *u = 0;
-  float *v = 0;
-  // these arrays store the rate of change of those chemicals:
-  float *du = 0;
-  float *dv = 0;
-
-  float *red = 0;
-  float *green = 0;
-  float *blue = 0;
-
-  u = allocate(full_width, g_height, "U array", true);
-  v = allocate(full_width, g_height, "V array", true);
-  du = allocate(full_width, g_height, "D_u array", true);
-  dv = allocate(full_width, g_height, "D_v array", true);
-  red = allocate(full_width, g_height, "red array", false);
-  green = allocate(full_width, g_height, "green array", false);
-  blue = allocate(full_width, g_height, "blue array", false);
-
-  // put the initial conditions into each cell
-  switch(g_module) {
-    default:
-    case READY_MODULE_GS_SCALAR:
-      init(u,v, g_width, g_height, g_density, g_lowback, g_ramprects);
-      break;
-
-    case READY_MODULE_GS_HWIV:
-      init(u,v, full_width, g_height, g_density, g_lowback, g_ramprects);
-      break;
-
-    case READY_MODULE_BRUSSELATOR:
-      bruss_init(u, v, g_width, g_height, bruss_A, bruss_B);
-      break;
-  }
-
-  if (g_video) {
-    frames_per_display = 500;
-  }
-
-  /* Pass any non-changing variables to the compute module */
-  switch(g_module) {
-    default:
-    case READY_MODULE_GS_SCALAR:
-      gs_scl_compute_setup(g_width, g_height, g_wrap, g_paramspace);
-      break;
-
-    case READY_MODULE_BRUSSELATOR:
-      bruss_compute_setup(g_width, g_height, g_wrap, g_paramspace);
-      break;
-
-    case READY_MODULE_GS_HWIV:
-      gs_hwi_compute_setup(g_width, g_height, g_wrap, g_paramspace);
-      break;
-    }
-
-  double iteration = 0;
-  double fps_avg = 0.0; // decaying average of fps
-  double Mcgs;
-  while(true) 
-  {
-    struct timeval tod_record;
-    double tod_before, tod_after, tod_elapsed;
-    double fps = 0.0;     // frames per second
-    long its_done;
-
-    gettimeofday(&tod_record, 0);
-    tod_before = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-    // compute:
-    its_done = 0;
-    switch(g_module) {
-      default:
-      case READY_MODULE_GS_SCALAR:
-        for(i=0; i<frames_per_display; i++) {
-          compute_gs_scalar(u, v, du, dv, D_u, D_v, g_F, g_k, speed);
-          its_done++;
-        }
-        break;
-
-      case READY_MODULE_GS_HWIV:
-        compute_dispatch(u, v, du, dv, D_u, D_v, g_F, g_k, speed,
-          frames_per_display, g_threads);
-        its_done = frames_per_display;
-        break;
-
-      case READY_MODULE_BRUSSELATOR:
-        for(i=0; i<frames_per_display; i++) {
-          compute_bruss(u, v, du, dv, bruss_A, bruss_B, bruss_D1, bruss_D2, bruss_speed);
-          its_done++;
-        }
-        break;
-    }
-    iteration += ((double)its_done);
-
-    if (g_color) {
-      colorize(u, v, du, uv_range, red, green, blue, full_width, g_height, g_oldcolor, g_pastel_mode);
-    }
-
-    gettimeofday(&tod_record, 0);
-    tod_after = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-    tod_elapsed = tod_after - tod_before;
-    fps = ((double)its_done) / tod_elapsed;
-    // We display an exponential moving average of the fps measurement
-    fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-    Mcgs = fps_avg * ((double) g_width) * ((double) g_height) / 1.0e6;
-
-    char msg[1000];
-    sprintf(msg, "%s - %0.2f fps (%.2f Mcgs)", g_rd_name, fps_avg, Mcgs);
-
-    // display:
-    {
-      int chose_quit;
-      if (g_color) {
-        chose_quit = display(g_width, wid_x, g_height, red, green, blue,
-          iteration, g_scale, g_autobright, 1.0f, 2, 10, msg, g_video);
-      } else {
-        chose_quit = display(g_width, wid_x, g_height, u, u, u, 
-          iteration, g_scale, g_autobright, uv_range*1.25, 2, 10, msg, g_video);
-      }
-
-      if (chose_quit) // did user ask to quit?
-        break;
-    }
-
-    /* Dynamically adjust frame rate for a balance between responsiveness and
-       efficiency */
-    if(!(g_video)) {
-      if (tod_elapsed < 0.2) {
-        frames_per_display *= 2;
-      } else if (tod_elapsed > 0.4) {
-        frames_per_display /= 2;
-        if (frames_per_display < 10) {
-          frames_per_display = 10;
-        }
-      }
-    }
-  }
-}
-
-/* pearson_bkg fills everything with the trivial state (U=1, V=0) combined
-   with random noise of magnitude 0.01, and it does this while keeping all
-   U and V values between 0 and 1. */
-void pearson_bkg(float *u, float *v, int width, int height)
-{
-  int i, j;
-  for(i=0; i<height; i++) {
-    for(j=0; j<width; j++) {
-      u[i*width+j] = ut_frand(0.99, 1.0);
-      v[i*width+j] = ut_frand(0.0, 0.01);
-    }
-  }
-}
-
-/* pearson_block creates a block filled with a given U and V value combined
-   with superimposed noise of amplitude 0.01 */
-void pearson_block(float *u, float *v, int width, int height,
-                 int vpos, int hpos, int h, int w, float U, float V)
-{
-  int i, j;
-
-  for(i=vpos; i<vpos+h; i++) {
-    if ((i >= 0) && (i < height)) {
-      for(j=hpos; j<hpos+w; j++) {
-        if ((j >= 0) && (j < width)) {
-          u[i*width+j] = ut_frand(U-0.005, U+0.005);
-          v[i*width+j] = ut_frand(V-0.005, V+0.005);
-        }
-      }
-    }
-  }
-}
-
-/* ramp_block creates a starting pattern for B-Z spirals and continuous
-   propagating wave fronts (pattern type xi in my paper). */
-void ramp_block(float *u, float *v, int width, int height,
-                int vpos, int hpos, int h, int w, int fliph)
-{
-  int i, j;
-  float U, V;
-
-  for(i=vpos; i<vpos+h; i++) {
-    if ((i >= 0) && (i < height)) {
-      for(j=hpos; j<hpos+w; j++) {
-        if ((j >= 0) && (j < width)) {
-          U = ((float) (j-hpos)) / ((float) w);
-          if (fliph) {
-            U = 1.0 - U;
-          }
-          V = U;
-
-          if (U < 0.1) {
-            U = U / 0.1;
-          } else {
-            U = (1.0 - U) / 0.9;
-          }
-          U = 1.0 - sin((1.0 - U) * 1.5708);
-          U = 1.0 - (U * 0.95); // formerly 0.85
-          
-          if (V < 0.08) { // formerly 0.05
-            V = V / 0.08; // formerly 0.05
-          } else if (V < 0.5) {
-            V = (0.5 - V) / 0.42; // formerly 0.45
-          } else {
-            V = 0;
-          }
-          V = 1.0 - sin((1.0 - V) * 1.5708);
-          V = V * 0.4;
-
-          u[i*width+j] = ut_frand(U-0.005, U+0.005);
-          v[i*width+j] = ut_frand(V-0.005, V+0.005);
-        }
-      }
-    }
-  }
-}
-
-/* Given an F and k, returns the U and V values for the homogeneous state at
- that F and k. Tries to return the secondary (blue) state, but if that
- doesn't exist it returns the trivial (red) state.
-   The formula is derived from Muratov and Osipov 2000 formula 2.12 (with A
- defined by 2.10, and the other variables defined as in 2.3, 2.4 and 2.5 (all
- of this is on pages 8-9 of "Muratov 2000 Spike.pdf").
-   There is a more obvious formula (expressed in terms of F and k) in
- "Leppanen 2004 Computational.pdf" page 40 (his equation 3.22).
-   To get the values for the center F and k settings, use (g_F_CTR, g_k_CTR) */
-void homogen_uv(float F, float k, float * hU, float * hV)
-{
-  float sqrt_F, A;
-  float U, V;
-
-  sqrt_F = sqrt(F);
-
-  if (k < (sqrt_F - 2.0 * F) / 2.0) {
-    A = sqrt_F / (F + k);
-    U = (A - sqrt(A*A - 4.0)) / (2.0 * A);
-    U = max(0.0, min(1.0, U));
-    V = sqrt_F * (A + sqrt(A*A - 4.0)) / 2.0;
-    V = max(0.0, min(1.0, V));
-  } else {
-    U = 1.0;
-    V = 0.0;
-  }
-  *hU = U;
-  *hV = V;
-}
-
-/* i5_bkg is the background for my init routines. It fills the space either
-   with the trivial stable state (V=0, U=1), or when the other stable "blue"
-   state exists it uses that state. This makes for much more interesting
-   initial patterns for the areas near the various bifurcation lines and for
-   Uskate world. */
-void i5_bkg(float *u, float *v, int width, int height, int which)
-{
-  int i, j;
-
-  for(i=0; i<height; i++) {
-    for(j=0; j<width; j++) {
-      float U, V;
-
-      if (which == 0) {
-        U = 1.0; V = 0.0;
-      } else {
-        homogen_uv(g_F, g_k, &U, &V);
-      }
-      // printf("F %g  k %g  A %g  U %g  V %g\n", F, k, A, U, V);
-      u[i*width+j] = U;
-      v[i*width+j] = V;
-    }
-  }
-}
-
-/*
- Execute one of the "-load" command-line options. This requires parsing out an X and Y coordinate, an option
-orientation, and a filename or pathname. It calls pattern_load to actually read the pattern into the grid.
- */
-void load_option(float * u, float * v, int width, int height, const char * option)
-{
-  int x;
-  int y;
-  int orient = 0;
-  char * p1;
-
-  p1 = (char *) option;
-  x = atol(p1);
-  while(digit_p(*p1)) { p1++; }
-
-  /* Now we expect to see "," */
-  if(!(*p1 == ',')) {
-    /* We got to the end of the string, or the X coordinate was followed by something other than a comma */
-    fprintf(stdout, "invalid load format: '%s'\n", option);
-    exit(-1);
-  }
-  p1++; // skip the ','
-  y = atol(p1);
-  while(digit_p(*p1)) { p1++; }
-
-  /* Now we can see "o" or ":" */
-  if (*p1 == 'o') {
-    /* get orientation */
-    p1++; // skip the 'o'
-    orient = atoi(p1);
-    while(digit_p(*p1)) { p1++; }
-  }
-  if(!(*p1 == ':')) {
-    /* We got to the end of the string, or the X and Y were not followed by ":" */
-    fprintf(stdout, "invalid load format: '%s'\n", option);
-    exit(-1);
-  }
-  p1++; // skip the ':'
-  if (*p1) {
-    pattern_load(u, v, width, height, p1, x, y, orient);
-  }
-}
-
-bool digit_p(char c)
-{
-  return((c == '-') || ((c >= '0') && (c <= '9')));
-}
-
-
-#define MIN_CLIPSIZE 4
-#define MAX_CLIPSIZE 256
-
-#if (defined(__APPLE__) || defined(__linux__))
-# define FILE_LENGTH(name, result) \
-    struct stat _FILESTAT; int _STAT_result; \
-    _FILESTAT.st_size = 0; \
-    _STAT_result = stat(name, &_FILESTAT); \
-    *(result) = (_STAT_result < 0) ? 0 : _FILESTAT.st_size;
-#else
-# ifdef _WIN32
-// Based on: vcpptips.wordpress.com/tag/get-the-size-of-a-file-without-opening-it/
-// TODO: This needs to be tested
-#   define FILE_LENGTH(name, result) \
-    ULONGLONG _LEN_res_tmp; \
-    WIN32_FIND_DATA _LEN_fff_dat = { 0 }; \
-    HANDLE _LEN_fff_hdl = FindFirstFile(name, &_LEN_fff_dat); \
-    if (_LEN_fff_hdl != INVALID_HANDLE_VALUE) { \
-      FindClose(_LEN_fff_hdl); \
-      _LEN_res_tmp = (_LEN_fff_dat.nFileSizeHigh) << (sizeof(_LEN_fff_dat.nFileSizeHigh)*8) | (_LEN_fff_dat.nFileSizeLow); \
-      *(result) = _LEN_res_tmp; \
-    } else { *(result) = 0; }
-# else
-// Neither Unix nor Windows
-#   define FILE_LENGTH(name, result) \
-    sprintf(stdout, "Getting length of file is not yet supported in this environment.\n"); exit(-1);
-# endif
-#endif
-
-/*
-   Load a PDE4 pattern file with the given filename (pathname) into the U and V arrays, at the given X and Y location.
-
-                        6   7          Starting from orientation 0: orient. 5 is rotated 90 degrees clockwise;
-  Orientation             ^            orient. 3 is rotated 180 degrees, and orient. 6 is rotated another 90 degrees
-   diagram:               |            clockwise.
-                   2      |      0       Orientation 1 is a mirror image of orientation 0, flipped around a horizontal
-                    <-----o----->      axis. Compared to orientation 1, orient. 4 is rotated 90 degrees clockwise;
-                   3      |      1     orient. 2 is rotated 180 degrees, and orient. 7 is rotated another 90 degrees
-                          |            clockwise.
-                          v
-                        4   5
-
-You'll notice an odd scaling factor of "0.63324". Most of the patterns in my collection were created in my PDE4 program
-using what I call the "standard model" parameters. PDE4 converts intervals of time (frames) and space (pixels) into the
-dimensionless time and space units in the actual Pearson equations, and is adjusted to use a grid that has about sqrt(2)
-finer resolution than the grid Pearson used in his 1993 paper.
-  Here are the relevant variables and the u formula as calculated by both programs:
-
-  PDE4:                                                            HWIV:
-    g_SPATIAL_REZ = 0.007                                            g_scale = 2.0
-    dxy = (1/143)^2 = 20449                                          speed = 0.5
-    delta_t = 0.05                                                   D_u = 0.164
-    D_u = 2.0e-5                                                     D_v = 0.082
-    D_v = 1.0e-5
-    u -> u + delta_t * (D_u * nabla_u/dxy^2 - u v^2 + F(1-u))        u -> u + speed * (D_u * nabla_u - u v^2 + F(1-u))
-       = u + 0.5 * (2e-5 * dxy * nabla_u - u v^2 + F(1-u))              = u + 0.5 * (0.164 * nabla_u - u v^2 + F(1-u))
-       = u + 0.5 * (0.409 * nabla_u - u v^2 + F(1-u))
-    (v equation is similar)                                          (v equation is similar)
-
-After all the constants are combined, the result is that HWIV differs from PDE4 only in using a different effective D_u
-value, namely 0.164 as compared to 0.409. This causes a difference in scale of all simulated patterns, the scale
-difference is sqrt(0.164/0.409) = 0.63324.
-
-*/
-
-void pattern_load(float * u, float * v, int width, int height,
-  const char * pattern_filename, int x, int y, int orient)
-{
-  int i, j;
-  FILE * patfile;
-  long fsize, csz;
-  double data[2];
-  char * test_endian;
-  int big_endian;
-  float scale = 0.63324 * sqrt(g_scale / 2.0);
-
-  data[0] = 1.0f;
-  test_endian = (char *) data;
-  // printf("endian test bytes: %02X .. %02X\n", test_endian[0], test_endian[7]);
-  if (test_endian[0] == 0x3F) {
-    // The first byte in memory is part of the exponent. This means we're on a big-endian machine.
-    big_endian = 1;
-  } else {
-    big_endian = 0;
-  }
-
-  FILE_LENGTH(pattern_filename, &fsize);
-
-  if (fsize <= 0) {
-    fprintf(stderr, "Could not find pattern file '%s'\n", pattern_filename);
-    exit(-1);
-  }
-
-  /* PDE4 pattern files are a square grid of pixels, each pixel is a U followed by a V, all values are IEEE
-  /* double-precision floating-point. The size of the square is a power of 2 or a power of 2 times 1.5, ranging from 4
-  /* to 256. */
-  csz = MIN_CLIPSIZE;
-  while((csz < MAX_CLIPSIZE) && (csz * csz * 2L * sizeof(double) < fsize)) {
-    csz = next_patsize(csz);
-  }
-
-  if (csz * csz * 2L * sizeof(double) == fsize) {
-    int i2, j2, i3, j3, prev_i2, prev_j2;
-    double u_avg, v_avg, perim_count;
-    int min_i = height; int max_i = 0;
-    int min_j = width; int max_j = 0;
-    prev_i2 = prev_j2 = -1;
-
-    // Size matched exactly
-    // printf("Clip file is %ld bytes long, size %ldx%ld.\n", fsize, csz, csz);
-    printf("Placing pattern %s at position (%d,%d) orientation %d\n", pattern_filename, x, y, orient);
-    patfile = fopen(pattern_filename, "r");
-
-    u_avg = v_avg = perim_count = 0.0;
-    for(i = 0; i < csz; i++) { // i is row number, Y dimension
-      // i2 = (i - csz/2L)*63L/100L;
-      i2 = i - csz/2L;
-      i2 = (int) (scale * ((float) i2));
-      for(j = 0; j < csz; j++) { // j is column number, X dimension
-        float u_val, v_val;
-        // j2 = (j - csz/2L)*63L/100L;
-        j2 = j - csz/2L;
-        j2 = (int) (scale * ((float) j2));
-
-        // Copy to i3, j3 because we're going to change the values
-        i3 = i2; j3 = j2;
-
-        // Now apply orientation
-        if (orient & 1) { i3 = - i3; }
-        if (orient & 2) { j3 = - j3; }
-        if (orient & 4) { int t = i3; i3 = j3; j3 = t; }
-
-        i3 = y + i3;
-        j3 = (width-1) - (x + j3);
-
-        fread((void *) data, sizeof(double), 2, patfile);
-        if (big_endian) {
-          byteswap_2_double(data);
-        }
-
-        /* Add this point to the perimeter average */
-        if ((i == 0) || (i == csz-1) || (j == 0) || (j == csz-1)) {
-          u_avg += data[0]; v_avg += data[1]; perim_count += 1.0;
-        }
-
-        /* Because we are ADDING the pattern to the existing grid values, rather than simply replacing grid values,
-           we need to modify each pixel at most once. */
-        if ((i2 != prev_i2) && (j2 != prev_j2)) {
-          /* Next we clip to the actual grid dimensions */
-          if ((i3 >= 0) && (i3 < height) && (j3 >= 0) && (j3 < width)) {
-            /* Keep track of the min and max values of both coordinates (used below) */
-            if (i3 < min_i) { min_i = i3; } if (i3 > max_i) { max_i = i3; }
-            if (j3 < min_j) { min_j = j3; } if (j3 > max_j) { max_j = j3; }
-            u_val = (float) data[0]; v_val = (float) data[1];
-            u[i3 * width + j3] += u_val;
-            v[i3 * width + j3] += v_val;
-          }
-        }
-        prev_j2 = j2;
-      }
-      prev_i2 = i2;
-    }
-    fclose(patfile);
-
-    /* Now we subtract the average perimeter value from all the pixels that were adjusted. This allows the user to place
-       two patterns very close to one another, even if the "bounding rectangles" would normally cause part of the first
-       pattern to get wiped out by the second.
-         Note that the global limit to the range [0,1] is enforced in the main init() routine after all patterns are
-       loaded. */
-    u_avg = u_avg / perim_count;
-    v_avg = v_avg / perim_count;
-    for(i=min_i; i<=max_i; i++) {
-      for(j=min_j; j<=max_j; j++) {
-        u[i * width + j] -= u_avg;
-        v[i * width + j] -= v_avg;
-      }
-    }
-  } else {
-    printf("Error: %s size %ld is not canonical.\n", pattern_filename, fsize);
-  }
-}
-
-int next_patsize(int former)
-{
-  int newval;
-
-  /* NOTE: We assume 2's complement notation, in which case (x & (x-1)) is zero iff x is a power of 2 */
-  if ((former & (former-1)) == 0) {
-    // it was a power of 2
-    newval = (former * 3) / 2;
-  } else {
-    newval = (former * 4) / 3;
-  }
-  if (newval > MAX_CLIPSIZE) {
-    newval = MAX_CLIPSIZE;
-  }
-  return(newval);
-}
-
-void byteswap_2_double(double * array)
-{
-  char * bytes;
-  char t;
-  bytes = (char *) array;
-  int i;
-
-  /* We have two doubles to swap */
-  for(i=0; i<2; i++) {
-    t = bytes[8*i+0]; bytes[8*i+0] = bytes[8*i+7]; bytes[8*i+7] = t;
-    t = bytes[8*i+1]; bytes[8*i+1] = bytes[8*i+6]; bytes[8*i+6] = t;
-    t = bytes[8*i+2]; bytes[8*i+2] = bytes[8*i+5]; bytes[8*i+5] = t;
-    t = bytes[8*i+3]; bytes[8*i+3] = bytes[8*i+4]; bytes[8*i+4] = t;
-  }
-}
-
-void init(float *u, float *v, int width, int height, int density, int lowback, int BZrects)
-{
-  int nsp, i;
-  int base, var;
-
-  srand((unsigned int)time(NULL));
-
-  if (density <= 0) {
-    nsp = 0;
-  } else {
-    if (density == 3) {
-      base = (height * width) / 512;
-      var = 2;
-    } else {
-      base = density * 20;
-      var = density * (height * width) / 1000;
-    }
-
-    nsp = base + (rand() % var);
-    printf("Adding %d random rectangles\n", nsp);
-  }
-
-  i5_bkg(u, v, width, height, lowback ? 0 : 1);
-
-  for(i=0; i<nsp; i++) {
-    int v1, vs, h1, hs;
-    float U, V;
-
-    vs = 10 + (rand() % 12) * (rand() % 12) / 5;
-    hs = 10 + (rand() % 12) * (rand() % 12) / 5;
-
-    v1 = (rand() % height) - (vs / 2);
-    h1 = (rand() % width) - (hs / 2);
-
-    U = ut_frand(0.0, 1.0);
-    if (BZrects) {
-      V = ut_frand(0.0, 1.0 - U);
-    } else {
-      V = ut_frand(0.0, 1.0);
-    }
-    if (BZrects) {
-      if ((rand() & 0x3) == 0) {
-        int h;
-        h = 40 + (rand() & 31); // formerly 20 + ...
-        ramp_block(u, v, width, height, v1, h1, h, 100, rand() & 1); // formerly 60
-      } else {
-        pearson_block(u, v, width, height, v1, h1, vs, hs, U, V);
-      }
-    } else {
-      pearson_block(u, v, width, height, v1, h1, vs, hs, U, V);
-    }
-  }
-    
-  if ((num_loads <= 0) && (nsp <= 0)) {
-    // old init pattern
-    for(int i = 0; i < height; i++) {
-      for(int j = 0; j < width; j++) {
-        // Make an approximate circle in the middle
-        if(hypot(i-height/2,(j-width/2)/1.5)<=ut_frand(2,5))
-        {
-          u[i*width+j] = ut_frand(0.0,0.1);
-          v[i*width+j] = ut_frand(0.9,1.0);
-        }
-      }
-    }
-  }
-
-  /* Execute any -load arguments */
-  for(int i=0; i<num_loads; i++) {
-    load_option(u, v, width, height, load_opts[i]);
-  }
-
-  /* Finally, enforce the limits for Gray-Scott, which are that U and V must be in the range [0.0,1.0]. */
-  for(int i = 0; i < height; i++) {
-    for(int j = 0; j < width; j++) {
-      u[i*width+j] = max(0.0, min(u[i*width+j], 1.0));
-      v[i*width+j] = max(0.0, min(v[i*width+j], 1.0));
-    }
-  }
-}
-
-/*
-
-The following command-line examples are for Linux or Mac OS shell.
-Change file pathnames to use these examples in Windows.
-
-# Show the Brusselator parameter space:
-./ReaDy -calc-module 3 -paramspace -color
-
-# An example of running from a saved pattern:
-./ReaDy -color -threads 2 -F 0.062 -k 0.0609 -wrap -load '128,128:patterns/interactions/U-curve-15-c'
-
-# Two patterns combined:
-./ReaDy -color -threads 2 -F 0.062 -k 0.0609 -wrap \
-  -load '80,140o5:patterns/halftargets/ht-4-10-3' \
-  -load '35,30:patterns/uskates/std'
-
-*/
diff --git a/Ready_old_CLI/patterns.zip b/Ready_old_CLI/patterns.zip
deleted file mode 100644
index 3ba6aa96a..000000000
Binary files a/Ready_old_CLI/patterns.zip and /dev/null differ
diff --git a/Ready_old_CLI/ready_display/ready_display.cpp b/Ready_old_CLI/ready_display/ready_display.cpp
deleted file mode 100644
index 6abb750d3..000000000
--- a/Ready_old_CLI/ready_display/ready_display.cpp
+++ /dev/null
@@ -1,443 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See ../../README.txt for more details.
-
-*/
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib
-#include <stdio.h>
-
-// local:
-#include "ready_display.h"
-#include "../util.h"
-
-bool display(int g_width, int wid_x, int g_height, float *r, float *g, float *b,
-             double iteration, float model_scale, bool auto_brighten,float manual_brighten,
-             int image_scale,int delay_ms,const char* message, bool write_video)
-{
-  int full_width = g_width + 2*wid_x;
-  static bool need_init = true;
-
-  static IplImage *im,*im2;
-  static int border = 0;
-  static CvFont font;
-  static CvVideoWriter *video;
-  static const CvScalar white = cvScalar(255,255,255);
-
-  const char *title = "ReaDy (ESC to quit)";
-
-  if(need_init)
-  {
-    need_init = false;
-
-    im = cvCreateImage(cvSize(g_width,g_height),IPL_DEPTH_8U,3);
-    cvSet(im,cvScalar(0,0,0));
-    im2 = cvCreateImage(cvSize(g_width*image_scale,g_height*image_scale),IPL_DEPTH_8U,3);
-        
-    cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-        
-    double hScale=0.4;
-    double vScale=0.4;
-    int lineWidth=1;
-    cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-    if(write_video)
-    {
-      video = cvCreateVideoWriter("vid-Gray-Scott.avi",CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im),1);
-      if(video == NULL) {
-        fprintf(stdout, "NULL from cvCreateVideoWriter\n"); exit(-1);
-      }
-    }
-  }
-
-  // convert float arrays to IplImage for OpenCV to display
-  float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-  for(int i=0;i<g_height;i++) {
-    for(int j=wid_x;j<g_width+wid_x;j++) {
-      val = r[i*full_width+j];
-      if(val<minR) minR=val; if(val>maxR) maxR=val;
-
-      val = g[i*full_width+j];
-      if(val<minG) minG=val; if(val>maxG) maxG=val;
-
-      val = b[i*full_width+j];
-      if(val<minB) minB=val; if(val>maxB) maxB=val;
-    }
-  }
-  // I use this to find out the range of the chemicals when adding a new R-D system to ReaDy
-  // printf("R:[%f..%f] G:[%f..%f] B:[%f..%f]\n", minR,maxR, minG,maxG, minB,maxB);
-
-  for(int i=0;i<g_height;i++)
-  {
-    for(int j=0;j<g_width;j++) {
-      int j2 = full_width-j-(wid_x+1);
-      float val;
-      val = r[i*full_width+j2];
-      if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-      else val = (val * 255.0 / manual_brighten);
-      if(val<0) val=0; if(val>255) val=255;
-      ((uchar *)(im->imageData + i*im->widthStep))[j*im->nChannels + 2] = (uchar)val;
-
-      val = g[i*full_width+j2];
-      if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-      else val = (val * 255.0 / manual_brighten);
-      if(val<0) val=0; if(val>255) val=255;
-      ((uchar *)(im->imageData + i*im->widthStep))[j*im->nChannels + 1] = (uchar)val;
-
-      val = b[i*full_width+j2];
-      if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-      else val = (val * 255.0 / manual_brighten);
-      if(val<0) val=0; if(val>255) val=255;
-      ((uchar *)(im->imageData + i*im->widthStep))[j*im->nChannels + 0] = (uchar)val;
-    }
-  }
-
-  cvResize(im,im2);
-
-  char txt[100];
-  const char * fmt;
-
-  if (g_width > 170) {
-    fmt = "Generation=%7g (model's t=%g)";
-  } else if (g_width > 140) {
-    fmt = "Gen.=%7g (model's t=%g)";
-  } else {
-    fmt = "G=%7g t=%g";
-  }
-  sprintf(txt,fmt,iteration, ((double) iteration) / model_scale);
-  cvPutText(im2,txt,cvPoint(5,20),&font,white);
-  cvPutText(im2,message,cvPoint(5,40),&font,white);
-
-  if(write_video)
-    cvWriteFrame(video,im);
-
-  cvShowImage(title,im2);
-    
-  int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-  if(key==27) // did user ask to quit?
-  {
-    if(write_video)
-      cvReleaseVideoWriter(&video);
-    cvDestroyWindow(title);
-    cvReleaseImage(&im);
-    cvReleaseImage(&im2);
-    return true;
-  }
-  return false;
-}
-
-void colorize(float *u, float *v, float *du, float uv_range,
-  float *red, float *green, float *blue, int width, int height,
-  int color_style, int pastel_mode)
-{
-  float r, g, b;
-  // Step by row
-  for(int i = 0; i < height*width; i++) {
-    float uval = u[i] / uv_range;
-    float vval = v[i] / uv_range;
-    float delta_u = du[i] / uv_range;
-
-    color_mapping(uval, vval, delta_u, color_style, &r, &g, &b, pastel_mode);
-    red[i] = r;
-    green[i] = g;
-    blue[i] = b;
-  }
-}
-
-/*
- These are all the colour schemes from Robert Munafo's PDE3 and PDE4 programs, some of which which date back to 1994.
-Included are Pearson's original colours from the 1993 paper, some enhanced versions of Pearson with the derivative shown
-in various ways, some monochrome mappings, and the colours used for all the figures and videos on
-mrob.com/pub/comp/xmorphia.
- Colormap 22 is the one that used to be selected by "-color" (that option now selects colormap 17).
-The rest, including colormap 0, are selected by "-oldcolor N"
- */
-void color_mapping(float u0, float v0, float dU, int pm, float *red, float *green, float *blue, int pastel)
-{
-  float diff, dif2;
-  float t1, t2, t3, t4;
-
-  // Old ratio calculation doesn't work any better than a simple linear
-  // scaling, so I've returned to that - 20090311
-  diff = (dU * 3.0e5) + 0.5; // 3e5
-  diff = minmax(diff, 0.0, 1.0);
-
-  dif2 = (dU * 3.0e7) + 0.5; // 3e7
-  dif2 = minmax(dif2, 0.0, 1.0);
-
-  t1 = diff;
-  t1 = minmax(t1, 0.0, 1.0);
-    
-  t2 = u0;
-  t2 = minmax(t2, 0.0, 1.0);
-
-  t3 = v0;
-  t3 = minmax(t3, 0.0, 1.0);
-
-  switch(pm) {
-    case 0:
-      // Original color scheme from the very very earliest versions of pde1.
-      *red = t2; *green = t3; *blue = t2;
-      break;
-    case 1:
-      // Like colormap 6: Red component shows whether U is increasing,
-      // decreasing or is unchanged as compared to the (previous/next)
-      // generation. (Green and Blue show U and V, like colormap 6).
-      // This system dates back to 19940826, when I had not yet eliminated
-      // the second frame buffer and the derivative calculation was thus
-      // considerably easier.
-      *red = t1; *green = t2; *blue = t3;
-      break;
-    case 2:
-      *red = t2; *green = t1; *blue = t3;
-      break;
-    case 3:
-      // The classic version of this colormap was:
-      // *red = 1.0-t3; *green = 1.0-t3; *blue = 1.0-t2
-      // But this just looks like a washed-out version of colormap 4, so I've
-      // replaced it with this which is a nice blue and green variation
-      *red = t3; *green = 1.0-t2; *blue = 1.0-t1;
-      break;
-    case 4:
-      // Looks way too much like colormap 6
-      *blue = 1.0-t2;
-      t4 = 1.0-t3 - (*blue/2.0); if (t4 < 0) t4 = 0;
-      *red = t4; *green = t4;
-      break;
-    case 5:
-      t2 = 1.0-t2;
-      t4 = 1.0-t3 - (t2/2.0); if (t4 < 0) t4 = 0;
-      if (t4 > t2) {
-        *red = t4-t2; *green = t4-t2; *blue = t2-t4+1.0;
-      } else {
-        *red = t2-t4; *green = t4-t2+1.0; *blue = t4-t2+1.0;
-      }
-      break;
-    case 6:
-      // This is the color mode I settled with early on; as of 2009
-      // I didn't even remember ever having other color modes.
-      t2 = 1.0-t2;
-      t4 = 1.0-t3 - (t2/2); if (t4 < 0) t4 = 0;
-      if (t4 > t2) {
-        *red = t4-t2; *green = t4-t2; *blue = t2-t4+1.0;
-      } else {
-        *red = t2-t4; *green = 0; *blue = t4-t2+1.0;
-      }
-      break;
-    case 7:
-      // I added this color mode on 20090116 because I wanted something
-      // pretty for the vision board, and for some variety -- all the
-      // other color modes had yellow in the large area to the right!
-      t2 = 1.0-t2;
-      t4 = 1.0-t3 - (t2/2); if (t4 < 0) t4 = 0;
-      if (t4 > t2) {
-        *red = 1.0-(t4-t2); *green = 1.0-(t4-t2); *blue = (t4-t2)/2.0;
-      } else {
-        *red = 1.0-(t2-t4); *green = 1.0; *blue = (t2-t4)/2.0;
-      }
-      break;
-    case 8:
-      // This color mode is meant to reflect Pearson's description of the
-      // two large plain states as "blue state" and "red state" (see the
-      // Pearson paper, 9304003.pdf, page 9, caption for figure 3),
-      // without actually using the HSV color space (for that use color
-      // mode 9)
-      *red = (1.0-t3)*0.75; *green = 0; *blue = 1.0-t2;
-      break;
-    default:
-    case 9:
-      // This is a reasonable facsimile of Pearson's coloring scheme. He
-      // shows only the U value, expressing it as a hue in the HSV color
-      // space.
-      t2 = 1.0 - t2;
-      t2 = t2 * 0.7 + 0.061;
-      t3 = 1.0;
-      t4 = 1.0;
-      if (t2 < 0.02136) {
-        t3 = 0.7324;
-      } else if (t2 < 0.3052) {
-        t3 = 0.7324 + (t2-0.2136) * 2.5;
-      }
-      if (t2 < 0.1526) {
-        t4 = 0.4883;
-      } else if (t2 < 0.2441) {
-        t4 = 0.4883 + (t2-0.1526) * 5.0;
-      }
-      go_hsv2rgb(t2, t3, t4, red, green, blue);
-      break;
-
-    case 10: // `a'
-      // Colorscheme created in 200901 for PDE3.
-      // Like colormap 6: Saturation shows whether U is increasing, decreasing
-      // or is unchanged as compared to the (previous/next) generation. Hue
-      // shows U; value constant at 1.0.
-      t1 = 0.5 + t1/2.0;
-      go_hsv2rgb(0.8 * (1.0 - t2), t1, 1.0, red, green, blue);
-      break;
-
-    case 11: // `b'
-      // First new colorscheme devised on 20090311. Rick likes this one.
-      t1 = 0.5 + t1/2.0;
-      go_hsv2rgb(0.8 * ((float) (1.0 - t2)), (float) t1, t2, red, green, blue);
-      break;
-
-    case 12: // `c'
-      // This one compresses into MP4 better because it avoids bright red
-      t1 = 0.25 + t1 * 0.75;
-      go_hsv2rgb(0.8 * ((float) (t2)), t1, t2, red, green, blue);
-      break;
-
-    case 13: // `d'
-      // Modification of scheme 10, uses a greater range of the color space
-      t1 = 0.5 + t1/2.0;
-      t4 = 4 * (1.0 - t2) / 5;
-      t4 = t4 + 0.75; t4 = t4 - floor(t4); // olive, yellow, red, pink, lavender
-      go_hsv2rgb(t4, t1, t2, red, green, blue);
-      break;
-
-    case 14: // `e'
-      // Modification of scheme 13, limiting use of saturated colors to force
-      // better quality out of H264/mp4
-      t1 = 0.25 + t1 * 0.4;
-      t4 = (1.0 - t2)*1.8; // more than one full turn around the hue wheel
-      t4 = t4 + 0.5798; t4 = t4 - floor(t4); // deep purple, deep blue, green, yellow, red, pink, blue, aqua
-      go_hsv2rgb(t4, t1, t2, red, green, blue);
-      break;
-
-    case 15: // `f'
-      // Another one that compresses well, with orange in the "desert"
-      t1 = 0.25 + t1 * 0.4;
-      t4 = 6 * (1.0 - t2) / 5; // more than one full turn around the hue wheel
-      t4 = t4 + 1.0681; t4 = t4 - floor(t4); // dark brown, deep purple, deep blue, green, yellow, orange
-      go_hsv2rgb(t4, t1, t2, red, green, blue);
-      break;
-
-    case 16: // `g'
-      // Modification of scheme 13, limiting use of saturated colors to force
-      // better quality out of H264/mp4
-      t1 = 0.25 + t1 * 0.4;
-      t4 = (1.0 - t2) * 1.4;
-      t4 = t4 + 0.6409; t4 = t4 - floor(t4); // deep purple, deep blue, teal, green, yellow, red, pink, blue
-      go_hsv2rgb(t4, t1, t2, red, green, blue);
-      break;
-
-    case 17: // colormode 17, originally letter `h'
-      // Back to colorscheme 11, but with subdued brightness and saturation
-      t1 = 0.25 + t1 * 0.6;
-      t4 = (1.0 - t2) * 1.0;
-      t4 = t4 + 1.9836; t4 = t4 - floor(t4); // deep purple, deep blue, teal, green yellow, red
-      go_hsv2rgb(t4, t1, 0.3815 + t2/2.0, red, green, blue);
-      break;
-
-    case 18: // `i'
-      // Created on 20090404 specifically for monochrome figures in uskate
-      // world. The first version of this was:
-      //   t2 = minmax((t2 * 2) - 0.6104, 0.0, 1.0)
-      // Then I did 5 shades of gray on thresholds 0.40, 0.418, 0.422, 0.55.
-      // Finally, on 20090405 I optimized it for the B&W illustrations in the
-      // paper I'm writing and came up with this combination of 3 linear
-      // sclings.
-      diff = (u0 - 0.4) * 16.28;
-      if (diff > 1.22) {
-        diff = (diff / 10.0) + 0.5;
-      } else if (diff > 0.4578) {
-        diff = (diff / 5.0) + 0.3662;
-      }
-      t2 = minmax(diff, 0.0, 1.0);
-      go_hsv2rgb(0, 0, t2, red, green, blue);
-      break;
-
-    case 19: // `j'
-      // Created on 20090404 specifically for monochrome figures in uskate world
-      go_hsv2rgb(0, 0, (1.0 - t1), red, green, blue);
-      break;
-
-    case 20:
-      // Created on 20090507, higher-contrast derivative
-      t4 = 1.0 * dif2;
-      t4 = minmax(t4, 0, 1.0);
-      go_hsv2rgb(0, 0, (1.0 - t4), red, green, blue);
-      break;
-
-    case 21:
-      // 20101122: Monochrome without contrast adjustment
-      go_hsv2rgb(0, 0, u0, red, green, blue);
-      break;
-
-    case 22:
-      // Created by Robert Munafo for Tim Hutton's reaction-diffusion project
-      // Something simple to start (-:
-      // different colour schemes result if you reorder these, or replace
-      // "x" with "1.0f-x" for any of the 3 variables
-      diff = dU * 1000.0f + 0.5f;
-      diff = minmax(diff, 0.0, 1.0);
-      *red = diff; // increasing U will look pink
-      *green = 1.0-u0; *blue = 1.0-v0;
-
-  }
-
-  // Pastel mode transformation simply reduces all three components by half --
-  // either as additive primaries (making it darker) or as subtractive
-  // primaries (making it lighter)
-  if (pastel == 1) {
-    *red = 0.5 + *red/2.0;
-    *green = 0.5 + *green/2.0;
-    *blue = 0.5 + *blue/2.0;
-  } else if (pastel == 2) {
-    *red = *red/2.0;
-    *green = *green/2.0;
-    *blue = *blue/2.0;
-  }
-}
-
-void go_hsv2rgb(float h, float s, float v, float *red, float *green, float *blue)
-{
-  float p16 = 65536.0;
-  float i, f, p, q, t, r, g, b;
-  int ii;
-
-  if (s <= 0.0) {
-    // Ignore hue
-    r = v;
-    g = v;
-    b = v;
-  } else {
-    h = h * 6.0;
-    ii = ((int) h);
-    i = (float) ii;
-    f = h - i;
-    p = v*(1.0 - s);
-    q = v*(1.0 - (s*f));
-    t = v*(1.0 - (s*(1.0 - f)));
-    switch(ii) {
-      case 0:
-        r = v; g = t; b = p;
-        break;
-      case 1:
-        r = q; g = v; b = p;
-        break;
-      case 2:
-        r = p; g = v; b = t;
-        break;
-      case 3:
-        r = p; g = q; b = v;
-        break;
-      case 4:
-        r = t; g = p; b = v;
-        break;
-      case 5:
-      default:
-        r = v; g = p; b = q;
-        break;
-    }
-  }
-  *red = r; *green = g; *blue = b;
-}
-
diff --git a/Ready_old_CLI/ready_display/ready_display.h b/Ready_old_CLI/ready_display/ready_display.h
deleted file mode 100644
index aa90aa499..000000000
--- a/Ready_old_CLI/ready_display/ready_display.h
+++ /dev/null
@@ -1,10 +0,0 @@
-bool display(int width, int wid_x, int height, float *r, float *g, float *b,
-             double iteration, float model_scale, bool auto_brighten,float manual_brighten,
-			 int scale,int delay_ms,const char* message, bool write_video);
-
-void colorize(float *u, float *v, float *du, float uv_range,
-             float *red, float *green, float *blue,
-             int width, int height, int color_style, int pastel_mode);
-void color_mapping(float u0, float v0, float dU, int pm, float *red, float *green, float *blue, int pastel);
-void go_hsv2rgb(float h, float s, float v, float *red, float *green, float *blue);
-
diff --git a/Ready_old_CLI/util.cpp b/Ready_old_CLI/util.cpp
deleted file mode 100644
index 02cf47411..000000000
--- a/Ready_old_CLI/util.cpp
+++ /dev/null
@@ -1,18 +0,0 @@
-/*
-
-  util.cpp  from ReaDy module
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
- */
-
-#include <stdlib.h>
-
-// return a random value between lower and upper
-float ut_frand(float lower, float upper)
-{
-  return lower + rand()*(upper-lower)/RAND_MAX;
-}
diff --git a/Ready_old_CLI/util.h b/Ready_old_CLI/util.h
deleted file mode 100644
index 6d7c63cfb..000000000
--- a/Ready_old_CLI/util.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
-
-  util.h  from ReaDy module
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
- */
-
-#ifndef max
-# define max(a,b) (((a) > (b)) ? (a) : (b))
-# define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-#ifndef minmax
-# define minmax(v, lo, hi) max(lo, min(v, hi))
-#endif
-
-float ut_frand(float lower, float upper);
diff --git a/Schlogl/CMakeLists.txt b/Schlogl/CMakeLists.txt
deleted file mode 100644
index 721dca237..000000000
--- a/Schlogl/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(Schlogl)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(Schlogl 
-  schlogl.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(Schlogl ${OpenCV_LIBS} )
diff --git a/Schlogl/schlogl.cpp b/Schlogl/schlogl.cpp
deleted file mode 100644
index d40505bb0..000000000
--- a/Schlogl/schlogl.cpp
+++ /dev/null
@@ -1,120 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y]);
-
-void compute(float a[X][Y],
-             float da[X][Y],
-             float speed);
-
-int main()
-{
-    // Schlogl model:
-    // F. Schlogl, "Chemical reaction models for non-equilibrium phase transitions", Zeitschrift fur Physik, 253, 147, (1972) 
-    // Following:
-    // http://wwwnlds.physik.tu-berlin.de/~hizanidis/talks/front_propagation_noise.pdf
-
-    // -- parameters --
-    float speed = 0.1f;
-    // ----------------
-    
-    // this array stores the chemical concentrations:
-    float a[X][Y];
-    // this array store the rate of change of those chemicals:
-    float da[X][Y];
-
-    // put the initial conditions into each cell
-    init(a);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,da,speed);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Schlogl - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,a,iteration,true,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] = frand(-1.0f,1.0f);
-        }
-    }
-}
-
-void compute(float a[X][Y],
-             float da[X][Y],
-             float speed)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        int iprev = (i + X - 1) % X;
-        int inext = (i + 1) % X;
-
-        for(int j = 0; j < Y; j++) 
-        {
-            int jprev = (j + Y - 1) % Y;
-            int jnext = (j + 1) % Y;
-
-            float aval = a[i][j];
-
-            // compute the Laplacian of a
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-
-            // compute the new rate of change of a
-            da[i][j] = dda + aval*(1-aval)*(1+aval); // a-a^3
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-        }
-    }
-}
diff --git a/Schnakenberg/CMakeLists.txt b/Schnakenberg/CMakeLists.txt
deleted file mode 100644
index b47fca9eb..000000000
--- a/Schnakenberg/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(Schnakenberg)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(Schnakenberg 
-  schnakenberg.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(Schnakenberg ${OpenCV_LIBS} )
diff --git a/Schnakenberg/schnakenberg.cpp b/Schnakenberg/schnakenberg.cpp
deleted file mode 100644
index 555531951..000000000
--- a/Schnakenberg/schnakenberg.cpp
+++ /dev/null
@@ -1,178 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float alpha,float beta,float gamma,float nu,
-             float speed,
-             bool parameter_space);
-
-int main()
-{
-    // J. Schnakenberg, Simple chemical reaction systems with limit cycle behaviour, J. Theor. Biol. 81 (1979) 389–400.
-    // Following:
-    // ftp://ftp.comlab.ox.ac.uk/pub/Documents/techreports/NA-03-16.pdf
-
-    // Not sure this is right though.
-
-    // -- parameters --
-    float alpha=1.0f;
-    float beta=0.9f;
-    float gamma=1.0f;
-    float nu=10.0f;
-
-    float speed = 0.001f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true) 
-    {
-    	start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,da,db,alpha,beta,gamma,nu,speed,false);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Schnakenberg - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,b,b,iteration,true,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-#define PI 3.1415926535
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    float x,y;
-    for(int i = 0; i < X; i++) 
-    {
-        x=float(i)/X;
-        for(int j = 0; j < Y; j++) 
-        {
-            /*y = float(j)/Y;
-            a[i][j] = 0.919145 + 0.0016*cos(2*PI*(x+y));
-            b[i][j] = 0.937903 + 0.0016*cos(2*PI*(x+y));
-            for(int t=1;t<=8;t++)
-            {
-                a[i][j] += 0.01 * cos(2*PI*t*x);
-                b[i][j] += 0.01 * cos(2*PI*t*x);
-            }*/
-            a[i][j] = frand(-1.0f,1.0f);
-            b[i][j] = frand(-1.0f,1.0f);
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float alpha,float beta,float gamma,float nu,
-             float speed,
-             bool parameter_space)
-{
-    const bool toroidal = true;
-
-    int iprev,inext,jprev,jnext;
-
-    // compute change in each cell
-    for(int i = 0; i < X; i++) {
-        if(toroidal) {
-            iprev = (i + X - 1) % X;
-            inext = (i + 1) % X;
-        }
-        else {
-            iprev = max(0,i-1);
-            inext = min(X-1,i+1);
-        }
-
-        for(int j = 0; j < Y; j++) {
-            if(toroidal) {
-                jprev = (j + Y - 1) % Y;
-                jnext = (j + 1) % Y;
-            }
-            else {
-                jprev = max(0,j-1);
-                jnext = min(Y-1,j+1);
-            }
-
-            if(parameter_space)	{
-                /*const float kmin=0.045f,kmax=0.07f,fmin=0.00f,fmax=0.14f;
-                // set f and k for this location (ignore the provided values of f and k)
-                k = kmin + i*(kmax-kmin)/X;
-                f = fmin + j*(fmax-fmin)/Y;*/
-            }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            //da[i][j] = 1000.0f*(0.126779f - aval + aval*aval*bval) + dda / (128.0f*128.0f);
-            //db[i][j] = 1000.0f*(0.792366f - aval*aval*bval) + 10.0f*ddb / (128.0f*128.0f);
-            da[i][j] = dda + gamma*(alpha - aval + aval*aval*bval);
-            db[i][j] = nu * ddb + gamma*(beta - aval*aval*bval);
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-        }
-    }
-}
-
diff --git a/SpeedComparisons/CMakeLists.txt b/SpeedComparisons/CMakeLists.txt
deleted file mode 100644
index c892dac05..000000000
--- a/SpeedComparisons/CMakeLists.txt
+++ /dev/null
@@ -1,23 +0,0 @@
-cmake_minimum_required(VERSION 2.6)
-
-project(SpeedComparisons)
-
-if (NOT CMAKE_BUILD_TYPE)
- message(STATUS "No build type selected, default to Release")
- set(CMAKE_BUILD_TYPE "Release")
-endif()
-
-add_subdirectory(GrayScott)
-add_subdirectory(GrayScott_HWIVector)
-add_subdirectory(GrayScott_double)
-add_subdirectory(GrayScott_OpenCV)
-add_subdirectory(GrayScott_SSE)
-add_subdirectory(GrayScott_SSE_OpenMP)
-add_subdirectory(GrayScott_OpenMP)
-add_subdirectory(GrayScott_OpenCL)
-add_subdirectory(GrayScott_OpenCL_float2)
-add_subdirectory(GrayScott_OpenCL_Local)
-add_subdirectory(GrayScott_OpenCL_2x2)
-add_subdirectory(GrayScott_OpenCL_Image)
-add_subdirectory(GrayScott_OpenCL_Image_2x2)
-
diff --git a/SpeedComparisons/Display/defs.h b/SpeedComparisons/Display/defs.h
deleted file mode 100644
index 6b34e615e..000000000
--- a/SpeedComparisons/Display/defs.h
+++ /dev/null
@@ -1,4 +0,0 @@
-// the size of the world:
-#define X 256
-#define Y 256
-// (must be powers of 2)
diff --git a/SpeedComparisons/Display/display.cpp b/SpeedComparisons/Display/display.cpp
deleted file mode 100644
index fb0fdb0ef..000000000
--- a/SpeedComparisons/Display/display.cpp
+++ /dev/null
@@ -1,145 +0,0 @@
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib
-#include <stdio.h>
-
-// local:
-#include "display.h"
-
-bool display(float r[X][Y],float g[X][Y],float b[X][Y],
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2,*im3;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-        im3 = cvCreateImage(cvSize(X*scale+border*2,Y*scale+border),IPL_DEPTH_8U,3);
-        
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-        
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-        if(write_video)
-        {
-            video = cvCreateVideoWriter(title,CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im3),1);
-            border = 20;
-        }
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<X;i++)
-        {
-            for(int j=0;j<Y;j++)
-            {
-                if(r) {
-                    val = r[i][j];
-                    if(val<minR) minR=val; if(val>maxR) maxR=val;
-                }
-                if(g) {
-                    val = g[i][j];
-                    if(val<minG) minG=val; if(val>maxG) maxG=val;
-                }
-                if(b) {
-                    val = b[i][j];
-                    if(val<minB) minB=val; if(val>maxB) maxB=val;
-                }
-            }
-        }
-    }
-    #pragma omp parallel for
-    for(int i=0;i<X;i++)
-    {
-        for(int j=0;j<Y;j++)
-        {
-            if(r) {
-                float val = r[i][Y-j-1];
-                if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 2] = (uchar)val;
-            }
-            if(g) {
-                float val = g[i][Y-j-1];
-                if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 1] = (uchar)val;
-            }
-            if(b) {
-                float val = b[i][Y-j-1];
-                if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 0] = (uchar)val;
-            }
-        }
-    }
-
-    cvResize(im,im2);
-    cvCopyMakeBorder(im2,im3,cvPoint(border*2,0),IPL_BORDER_CONSTANT);
-
-    char txt[100];
-    if(!write_video)
-    {
-        sprintf(txt,"%d",iteration);
-        cvPutText(im3,txt,cvPoint(20,20),&font,white);
-
-        // DEBUG:
-        sprintf(txt,"%.4f,%.4f,%.4f",r[0][0],g[0][0],b[0][0]);
-        //cvPutText(im3,txt,cvPoint(20,40),&font,white);
-    }
-
-    // DEBUG:
-    if(write_video)
-    {
-        cvPutText(im3,"0.06",cvPoint(5,15),&font,white);
-        cvPutText(im3,"F",cvPoint(5,im2->height/2),&font,white);
-        cvPutText(im3,"0.00",cvPoint(5,im2->height),&font,white);
-        cvPutText(im3,"0.03",cvPoint(border*2-10,im2->height+15),&font,white);
-        cvPutText(im3,"K",cvPoint(border*2+im2->width/2,im2->height+15),&font,white);
-        cvPutText(im3,"0.07",cvPoint(im3->width-35,im2->height+15),&font,white);
-    }
-
-    cvPutText(im3,message,cvPoint(20,40),&font,white);
-
-    if(write_video)
-        cvWriteFrame(video,im3);
-
-    cvShowImage(title,im3);
-    
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        if(write_video)
-            cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
-
diff --git a/SpeedComparisons/Display/display.h b/SpeedComparisons/Display/display.h
deleted file mode 100644
index dcd63ad58..000000000
--- a/SpeedComparisons/Display/display.h
+++ /dev/null
@@ -1,5 +0,0 @@
-#include "defs.h"
-
-bool display(float r[X][Y],float g[X][Y],float b[X][Y],
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
diff --git a/SpeedComparisons/GrayScott/CMakeLists.txt b/SpeedComparisons/GrayScott/CMakeLists.txt
deleted file mode 100644
index a9986d64a..000000000
--- a/SpeedComparisons/GrayScott/CMakeLists.txt
+++ /dev/null
@@ -1,15 +0,0 @@
-project(GrayScott)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(GrayScott 
-  gray_scott.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-  ../Display/defs.h
-)
-
-TARGET_LINK_LIBRARIES(GrayScott ${OpenCV_LIBS} )
diff --git a/SpeedComparisons/GrayScott/gray_scott.cpp b/SpeedComparisons/GrayScott/gray_scott.cpp
deleted file mode 100644
index 52953132a..000000000
--- a/SpeedComparisons/GrayScott/gray_scott.cpp
+++ /dev/null
@@ -1,238 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float r_a,float r_b,float f,float k,
-             float speed,
-             bool parameter_space);
-
-static int g_wrap = 1;
-static bool g_paramspace = 0;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if (strcmp(argv[i],"-paramspace")==0) {
-           // do a parameter space plot, like in the Pearson paper
-           g_paramspace = true;
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            fprintf(stderr, "Unrecognized argument: '%s'\n", argv[i]);
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    // for spots:
-    float k = 0.064f;
-    float f = 0.035f;
-    // for stripes:
-    //float k = 0.06f;
-    //float f = 0.035f;
-    // for long stripes
-    //float k = 0.065f;
-    //float f = 0.056f;
-    // for dots and stripes
-    //float k = 0.064f;
-    //float f = 0.04f;
-    // for spiral waves:
-    //float k = 0.0475f;
-    //float f = 0.0118f;
-    float speed = 1.0f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-
-    const int N_FRAMES_PER_DISPLAY = 200;
-    int iteration = 0;
-    double fps_avg = 0.0; // decaying average of fps
-    while(true) 
-    {
-        struct timeval tod_record;
-        double tod_before, tod_after, tod_elapsed;
-        double fps = 0.0;     // frames per second
-
-        gettimeofday(&tod_record, 0);
-        tod_before = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        // compute:
-        for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-        {
-            compute(a,b,da,db,r_a,r_b,f,k,speed,g_paramspace);
-            iteration++;
-        }
-
-        gettimeofday(&tod_record, 0);
-        tod_after = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        tod_elapsed = tod_after - tod_before;
-        fps = ((double)N_FRAMES_PER_DISPLAY) / tod_elapsed;
-        // We display an exponential moving average of the fps measurement
-        fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-		double Mcgs = fps_avg * ((double) X) * ((double) Y) / 1.0e6;
-
-        char msg[1000];
-        sprintf(msg, "GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-        // display:
-        if(display(a,a,a,iteration,false,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            
-            //if(hypot(i%50-25/*-X/2*/,j%50-25/*-Y/2*/)<=frand(2,5))
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5)) // start with a uniform field with an approximate circle in the middle
-            {
-                a[i][j] = 0.0f;
-                b[i][j] = 1.0f;
-            }
-            else {
-                a[i][j] = 1;
-                b[i][j] = 0;
-            }
-            //float v = frand(0.0f,1.0f);
-            //a[i][j] = v;
-            //b[i][j] = 1.0f-v;
-            //a[i][j] += frand(-0.01f,0.01f);
-            //b[i][j] += frand(-0.01f,0.01f);
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float r_a,float r_b,float f,float k,float speed,
-             bool parameter_space)
-{
-    //const bool toroidal = false;
-
-    //int iprev,inext,jprev,jnext;
-
-    // compute change in each cell
-    for(int i = 0; i < X; i++) {
-        int iprev,inext;
-        if (g_wrap) {
-            iprev = (i + X - 1) % X;
-            inext = (i + 1) % X;
-        } else {
-            iprev = max(0,i-1);
-            inext = min(X-1,i+1);
-        }
-
-        for(int j = 0; j < Y; j++) {
-            int jprev,jnext;
-            if (g_wrap) {
-                jprev = (j + Y - 1) % Y;
-                jnext = (j + 1) % Y;
-            } else {
-                jprev = max(0,j-1);
-                jnext = min(Y-1,j+1);
-            }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            if(parameter_space)	{
-                const float kmin=0.045f,kmax=0.07f,fmin=0.01f,fmax=0.09f;
-                // set f and k for this location (ignore the provided values of f and k)
-                k = kmin + i*(kmax-kmin)/X;
-                f = fmin + j*(fmax-fmin)/Y;
-            }
- 
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = r_a * dda - aval*bval*bval + f*(1-aval);
-            db[i][j] = r_b * ddb + aval*bval*bval - (f+k)*bval;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) 
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] += speed * da[i][j];
-            b[i][j] += speed * db[i][j];
-            // kill denormals by adding a teeny tiny something (http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.20.1348&rank=4)
-            a[i][j] += 1e-10f;
-            b[i][j] += 1e-10f;
-        }
-    }
-}
-
diff --git a/SpeedComparisons/GrayScott_HWIVector/CMakeLists.txt b/SpeedComparisons/GrayScott_HWIVector/CMakeLists.txt
deleted file mode 100644
index e8e3a1ac8..000000000
--- a/SpeedComparisons/GrayScott_HWIVector/CMakeLists.txt
+++ /dev/null
@@ -1,20 +0,0 @@
-project(GrayScott_HWIVector)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-if(MSVC)
-  add_definitions(/arch:SSE2)
-else()
-  add_definitions(-msse2)
-endif()
-
-add_executable(GrayScott_HWIVector 
-  gray_scott_hwivector.cpp 
-  hwi_vector.h
-  dicek.h
-  display_hwiv.cpp
-  display_hwiv.h
-)
-
-TARGET_LINK_LIBRARIES(GrayScott_HWIVector ${OpenCV_LIBS} )
diff --git a/SpeedComparisons/GrayScott_HWIVector/dicek.h b/SpeedComparisons/GrayScott_HWIVector/dicek.h
deleted file mode 100644
index 432e9c73a..000000000
--- a/SpeedComparisons/GrayScott_HWIVector/dicek.h
+++ /dev/null
@@ -1,507 +0,0 @@
-/*
-  dicek.h                                                                                                ^u 120 ^x f
-
-This provides macros for multi-threaded programming. The intention is to allow a single program source to be compiled
-for Windows, Linux or Mac OS, and accomplish multi-threading without having to have a lot of #ifdefs around the
-OS-specific code.
-
-REVISION HISTORY
- 20110929 First version of simple DICEK_SPLIT_MERGE function. This is a blocking, one-time N-way parallel subroutine
-call (no provision for threads to continue through multiple synchronization barriers). Works in DICEK_USE_THREADS and
-DICEK_EMULATE modes (tested with math3000.cxx A094358() routine).
- 20110930 Add a first (extremely speculative and untested) shot at the Windows implementation, currently protected by a
-fall-back #ifdef block at the beginning that checks for Windows and reverts to EMULATE mode, pending testing by a real
-Windows programmer.
- 20111001 Add the thread interlock macros and get them working in EMULATE and POSIX modes.
- 20111009 Initialize mutexes in Win32 version of DICEK_SPLIT_1
- 20111011 Use semaphores instead of mutexes in Win32 version.
-
-*/
-
-/* First we test for each of the known operating system environments */
-
-#if (defined(__linux__) || defined(__APPLE__))
-# ifndef DICEK_USE_POSIX
-#   ifndef DICEK_EMULATE
-#     define DICEK_USE_POSIX
-#   endif
-# endif
-#endif
-
-
-#ifdef _WIN32
-# ifndef DICEK_EMULATE
-#   define DICEK_USE_PROCESS_H
-# endif
-#else
-# define __stdcall
-#endif
-
-
-#ifndef DICEK_USE_PROCESS_H
-# ifndef DICEK_USE_POSIX
-#   ifndef DICEK_EMULATE
-#     define DICEK_EMULATE
-#   endif
-# endif
-#endif
-
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//  #include<stdio.h>                                                                                                //
-//  /*                                                   xx*/                    /*                             */   //
-//  /*                                                     */                    /*                             */   //
-//  int                 main()  {char      a[]     ={     'T',    'h','i',     's',32,     'p','r'        ,'o','g'   //
-//  ,'r','a','m',      32,   'j',   'u',   's',    't'    ,32           ,'d',    'o',   'e'     ,'s'    ,32    ,'i'  //
-//  ,'t'              ,32    ,'t'    ,'h'  ,'e'   ,32,    'h'    ,97,'r','d'   ,32,    'w',97,121,33,   32,    40,   //
-//  68,               'o',   'n',    39,   't'    ,32     ,'y'   ,'o',   117,    32    ,'t'             ,'h'   ,'i'  //
-//  ,'n'               /*     Xy     =a     +3     +n      ++     ;a=     b-     (*     x/z              );     if   //
-//  (Xy-++n<(z+*x))z  =b;a   +b,     z+=     x*/,107 ,    63,63    ,63,41,'\n'   ,00};    puts(a);}      /*.RPM.*/   //
-//                                                                                                                   //
-//        Emulated versions of the DICEK macros. (These also serve as documentation for what each macro does)        //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-#ifdef DICEK_EMULATE
-
-/*
- We cannot do inter-thread communication because we execute each "spawned thread" to completion before starting the next
-one. Therefore, the first time any child thread tries to DICEK_INTERLOCK, the program will deadlock because there is no
-way for the parent to get to the DICEK_CH_BEGIN until the child has exited.
- */
-#define DICEK_SUPPORTS_BLOCKING 0
-
-/*
-DICEK_THREAD_VARS defines variables which need to be included in the parameter block of any subroutine called by
-DICEK_SPLIT_MERGE
-  Since this is the emulated version, there are no master_wkg and child_wkg semaphores.
- */
-#define DICEK_THREAD_VARS \
-  long DICEK_tnum; \
-  void * DICEK_thread; \
-  void * DICEK_return;
-
-/*
- DICEK_INIT_NTHR declares an integer-type variable named "nth" and initializes it to the number of hardware threads
-supported by this system. Place it in your main() or a function from which threads will be launched.
-  If you want your threads to have access to the value of nth, you should do "glob = nth;" right after the
-DICEK_INIT_NTHR(), where "glob" is a suitably scoped global variable.
-
-In emulated mode we return 3 as the number of "hardware threads", partly because 3 is a fairly rare value in real
-computers (the Athlon X3 being about the only one anyone will have) and therefore if you find that DICEK is running 3
-"threads" then you know it probably failed to auto-detect the environment.
-  Also, as of this writing (2011) 3 threads is about the average number of hardware threads across all portable and
-desktop computers that are out there. (Anything by Intel with the "Core i3" or higher brand has 4 threads, and most
-desktop machines have at least 2 cores and 4 threads as well)
-   However, since this is the emulated-mode version of the macros, the "threads" are actually going to just run one
-after the next.
- */
-#define DICEK_INIT_NTHR(nth) int nth = 3;
-
-/*
- Place DICEK_DATA in the variable declarations area of the function containing an DICEK_FORK directive. dtype should be
-a struct which contains the DICEK_THREAD_VARS macro, and as many other fields as you want. nth is the number of threads
-that you will be creating with DICEK_SPLIT_MERGE.
-  It will declare an array aname (dynamically allocated) which will be an array of structs of type dtype, and fill the
-DICEK-specific fields with a thread number and a pointer to the pthreads data for each thread that will be used by
-DICEK_SPLIT_MERGE.
- */
-#define DICEK_DATA(dtype, arrayname, nth) \
-    dtype * arrayname; \
-    arrayname = (dtype *)malloc(nth * sizeof(dtype)); \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      arrayname[_DICEK_i].DICEK_tnum = _DICEK_i; \
-      arrayname[_DICEK_i].DICEK_thread = 0; \
-    }
-
-/*
- DICEK_SPLIT_1 starts a thread with the given index; the array "arrayname" should be allocated by DICEK_DATA and should
-contain an element at the given index. The thread will call a function called funcname, which should be type
-
-  void(* funcname)(void *)
-
-The instance of funcname will be passed the index'th element of arrayname.
-  After the thread has exited, execution proceeds with the instruction following DICEK_SPLIT_1.
- */
-#define DICEK_SPLIT_1(funcname, arrayname, index) \
-    arrayname[index].DICEK_return = 0; \
-    arrayname[index].DICEK_thread = (void *) index; \
-    (funcname)((void *) &(arrayname[index]));
-
-#define DICEK_MERGE_1(arrayname, index)  /* nop */
-
-/*
- DICEK_SPLIT_MERGE starts nth threads vis DICEK_SPLIT_1. After the last of the threads has exited, execution proceeds
-with the instruction following DICEK_SPLIT_MERGE.
- */
-#define DICEK_SPLIT_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_1((funcname), (arrayname), (_DICEK_i)) \
-    }
-
-#define DICEK_SPLIT_M(funcname, arrayname, index) DICEK_SPLIT_1(funcname, arrayname, index)
-
-/* Just start the threads, without a merge. This is used for applications that do inter-thread synchronization */
-#define DICEK_SPLIT(funcname, arrayname, nth) DICEK_SPLIT_MERGE(funcname, arrayname, nth)
-
-/*
- Place a DICEK_SUB at the beginning of any routine called by a DICEK_SPLIT_MERGE. dtype is the type of the struct which
-is the array elements of arrayname, and argname is the name of the subroutine's argument. For example:
-
-  typedef struct f_vars {
-    DICEK_THREAD_VARS
-    int foo;
-    char * bar;
-  } f_vars;
-
-  (void *)myfunc(void * params)
-  {
-    DICEK_SUB(f_vars, params);
-    // other local variable declarations
-    ...
-  }
-
-For an example of a thread that performs inter-thread synchronization, see the example below (DICEK_CH_BEGIN macro)
-
-In emulated mode, we need a pointer to the struct that was passed in. This pointer is used by DICEK_RETURN
- */
-#define DICEK_SUB(dtype, argname) \
-  dtype * _DICEK_params = (dtype *) argname;
-
-/*
- The child thread should start with DICEK_CH_BEGIN, which grabs the semaphore indicating that the child is working.
-Then it should do its work, followed DICEK_CH_SYNC to tell the other threads in its group that it has finished
-its section of work.
- The master thread will eventually release the semaphore used by DICEK_CH_BEGIN allowing the child to continue
-with the next piece of work.
-
-  (void *)myfunc(void * params)
-  {
-    DICEK_SUB(f_vars, params);
-    // other local variable declarations
-    while(working) {
-      DICEK_CH_BEGIN
-      // Do computation here
-      DICEK_CH_SYNC
-    }
-    // We're done performing useful work
-    DICEK_CH_END
-    // Here we deallocate any temporary private storage
-    DICEK_RETURN(result)
-  }
-
-*/
-#define DICEK_CH_BEGIN \
-  fprintf(stderr, "Runtime error: DICEK does not support inter-thread communication in EMULATE mode.\n"); \
-  exit(-1);
-
-/*
- DICEK_CH_SYNC releases the "master working" semaphore, which tells the master that this thread is done with its
-work and the master can proceed to use the results that were produced by the child.
- */
-#define DICEK_CH_SYNC /* nop */
-
-/*
-  DICEK_INTERLOCK is called by the master thread to wait for the children to be done with their bits of work.
-It attempts to take all of the master_working semaphores, which are initially locked and don't get unlocked
-until the child threads reach their DICEK_CH_SYNC.
- */
-#define DICEK_INTERLOCK(arrayname, nth) DICEK_CH_BEGIN
-
-/*
-  DICEK_RESUME is called by the master thread to tell the child threads that they cal now continue with another
-piece of work. It unlocks the child_working semaphore so that the children can execute their DICEK_CH_BEGIN
-statement a second time.
- */
-#define DICEK_RESUME(arrayname, nth) /* nop */
-
-/*
- Put DICEK_CH_END at the point where a child thread has finished performing any actions that require
-cooperation or synchronization, and before the process ends. This macro releases the child_working semaphore
-so that the semaphore can be deallocated by the master process.
- The typical sequence is: DICEK_CH_END; deallocate private storage; DICEK_RETURN; exit. See the DICEK_CH_BEGIN
-block comment for a pseudo-code example.
- */
-#define DICEK_CH_END /* nop */
-
-/*
- Place a DICEK_RETURN at the very end of any routine called by a DICEK_SPLIT_MERGE. It passes returnv back
-to the calling (master) thread via a field routine's parameter block.
- WARNING: On some operating systems, this return value is passed through the kernel, while on others it is actually
-written directly into the parameter block. On some systems, this routine causes the thread to exit immediately, while on
-others it is possible to execute statements after the DICEK_RETURN. Users of this package should not rely on any of
-these specifics to be true in runtime.
- */
-#define DICEK_RETURN(returnv) _DICEK_params->DICEK_return = (void*)(returnv);
-
-#define DICEK_MERGE_M(arrayname, index)  /* nop */
-
-#define DICEK_MERGE(funcname, arrayname, nth) /* nop */
-
-#endif
-
-/* - - - - - - - - - - - - - - - - - - - - - End of the EMULATED section  - - - - - - - - - - - - - - - - - - - - - - */
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                ,        ,         //
-//                                                                                               /(        )`        //
-//                                                                                               \ \___   / |        //
-//         _nnnn_                                                                                /- _  `-/  '        //
-//        dGGGGMMb                                                                              (/\/ \ \   /\        //
-//       @p~qp~~qMb                                                                             / /   | `    \       //
-//       M|@||@) M|       ooooooooo.     .oooooo.    .oooooo..o ooooo ooooooo  ooooo            O O   ) /    |       //
-//       @,----.JM|       `888   `Y88.  d8P'  `Y8b  d8P'    `Y8 `888'  `8888    d8'             `-^--'`<     '       //
-//      JS^\__/  qKL       888   .d88' 888      888 Y88bo.       888     Y888..8P              (_.)  _  )   /        //
-//     dZP        qKRb     888ooo88P'  888      888  `"Y8888o.   888      `8888'                `.___/`    /         //
-//    dZP          qKKb    888         888      888      `"Y88b  888     .8PY888.                 `-----' /          //
-//   fZP            SMMb   888         `88b    d88' oo     .d8P  888    d8'  `888b   <----.     __ / __   \          //
-//   HZM            MMMM  o888o         `Y8bood8P'  8""88888P'  o888o o888o  o88888o <----|====O)))==) \) /====      //
-//   FqM            MMMM                                                             <----'    `--' `.__,' \         //
-// __| ".        |\dS"qML                                                                         |        |         //
-// |    `.       | `' \Zq                                                                          \       /       /\//
-//_)      \.___.,|     .'                                                                     ______( (_  / \______/ //
-//\____   )MMMMMP|   .'           Versions of the DICEK macros for Linux, Mac OS,           ,'  ,-----'   |          //
-//     `-'       `--' hjm             and other POSIX-compliant environments                `--{__________)          //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-#ifdef DICEK_USE_POSIX
-
-#define DICEK_SUPPORTS_BLOCKING 1
-
-#include <pthread.h>
-
-// Mach-based system (most notably Apple) use sysctlbyname to query the number of CPUs/cores/threads
-# ifdef __MACH__
-#   include <sys/types.h>
-#   include <sys/sysctl.h>
-# endif
-
-// TODO: Some Linux systems might need an #include to access the mechanism for finding out the number of threads.
-
-
-#define DICEK_THREAD_VARS \
-  long DICEK_tnum; \
-  void * DICEK_thread; \
-  pthread_mutex_t DICEK_master_wkg; \
-  pthread_mutex_t DICEK_child_wkg; \
-  void * DICEK_return;
-
-
-#ifdef __APPLE__
-# define DICEK_INIT_NTHR(nth) int nth; { \
-    size_t _DICEK_sz_in = sizeof(nth); \
-    long _DICEK_rv2 = sysctlbyname("hw.ncpu", (void *) (&nth), &_DICEK_sz_in, 0, 0); \
-    if (nth <= 0) { nth = 1; } \
-    if (nth > 64) { nth = 64; } }
-#else
-// TODO: How to query number of threads in Linux (I suspect reading /proc/cpuinfo will work on most,
-// but not all, Linuces)
-# define DICEK_INIT_NTHR(nth) int nth = 3;
-#endif
-
-#define DICEK_DATA(dtype, arrayname, nth) \
-    dtype * arrayname; \
-    arrayname = (dtype *)malloc(nth * sizeof(dtype)); \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      arrayname[_DICEK_i].DICEK_tnum = _DICEK_i; \
-      arrayname[_DICEK_i].DICEK_thread = 0; \
-      arrayname[_DICEK_i].DICEK_return = 0; \
-    }
-
-/* Just start a single thread. */
-#define DICEK_SPLIT_1(funcname, arrayname, index) \
-    arrayname[index].DICEK_return = 0; \
-    pthread_create((pthread_t *) (&(arrayname[index].DICEK_thread)), \
-        0, (void* (*)(void*))funcname, (void *) &(arrayname[index]));
-
-/* End a single thread. */
-#define DICEK_MERGE_1(arrayname, index) \
-    pthread_join((pthread_t) (arrayname[index].DICEK_thread), \
-        &(arrayname[index].DICEK_return));
-
-/* In the POSIX implementation, semaphore communication is supported, so the SPLIT_MERGE function
-(which by definition has no communication between master and spawned threads) does not need to
-initialize the semaphores, whereas DICEK_SPLIT_1 does. */
-#define DICEK_SPLIT_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_1(funcname, arrayname, _DICEK_i) \
-    } \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_MERGE_1(arrayname, _DICEK_i) \
-    }
-
-/* DICEK_SPLIT_M also sets up the semaphores in case the client is going to want to do that. */
-#define DICEK_SPLIT_M(funcname, arrayname, index) \
-    pthread_mutex_init(&(arrayname[index].DICEK_child_wkg), 0); \
-    pthread_mutex_init(&(arrayname[index].DICEK_master_wkg), 0); \
-    pthread_mutex_trylock(&(arrayname[index].DICEK_master_wkg)); \
-    DICEK_SPLIT_1(funcname, arrayname, index)
-
-/* Just start the threads, without a merge. This is used for applications that do inter-thread synchronization */
-#define DICEK_SPLIT(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_M((void* (*)(void*))funcname, (arrayname), _DICEK_i) \
-    }
-
-#define DICEK_SUB(dtype, argname) dtype * _DICEK_params = (dtype *) argname;
-
-#define DICEK_CH_BEGIN \
-  pthread_mutex_lock(&(_DICEK_params->DICEK_child_wkg));
-
-#define DICEK_CH_SYNC \
-  pthread_mutex_unlock(&(_DICEK_params->DICEK_master_wkg));
-
-#define DICEK_INTERLOCK(arrayname, nth) \
-  for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-    pthread_mutex_lock(&(arrayname[_DICEK_i].DICEK_master_wkg)); \
-  }
-
-#define DICEK_RESUME(arrayname, nth) \
-  for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-    pthread_mutex_unlock(&(arrayname[_DICEK_i].DICEK_child_wkg)); \
-  }
-
-#define DICEK_CH_END \
-  pthread_mutex_unlock(&(_DICEK_params->DICEK_child_wkg));
-
-#define DICEK_RETURN(returnv) pthread_exit((void*)(returnv));
-
-#define DICEK_MERGE_M(arrayname, index) \
-    DICEK_MERGE_1(arrayname, index) \
-    pthread_mutex_destroy(&(arrayname[index].DICEK_child_wkg)); \
-    pthread_mutex_unlock(&(arrayname[index].DICEK_master_wkg)); \
-    pthread_mutex_destroy(&(arrayname[index].DICEK_master_wkg));
-
-#define DICEK_MERGE(funcname, arrayname, nth) \
-  for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-    DICEK_MERGE_M((arrayname), _DICEK_i) \
-  }
-
-#endif
-
-/* - - - - - - - - - - - - - - - - - - - - - - End of the POSIX section - - - - - - - - - - - - - - - - - - - - - - - */
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//       _-^^^^-_                                  ,,                    ,,                                          //
-//      /      './^-____-^/  `7MMF'     A     `7MF'db                  `7MM                                          //
-//     /       ./.'      /     `MA     ,MA     ,V                        MM                                          //
-//     /      './.'      /      VM:   ,VVM:   ,V `7MM  `7MMpMMMb.   ,M""bMM  ,pW"Wq.`7M'    ,A    `MF',pP"Ybd        //
-//    /_-^^^^-_/.'      /        MM.  M' MM.  M'   MM    MM    MM ,AP    MM 6W'   `Wb VA   ,VAA   ,V  8I   `"        //
-//    /      './^-____-^/        `MM A'  `MM A'    MM    MM    MM 8MI    MM 8M     M8  VA ,V  VA ,V   `YMMMa.        //
-//   /       '/.'      /          :MM;    :MM;     MM    MM    MM `Mb    MM YA.   ,A9   VVV    VVV    L.   I8        //
-//   /      .'/.'      /           VF      VF    .JMML..JMML  JMML.`Wbmd"MML.`Ybmd9'     W      W     M9mmmP'        //
-//  /_-^^^^-_/.'      /                                                                                              //
-//            ^-___rpm           Versions of the DICEK macros for Microsoft Windows                                  //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-#ifdef DICEK_USE_PROCESS_H
-
-// This is based on:
-//   http://msdn.microsoft.com/en-US/library/kdzttdcb(v=VS.80).aspx  (_beginthread)
-//   http://msdn.microsoft.com/en-us/library/ms687032(v=VS.85).aspx  (WaitForSingleObject)
-//   http://msdn.microsoft.com/en-us/library/ms685071(v=VS.85).aspx  (ReleaseSemaphore)
-//   http://msdn.microsoft.com/en-us/library/ms682438(v=VS.85).aspx  (CreateSemaphore)
-//
-//   http://msdn.microsoft.com/en-us/library/ms682411(v=VS.85).aspx  (CreateMutex, no longer used)
-//   http://msdn.microsoft.com/en-us/library/ms685066(v=VS.85).aspx  (ReleaseMutex, no longer used)   
-
-#define DICEK_SUPPORTS_BLOCKING 1
-
-#include <process.h>
-#include <windows.h>
-
-#define DICEK_THREAD_VARS \
-  long DICEK_tnum; \
-  HANDLE DICEK_thread; \
-  HANDLE DICEK_master_wkg; \
-  HANDLE DICEK_child_wkg; \
-  void * DICEK_return;
-
-#define DICEK_INIT_NTHR(nth) \
-    int nth; \
-    { SYSTEM_INFO si; GetSystemInfo(&si); nth = si.dwNumberOfProcessors; }
-
-#define DICEK_DATA(dtype, arrayname, nth) \
-    dtype * arrayname; \
-    arrayname = (dtype *)malloc(nth * sizeof(dtype)); \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      arrayname[_DICEK_i].DICEK_tnum = _DICEK_i; \
-      arrayname[_DICEK_i].DICEK_thread = (HANDLE) NULL; \
-      arrayname[_DICEK_i].DICEK_return = 0; \
-    }
-
-#define DICEK_SPLIT_1(funcname, arrayname, index) \
-    arrayname[index].DICEK_return = 0; \
-    arrayname[index].DICEK_thread = (HANDLE) _beginthreadex(NULL,0,funcname, \
-        (void *) &(arrayname[index]), 0, NULL);
-
-#define DICEK_MERGE_1(arrayname, index) \
-    WaitForSingleObject(arrayname[index].DICEK_thread, INFINITE);
-
-#define DICEK_SPLIT_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_1(funcname, arrayname, _DICEK_i) \
-    } \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_MERGE_1(arrayname, _DICEK_i) \
-    }
-
-// DICEK_SPLIT_M also sets up the semaphores in case the client is going to want to do that.
-// child semaphore is initially free (value 1); master semaphore is initially taken
-#define DICEK_SPLIT_M(funcname, arrayname, index) \
-    arrayname[index].DICEK_child_wkg = CreateSemaphore(NULL, 1, 1, NULL); \
-    arrayname[index].DICEK_master_wkg = CreateSemaphore(NULL, 0, 1, NULL); \
-    DICEK_SPLIT_1(funcname, arrayname, index)
-
-/* Just start the threads, without a merge. This is used for applications that do inter-thread synchronization */
-#define DICEK_SPLIT(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_SPLIT_M((funcname), (arrayname), (_DICEK_i)) \
-    }
-
-/* Save a pointer to the param block for use by DICEK_RETURN */
-#define DICEK_SUB(dtype, argname) dtype * _DICEK_params = (dtype *) argname;
-
-#define DICEK_CH_BEGIN \
-    WaitForSingleObject(_DICEK_params->DICEK_child_wkg,INFINITE);
-
-#define DICEK_CH_SYNC \
-    ReleaseSemaphore(_DICEK_params->DICEK_master_wkg, 1, NULL);
-
-#define DICEK_INTERLOCK(arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-        WaitForSingleObject(arrayname[_DICEK_i].DICEK_master_wkg,INFINITE); \
-    }
-
-#define DICEK_RESUME(arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      ReleaseSemaphore(arrayname[_DICEK_i].DICEK_child_wkg, 1, NULL); \
-    }
-
-#define DICEK_CH_END \
-    ReleaseSemaphore(_DICEK_params->DICEK_child_wkg, 1, NULL);
-
-#define DICEK_RETURN(returnv) \
-    _DICEK_params->DICEK_return = (void*)(returnv); \
-    _endthread();
-
-#define DICEK_MERGE_M(arrayname, index) \
-    DICEK_MERGE_1(arrayname, index) \
-    CloseHandle(arrayname[index].DICEK_thread); \
-    CloseHandle(arrayname[index].DICEK_child_wkg); \
-    ReleaseSemaphore(arrayname[index].DICEK_master_wkg, 1, NULL); \
-    CloseHandle(arrayname[index].DICEK_master_wkg);
-
-#define DICEK_MERGE(funcname, arrayname, nth) \
-    for(int _DICEK_i=0; _DICEK_i<nth; _DICEK_i++) { \
-      DICEK_MERGE_M((arrayname), _DICEK_i) \
-    }
-
-#endif
-
-/* - - - - - - - - - - - - - - - - - - - - - End of the PROCESS_H section - - - - - - - - - - - - - - - - - - - - - - */
diff --git a/SpeedComparisons/GrayScott_HWIVector/display_hwiv.cpp b/SpeedComparisons/GrayScott_HWIVector/display_hwiv.cpp
deleted file mode 100644
index aad53bfa7..000000000
--- a/SpeedComparisons/GrayScott_HWIVector/display_hwiv.cpp
+++ /dev/null
@@ -1,129 +0,0 @@
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib
-#include <stdio.h>
-
-// local:
-#include "display_hwiv.h"
-
-#define INDEX(a,x,y) ((a)[(x)*full_width+(y)])
-
-bool display(int g_width, int g_height, float *r, float *g, float *b,
-             double iteration, float model_scale, bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message, bool write_video)
-{
-  int full_width = g_width + 8;
-  static bool need_init = true;
-
-  static IplImage *im,*im2;
-  static int border = 0;
-  static CvFont font;
-  static CvVideoWriter *video;
-  static const CvScalar white = cvScalar(255,255,255);
-
-  const char *title = "Press ESC to quit";
-
-  if(need_init)
-  {
-    need_init = false;
-
-    im = cvCreateImage(cvSize(g_width,g_height),IPL_DEPTH_8U,3);
-    cvSet(im,cvScalar(0,0,0));
-    im2 = cvCreateImage(cvSize(g_width*scale,g_height*scale),IPL_DEPTH_8U,3);
-    
-    cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-    
-    double hScale=0.4;
-    double vScale=0.4;
-    int lineWidth=1;
-    cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-    if(write_video)
-    {
-      video = cvCreateVideoWriter("vid-Gray-Scott.avi",CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im),1);
-      if(video == NULL) {
-        fprintf(stdout, "NULL from cvCreateVideoWriter\n"); exit(-1);
-      }
-    }
-  }
-
-  // convert float arrays to IplImage for OpenCV to display
-  float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-  if(auto_brighten)
-  {
-    for(int i=0;i<g_height;i++)
-    {
-      for(int j=4;j<g_width+4;j++)
-      {
-        val = INDEX(r,i+4,j);
-        if(val<minR) minR=val; if(val>maxR) maxR=val;
-
-        val = INDEX(g,i+4,j);
-        if(val<minG) minG=val; if(val>maxG) maxG=val;
-
-        val = INDEX(b,i+4,j);
-        if(val<minB) minB=val; if(val>maxB) maxB=val;
-      }
-    }
-  }
-  for(int i=0;i<g_height;i++)
-  {
-    for(int j=0;j<g_width;j++)
-    {
-      float val;
-      val = INDEX(r,i,full_width-j-5);
-      if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-      else val *= manual_brighten;
-      if(val<0) val=0; if(val>255) val=255;
-      ((uchar *)(im->imageData + i*im->widthStep))[j*im->nChannels + 2] = (uchar)val;
-
-      val = INDEX(g,i,full_width-j-5);
-      if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-      else val *= manual_brighten;
-      if(val<0) val=0; if(val>255) val=255;
-      ((uchar *)(im->imageData + i*im->widthStep))[j*im->nChannels + 1] = (uchar)val;
-
-      val = INDEX(b,i,full_width-j-5);
-      if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-      else val *= manual_brighten;
-      if(val<0) val=0; if(val>255) val=255;
-      ((uchar *)(im->imageData + i*im->widthStep))[j*im->nChannels + 0] = (uchar)val;
-    }
-  }
-
-  cvResize(im,im2);
-
-  char txt[100];
-  const char * fmt;
-
-  if (g_width > 170) {
-    fmt = "Generation=%7g (model's t=%g)";
-  } else if (g_width > 140) {
-    fmt = "Gen.=%7g (model's t=%g)";
-  } else {
-    fmt = "G=%7g t=%g";
-  }
-  sprintf(txt,fmt,iteration, ((double) iteration) / model_scale);
-  cvPutText(im2,txt,cvPoint(5,20),&font,white);
-  cvPutText(im2,message,cvPoint(5,40),&font,white);
-
-  if(write_video)
-    cvWriteFrame(video,im);
-
-  cvShowImage(title,im2);
-  
-  int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-  if(key==27) // did user ask to quit?
-  {
-    if(write_video)
-      cvReleaseVideoWriter(&video);
-    cvDestroyWindow(title);
-    cvReleaseImage(&im);
-    cvReleaseImage(&im2);
-    return true;
-  }
-  return false;
-}
-
diff --git a/SpeedComparisons/GrayScott_HWIVector/display_hwiv.h b/SpeedComparisons/GrayScott_HWIVector/display_hwiv.h
deleted file mode 100644
index b174670d5..000000000
--- a/SpeedComparisons/GrayScott_HWIVector/display_hwiv.h
+++ /dev/null
@@ -1,3 +0,0 @@
-bool display(int width, int height, float *r, float *g, float *b,
-             double iteration, float model_scale, bool auto_brighten,float manual_brighten,
-			 int scale,int delay_ms,const char* message, bool write_video);
diff --git a/SpeedComparisons/GrayScott_HWIVector/gray_scott_hwivector.cpp b/SpeedComparisons/GrayScott_HWIVector/gray_scott_hwivector.cpp
deleted file mode 100644
index ea1454906..000000000
--- a/SpeedComparisons/GrayScott_HWIVector/gray_scott_hwivector.cpp
+++ /dev/null
@@ -1,904 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// hardware
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-# include <xmmintrin.h>
-#endif
-
-// To convince yourself that the macro library works on any hardware,
-// un-comment this "#define HWIV_EMULATE", and you'll get the macros inside
-// the #ifdef HWIV_V4F4_EMULATED block in hwi_vector.h.  The emulated
-// macros do everything with normal floats and arrays, and run about 3-4 times
-// slower.
-//#define HWIV_EMULATE
-#define HWIV_WANT_V4F4
-#include "hwi_vector.h"
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-
-	// TODO: Make sure this is the correct include file for the FindFirstFile function (used in FILE_LENGTH macro below)
-    #include <windows.h>
-#else
-    #include <sys/stat.h>
-    #include <sys/time.h>
-#endif
-
-// local:
-//#define DICEK_EMULATE
-#include "dicek.h"
-#include "display_hwiv.h"
-
-static long g_width = 256;
-static long g_height = 256;
-
-float *allocate(long width, long height, const char * error_text, bool for_mm);
-float *allocate(long width, long height, const char * error_text, bool for_mm)
-{
-  long sz;
-  float * rv;
-
-  sz = width *  height * sizeof(*rv);
-  if (for_mm) {
-    rv = (float *) _mm_malloc(sz, 16);
-  } else {
-    rv = (float *) malloc(((size_t)sz));
-  }
-
-   if (rv == NULL) {
-    fprintf(stderr, "allocate: Could not get %ld bytes for %s\n", ((long) sz),
-      error_text);
-    exit(-1);
-  }
-  return (rv);
-}
-
-void init(float *a, float *b, long width, long height);
-
-void compute_allocate(void);
-void compute_import(float *u, float *v, long width, long height);
-
-typedef struct compute_params {
-  DICEK_THREAD_VARS;
-  float *u;
-  float *v;
-  float *du;
-  float *dv;
-  float D_u;
-  float D_v;
-  float F;
-  float k;
-  float speed;
-  int parameter_space;
-  int num_its;
-  long start_row;
-  long end_row;
-  int interlock_type;
-} compute_params;
-
-unsigned __stdcall  compute(void * param_block);  // Arg is really "compute_params * param_block"
-
-void compute_dispatch(float *u, float *v, float *du, float *dv,
-  float D_u, float D_v, float F, float k, float speed,
-  int parameter_space, int num_its, int num_threads);
-
-static int g_paramspace = 0;
-static int g_wrap = 1;
-static float g_k = 0.064;
-static float g_F = 0.035;
-static bool g_video = false;
-
-#define VECSIZE 4
-
-int main(int argc, char * * argv)
-{
-  DICEK_INIT_NTHR(g_threads)
-  printf("DiceK reports %d threads.\n", g_threads);
-
-  // Here we implement the Gray-Scott model, as described here:
-  // http://arxiv.org/abs/patt-sol/9304003
-  //    (the seminal paper by Pearson)
-  // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-  //    (a present university course project, by Greg Turk at Georgia Tech)
-  // http://www.mrob.com/pub/comp/xmorphia/index.html
-  //    (a web exhibit with over 100 videos and 500 images)
-
-  // -- parameters --
-  float D_u = 0.082;
-  float D_v = 0.041;
-
-  // The default is equivalent to options:   -F 0.035 -k 0.064
-  g_k = 0.064;  g_F = 0.035;
-
-  // Other pattern-types to try (pass -F and -k as argument on command line, some require "-density 1" as well):
-  //
-  // -F 0.0118 -k 0.0475              Spiral waves
-  // -F 0.022  -k 0.059               Spots that multiply and keep killing each other off
-  // -F 0.035  -k 0.06                Stripes with branching (fingerprint)
-  // -F 0.04   -k 0.064               For spots that multiply, with stripes mixed in
-  // -F 0.056  -k 0.065               Long stripes ("-density 1" option helps here)
-  // -F 0.062  -k 0.0609 -density 2   "Uskate world", where I found all the Wolfram-class-4 behaviour
-  // -F 0.094  -k 0.059  -density 1   "soap bubbles"
-  // -F 0.094  -k 0.057 -lowback -density 1   Inverse soap bubbles
-
-  float speed = 1.0;
-
-  for (int i = 1; i < argc; i++) {
-    if (0) {
-    } else if ((i+1<argc) && (strcmp(argv[i],"-height")==0)) {
-      // set height
-      i++; g_height = atoi(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-size")==0)) {
-      // set both width and height
-      i++; g_height = g_width = atoi(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-threads")==0)) {
-      // specify number of threads to use
-      i++; g_threads = atoi(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-width")==0)) {
-      // set width
-      i++; g_width = atoi(argv[i]);
-    } else if (strcmp(argv[i],"-wrap")==0) {
-      // patterns wrap around ("torus", also called "continuous boundary
-      // condition")
-      g_wrap = 1;
-    } else {
-      fprintf(stderr, "Unrecognized argument, or parameter needed: '%s'\n", argv[i]);
-      exit(-1);
-    }
-  }
-
-  if (g_width % VECSIZE) {
-    g_width = ((g_width/VECSIZE)+1)*VECSIZE;
-  }
-  long full_width = g_width + 2*VECSIZE;
-
-  /* We cannot use multiple threads unless DiceK supports thread synchronization */
-  if (!(DICEK_SUPPORTS_BLOCKING)) {
-    g_threads = 1;
-  }
-
-  printf("Will use %d threads.\n", g_threads);
-
-  // ----------------
-  
-  // these arrays store the chemical concentrations:
-  float *u = 0;
-  float *v = 0;
-  // these arrays store the rate of change of those chemicals:
-  float *du = 0;
-  float *dv = 0;
-
-  u = allocate(full_width, g_height, "U array", true);
-  v = allocate(full_width, g_height, "V array", true);
-  du = allocate(full_width, g_height, "D_u array", true);
-  dv = allocate(full_width, g_height, "D_v array", true);
-
-  // put the initial conditions into each cell
-  init(u,v, full_width, g_height);
-
-#define N_FRAMES_PER_DISPLAY 1000
-
-  double iteration = 0;
-  double fps_avg = 0.0; // decaying average of fps
-  double Mcgs;
-  while(true) 
-  {
-    struct timeval tod_record;
-    double tod_before, tod_after, tod_elapsed;
-    double fps = 0.0;     // frames per second
-
-    gettimeofday(&tod_record, 0);
-    tod_before = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-    // compute:
-    compute_dispatch(u, v, du, dv, D_u, D_v, g_F, g_k, speed, g_paramspace, N_FRAMES_PER_DISPLAY, g_threads);
-    iteration += ((double)N_FRAMES_PER_DISPLAY);
-
-    gettimeofday(&tod_record, 0);
-    tod_after = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-    tod_elapsed = tod_after - tod_before;
-    fps = ((double)N_FRAMES_PER_DISPLAY) / tod_elapsed;
-    // We display an exponential moving average of the fps measurement
-    fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-    Mcgs = fps_avg * ((double) full_width) * ((double) g_height) / 1.0e6;
-
-    char msg[1000];
-    sprintf(msg,"GrayScott - %0.2f fps (%.2f Mcgs)", fps_avg, Mcgs);
-
-    // display:
-    {
-      int chose_quit;
-      chose_quit = display(g_width,g_height,u,u,u,iteration,1.0,false,200.0f,2,10,msg,false);
-      if (chose_quit) // did user ask to quit?
-        break;
-    }
-  }
-}
-
-/* Spawn threads and dispatch to compute() routine inside threads */
-void compute_dispatch(float *u, float *v, float *du, float *dv,
-  float D_u, float D_v, float F, float k, float speed,
-  int parameter_space, int num_its, int nthreads)
-{
-  if (nthreads <= 0) {
-    nthreads = 1;
-  }
-  DICEK_DATA(compute_params, cp, nthreads);
-
-  long i;
-  long a_row = 0;
-
-  /* Set up all the parameter blocks */
-  for(i=0; i<nthreads; i++) {
-    cp[i].u = u; cp[i].v = v; cp[i].du = du; cp[i].dv = dv; cp[i].D_u = D_u; cp[i].D_v = D_v;
-    cp[i].F = F; cp[i].k = k; cp[i].speed = speed; cp[i].parameter_space = parameter_space;
-    cp[i].num_its = num_its; cp[i].start_row = a_row;
-    a_row = (g_height * (i+1)) / nthreads;
-    cp[i].end_row = a_row;
-    cp[i].interlock_type = (nthreads>1) ? 1 : 0;
-  }
-
-  if (nthreads > 1) {
-    /* Start N threads, each will immediately begin the first part of its computation */
-    DICEK_SPLIT(compute, cp, nthreads);
-
-    /* Now for each iteration we need to sync the threads once. Each iteration consists of two
-       work phases.
-         During the first phase u and v are being read and du,dv are written; each thread
-       reads neighboring thread's data (in a single row along its top and bottom borders)
-         During the second phase u,v are overwritten with the next generation; each thread keeps
-       to its own area.
-         Because inter-thread communication only happens during phase 1, there need be only
-       one barrier per loop. */
-    for(i=0; i<num_its; i++) {
-      DICEK_INTERLOCK(cp, nthreads); // Wait for all threads to get to the end of the i'th derivative phase
-      DICEK_RESUME(cp, nthreads);  // Let them continue
-    }
-
-    // Now wait for the threads to finish updating u and v arrays the final time, and exit
-    DICEK_MERGE(compute, cp, nthreads);
-  } else {
-    /* With 1 thread it's more efficient to just call the compute routine directly */
-    compute((void *)&(cp[0]));
-  }
-}
-
-#ifndef max
-# define max(a,b) (((a) > (b)) ? (a) : (b))
-# define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-#ifndef minmax
-# define minmax(v, lo, hi) max(lo, min(v, hi))
-#endif
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-  float rv;
-
-  rv = lower + rand()*(upper-lower)/RAND_MAX;
-  rv = minmax(rv, 0.0, 1.0);
-  return rv;
-}
-
-void init(float *u, float *v, long width, long height)
-{
-  long nsp, i;
-  long base, var;
-
-  srand((unsigned int)time(NULL));
-
-  for(int i = 0; i < height; i++) {
-    for(int j = 0; j < width; j++) {
-      // start with a uniform field with an approximate circle in the middle
-      if (hypot(i-height/2,j-width/3)<=frand(2,5))
-      {
-        u[i*width+j] = frand(0.0,0.1);
-        v[i*width+j] = frand(0.9,1.0);
-      } else {
-        u[i*width+j] = frand(0.9,1.0);
-        v[i*width+j] = frand(0.0,0.1);
-      }
-    }
-  }
-}
-
-static float * c_u;
-static float * c_v;
-static float * c_du;
-static float * c_dv;
-static long c_full_width;
-
-void compute_allocate(long width, long height)
-{
-  c_full_width = width + 2*VECSIZE;
-  c_u = allocate(c_full_width, height, "U array", true);
-  c_v = allocate(c_full_width, height, "V array", true);
-  c_du = allocate(c_full_width, height, "D_u array", true);
-  c_dv = allocate(c_full_width, height, "D_v array", true);
-}
-
-void compute_import(float *u, float *v, long width, long height)
-{
-  for(long i = 0; i < height; i++) {
-    for(long j = 0; j < width; j++) {
-      c_u[(i+VECSIZE)*width+j] = u[i*width+j];
-      c_v[(i+VECSIZE)*width+j] = v[i*width+j];
-    }
-  }
-}
-
-unsigned __stdcall compute(void * gpb)
-{
-  DICEK_SUB(compute_params, gpb);
-  const int full_width = g_width + 2*VECSIZE;
-  const int wid_v = full_width / VECSIZE;
-
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-  /* On Intel we disable accurate handling of denorms and zeros. This is an
-     important speed optimization. */
-  int oldMXCSR = _mm_getcsr(); //read the old MXCSR setting
-  int newMXCSR = oldMXCSR | 0x8040; // set DAZ and FZ bits
-  _mm_setcsr( newMXCSR ); //write the new MXCSR setting to the MXCSR
-#endif
-
-  compute_params * param_block;
-  param_block = (compute_params *) gpb;
-  float *u = param_block->u;
-  float *v = param_block->v;
-  float *du = param_block->du;
-  float *dv = param_block->dv;
-  float D_u = param_block->D_u;
-  float D_v = param_block->D_v;
-  float F = param_block->F;
-  float k = param_block->k;
-  float speed = param_block->speed;
-  int parameter_space = param_block->parameter_space;
-  int num_its = param_block->num_its;
-  int start_row = param_block->start_row;
-  int end_row = param_block->end_row;
-  int interlock = param_block->interlock_type;
-
-
-
-  int iter;
-
-  if (interlock) { DICEK_CH_BEGIN }
-
-  // Scan per iteration
-  for(iter = 0; iter < num_its; iter++) {
-
-//printf("iter %d rows [%ld,%ld)\n",iter,start_row,end_row);
-
-  // Scan per row
-  for(int i = start_row; i < end_row; i++) {
-    const V4F4 v4_F = v4SPLAT(F);
-    const V4F4 v4_k = v4SPLAT(k);
-    const V4F4 v4_Du = v4SPLAT(D_u);
-    const V4F4 v4_Dv = v4SPLAT(D_v);
-    int iprev,inext;
-    if (g_wrap) {
-      /* Periodic boundary condition */
-      iprev = (i+g_height-1) % g_height;
-      inext = (i+1) % g_height;
-    } else {
-      /* The edges are their own neighbors. This amounts to a von Neumann boundary condition. */
-      iprev = max(i-1, 0);
-      inext = min(i+1, g_height-1);
-    }
-
-    // Scan per column in steps of vector width
-    for(int j = 1; j < wid_v-1; j++) {
-      V4F4 * v_ubase = ((V4F4 *)u)+i*wid_v+j;
-      V4F4 * v_vbase = ((V4F4 *)v)+i*wid_v+j;
-      V4F4 * v_dubase = ((V4F4 *)du)+i*wid_v+j;
-      V4F4 * v_dvbase = ((V4F4 *)dv)+i*wid_v+j;
-      V4F4 u_left = _mm_loadu_ps(((float*)v_ubase)-1);
-      V4F4 u_right = _mm_loadu_ps(((float*)v_ubase)+1);
-      V4F4 v_left = _mm_loadu_ps(((float*)v_vbase)-1);
-      V4F4 v_right = _mm_loadu_ps(((float*)v_vbase)+1);
-      V4F4 * v_ub_prev = ((V4F4 *)u)+iprev*wid_v+j;
-      V4F4 * v_ub_next = ((V4F4 *)u)+inext*wid_v+j;
-      V4F4 * v_vb_prev = ((V4F4 *)v)+iprev*wid_v+j;
-      V4F4 * v_vb_next = ((V4F4 *)v)+inext*wid_v+j;
-
-      // To compute the Laplacians of u and v, we use the 5-point neighbourhood for the Euler discrete method:
-      //    nabla(x) = x[i][j-1]+x[i][j+1]+x[i-1][j]+x[i+1][j] - 4*x[i][j];
-      // ("nabla" is the name of the "upside down delta" symbol used for the Laplacian in equations)
-#     define NABLA_5PT(ctr,left,right,up,down) \
-         v4SUB(v4ADD(v4ADD(v4ADD(left,right),up),down),v4MUL(ctr,v4SPLAT(4.0f)))
-
-      // compute the new rate of change of u and v
-      V4F4 v4_uvv = v4MUL(*v_ubase,v4MUL(*v_vbase,*v_vbase)); // u*v^2 is used twice
-
-      /* Scalar code is:     du[i][j] = D_u * nabla_u - u*v^2 + F*(1-u);
-         We treat it as:                D_u * nabla_u - (u*v^2 - F*(1-u)) */
-      *v_dubase = v4SUB(v4MUL(v4_Du,
-         NABLA_5PT(*v_ubase, u_left, u_right, *v_ub_prev, *v_ub_next)),
-                           v4SUB(v4_uvv,v4MUL(v4_F,v4SUB(v4SPLAT(1.0f),*v_ubase))));
-
-      /* dv formula is similar:  dv[i][j] = D_v * nabla_v + u*v^2 - (F+k)*v; */
-      *v_dvbase = v4ADD(v4MUL(v4_Dv,
-         NABLA_5PT(*v_vbase, v_left, v_right, *v_vb_prev, *v_vb_next)),
-                           v4SUB(v4_uvv,v4MUL(v4ADD(v4_F,v4_k),*v_vbase)));
-    }
-
-  } // End of scan per row
-
-  if (interlock) { DICEK_CH_SYNC }
-  if (interlock) { DICEK_CH_BEGIN }
-
-  {
-    int right_b, left_b;
-    if (g_wrap) {
-      right_b = wid_v-2;
-      left_b = 1;
-    } else {
-      right_b = 1;
-      left_b = wid_v-2;
-    }
-  // effect change
-    for(int i = start_row; i < end_row; i++) {
-      for(int j = 1; j < wid_v-1; j++) {
-        const V4F4 v4_speed = v4SPLAT(speed);
-        V4F4 * v_ubase = ((V4F4 *)u)+i*wid_v+j;
-        V4F4 * v_vbase = ((V4F4 *)v)+i*wid_v+j;
-        V4F4 * v_dubase = ((V4F4 *)du)+i*wid_v+j;
-        V4F4 * v_dvbase = ((V4F4 *)dv)+i*wid_v+j;
-        // u[i][j] = u[i][j] + speed * du[i][j];
-        *v_ubase = v4ADD(v4MUL(v4_speed, *v_dubase), *v_ubase);
-        // v[i][j] = v[i][j] + speed * dv[i][j];
-        *v_vbase = v4ADD(v4MUL(v4_speed, *v_dvbase), *v_vbase);
-      }
-      // Update cells on boundary from one row inland
-      *(((V4F4 *)u)+i*wid_v) = *(((V4F4 *)u)+i*wid_v+right_b);
-      *(((V4F4 *)v)+i*wid_v) = *(((V4F4 *)v)+i*wid_v+right_b);
-      *(((V4F4 *)u)+i*wid_v+wid_v-1) = *(((V4F4 *)u)+i*wid_v+left_b);
-      *(((V4F4 *)v)+i*wid_v+wid_v-1) = *(((V4F4 *)v)+i*wid_v+left_b);
-    }
-  }
-
-  } // End of scan per iteration
-
-  // finish last phase 2
-  if (interlock) { DICEK_CH_SYNC }
-
-  DICEK_CH_END
-
-  return 0;
-}
-
-
-
-#ifdef COMPUTE_A
-
-/* The parameter space code, specifically the "if(parameter_space)" test itself, causes a 2.5% slowdown even when the
-   parameter_space flag is false */
-//#define SUPPORT_PARAM_SPACE
-
-void * compute(void * gpb)
-{
-  DICEK_SUB(compute_params, gpb);
-
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-  /* On Intel we disable accurate handling of denorms and zeros. This is an
-     important speed optimization. */
-  int oldMXCSR = _mm_getcsr(); //read the old MXCSR setting
-  int newMXCSR = oldMXCSR | 0x8040; // set DAZ and FZ bits
-  _mm_setcsr( newMXCSR ); //write the new MXCSR setting to the MXCSR
-#endif
-
-  compute_params * param_block;
-  param_block = (compute_params *) gpb;
-  float *u = param_block->u;
-  float *v = param_block->v;
-  float *du = param_block->du;
-  float *dv = param_block->dv;
-  float D_u = param_block->D_u;
-  float D_v = param_block->D_v;
-  float F = param_block->F;
-  float k = param_block->k;
-  float speed = param_block->speed;
-  int parameter_space = param_block->parameter_space;
-  int num_its = param_block->num_its;
-  long start_row = param_block->start_row;
-  long end_row = param_block->end_row;
-  int interlock = param_block->interlock_type;
-
-  int iter;
-#ifndef HWIV_HAVE_V4F4
-  fprintf(stdout, "Did not get vector macros from HWIV\n");
-  exit(-1);
-#endif
-  // Vector "constants": speed, F, k, D_u, D_v
-  V4F4 v4_speed, v4_F, v4_k, v4_Du, v4_Dv;
-  // Pointers used to load data from rows of the grid
-  V4F4 *v_ub_prev, *v_ubase, *v_ub_next;
-  V4F4 *v_vb_prev, *v_vbase, *v_vb_next;
-  // Actual grid data
-  V4F4 v4_u_l, v4_u, v4_u_r;
-  V4F4 v4_v_l, v4_v, v4_v_r;
-  V4F4 v4_uvv;
-  // Pointers to second grid where we write the results of the main computation
-  V4F4 *v_dubase, *v_dvbase;
-
-#ifdef SUPPORT_PARAM_SPACE
-  const float k_min=0.045f, k_max=0.07f, F_min=0.01f, F_max=0.09f;
-  float k_diff;
-  V4F4 v4_kdiff;
-  k_diff = (k_max-k_min)/g_height;
-  v4_kdiff = v4SET(0, -k_diff, -2*k_diff, -3*k_diff);
-#endif
-
-  // Initialize our vectorized scalars
-  v4_speed = v4SPLAT(speed);
-  v4_F = v4SPLAT(F);
-  v4_k = v4SPLAT(k);
-  v4_Du = v4SPLAT(D_u);
-  v4_Dv = v4SPLAT(D_v);
-
-  // Scan per iteration
-  for(iter = 0; iter < num_its; iter++) {
-
-  if (interlock) { DICEK_CH_BEGIN }
-
-//printf("iter %d rows [%ld,%ld)\n",iter,start_row,end_row);
-
-  // Scan per row
-  for(long i = start_row; i < end_row; i++) {
-    long iprev,inext;
-    long v_j2;
-    if (g_wrap) {
-      /* Periodic boundary condition */
-      iprev = (i+g_height-1) % g_height;
-      inext = (i+1) % g_height;
-    } else {
-      /* The edges are their own neighbors. This amounts to a Neumann boundary condition. */
-      iprev = max(i-1, 0);
-      inext = min(i+1, g_height-1);
-    }
-
-    /* Get pointers to beginning of rows for each of the grids. We access
-       3 rows each for u and v, and 1 row each for du and dv. */
-    v_ubase = (V4F4 *)&INDEX(u,i,0);
-    v_ub_prev = (V4F4 *)&INDEX(u,iprev,0);
-    v_ub_next = (V4F4 *)&INDEX(u,inext,0);
-    v_vbase = (V4F4 *)&INDEX(v,i,0);
-    v_vb_prev = (V4F4 *)&INDEX(v,iprev,0);
-    v_vb_next = (V4F4 *)&INDEX(v,inext,0);
-    v_dubase = (V4F4 *)&INDEX(du,i,0);
-    v_dvbase = (V4F4 *)&INDEX(dv,i,0);
-
-#ifdef SUPPORT_PARAM_SPACE
-    if (parameter_space) {
-      // set F for this row (ignore the provided value)
-      F = F_min + (g_height-i-1) * (F_max-F_min)/g_width;
-      v4_F = v4SPLAT(F);
-    }
-#endif
-
-    /* Pre-load the first two blocks of data we need, which are the "center"
-       and "right" blocks from the end of the row (as if we have just wrapped
-       around from the end of the row back to the beginning) */
-    v_j2 = g_wrap ? ((g_width-4)/VECSIZE) : 0;
-
-    v4_u = *(v_ubase+v_j2);
-    v4_v = *(v_vbase+v_j2);
-    v4_u_r = *v_ubase++;
-    v4_v_r = *v_vbase++;
-
-    // Scan per column in steps of vector width
-    for(long j = 0; j < g_width-VECSIZE; j+=VECSIZE) {
-      // Get a new 4 pixels from the current row and shift the other 8 pixels over
-      v4_u_l = v4_u; v4_u = v4_u_r; v4_u_r = *v_ubase;
-      v4_v_l = v4_v; v4_v = v4_v_r; v4_v_r = *v_vbase;
-
-#ifdef SUPPORT_PARAM_SPACE
-      if (parameter_space) {
-        // set k for this column (ignore the provided value)
-        k = k_min + (g_width-j-1)*k_diff;
-        // k decreases by k_diff each time j increases by 1, so this vector
-        // needs to contain 4 different k values. We use v4_kdiff, pre-computed
-        // above, to accomplish this.
-        v4_k = v4ADD(v4SPLAT(k),v4_kdiff);
-      }
-#endif
-
-      // To compute the Laplacians of u and v, we use the 5-point neighbourhood for the Euler discrete method:
-      //    nabla(x) = x[i][j-1]+x[i][j+1]+x[i-1][j]+x[i+1][j] - 4*x[i][j];
-      // ("nabla" is the name of the "upside down delta" symbol used for the Laplacian in equations)
-#     define NABLA_5PT(ctr,left,right,up,down) \
-         v4SUB(v4ADD(v4ADD(v4ADD(left,right),up),down),v4MUL(ctr,v4SPLAT(4.0f)))
-
-      // compute the new rate of change of u and v
-      v4_uvv = v4MUL(v4_u,v4MUL(v4_v,v4_v)); // u*v^2 is used twice
-
-      /* Scalar code is:     du[i][j] = D_u * nabla_u - u*v^2 + F*(1-u);
-         We treat it as:                D_u * nabla_u - (u*v^2 - F*(1-u)) */
-      *v_dubase = v4SUB(v4MUL(v4_Du,
-         NABLA_5PT(v4_u, v4RAISE(v4_u,v4_u_l), v4LOWER(v4_u,v4_u_r), *v_ub_prev, *v_ub_next)),
-                           v4SUB(v4_uvv,v4MUL(v4_F,v4SUB(v4SPLAT(1.0f),v4_u))));
-
-      /* dv formula is similar:  dv[i][j] = D_v * nabla_v + u*v^2 - (F+k)*v; */
-      *v_dvbase = v4ADD(v4MUL(v4_Dv,
-         NABLA_5PT(v4_v, v4RAISE(v4_v,v4_v_l), v4LOWER(v4_v,v4_v_r), *v_vb_prev, *v_vb_next)),
-                           v4SUB(v4_uvv,v4MUL(v4ADD(v4_F,v4_k),v4_v)));
-
-      v_ub_prev++; v_ub_next++;
-      v_ubase++;   v_vbase++;
-      v_vb_prev++; v_vb_next++;
-      v_dubase++;  v_dvbase++;
-    }
-
-    /* Now we do the last 4 pixels. This is unrolled out of the main loop just to avoid having to do the g_wrap
-       test every time in the j loop. */
-    v4_u_l = v4_u; v4_u = v4_u_r;
-    v4_v_l = v4_v; v4_v = v4_v_r;
-    if (g_wrap) {
-      /* The 4 cells to the "right" are the first 4 in this row */
-      v4_u_r = *((V4F4 *)&INDEX(u,i,0));
-      v4_v_r = *((V4F4 *)&INDEX(v,i,0));
-    } else {
-      /* just leave them alone, retaining the rightmost 4 values in this row, which were loaded on the last iteration
-         through the loop */
-    }
-    v4_uvv = v4MUL(v4_u,v4MUL(v4_v,v4_v));
-    *v_dubase = v4SUB(v4MUL(v4_Du,
-         NABLA_5PT(v4_u, v4RAISE(v4_u,v4_u_l), v4LOWER(v4_u,v4_u_r), *v_ub_prev, *v_ub_next)),
-                           v4SUB(v4_uvv,v4MUL(v4_F,v4SUB(v4SPLAT(1.0f),v4_u))));
-    *v_dvbase = v4ADD(v4MUL(v4_Dv,
-         NABLA_5PT(v4_v, v4RAISE(v4_v,v4_v_l), v4LOWER(v4_v,v4_v_r), *v_vb_prev, *v_vb_next)),
-                           v4SUB(v4_uvv,v4MUL(v4ADD(v4_F,v4_k),v4_v)));
-  } // End of scan per row
-
-  // First thread interlock goes here
-  if (interlock) { DICEK_CH_SYNC }
-  if (interlock) { DICEK_CH_BEGIN }
-
-  {
-  // effect change
-    for(long i = start_row; i < end_row; i++) {
-      v_ubase = ((V4F4 *) (&INDEX(u,i,0)));
-      v_vbase = ((V4F4 *) (&INDEX(v,i,0)));
-      v_dubase = ((V4F4 *) (&INDEX(du,i,0)));
-      v_dvbase = ((V4F4 *) (&INDEX(dv,i,0)));
-      for(long j = 0; j < g_width; j+=VECSIZE) {
-        // u[i][j] = u[i][j] + speed * du[i][j];
-        *v_ubase = v4ADD(v4MUL(v4_speed, *v_dubase), *v_ubase); v_ubase++; v_dubase++;
-        // v[i][j] = v[i][j] + speed * dv[i][j];
-        *v_vbase = v4ADD(v4MUL(v4_speed, *v_dvbase), *v_vbase); v_vbase++; v_dvbase++;
-      }
-    }
-  }
-
-  // second thread interlock goes here
-  if (interlock) { DICEK_CH_SYNC }
-
-  } // End of scan per iteration
-}
-#endif
-
-#ifdef COMPUTE_PRE_THREADS
-
-/* This is the old version of compute() that used all "assembly-language" syntax */
-
-void compute(float *u, float *v, float *du, float *dv,
-             float D_u,float D_v,float F,float k,float speed,
-             int parameter_space)
-{
-#ifndef HWIV_HAVE_V4F4
-  fprintf(stdout, "Did not get vector macros from HWIV\n");
-  exit(-1);
-#endif
-  V4F4 v4_speed; // vectorized version of speed scalar
-  V4F4 v4_F; // vectorized version of F scalar
-  V4F4 v4_k; // vectorized version of k scalar
-  HWIV_4F4_ALIGNED talign; // used by FILL_4F4
-  V4F4 v4_u; V4F4 v4_du;
-  V4F4 v4_v; V4F4 v4_dv;
-  HWIV_INIT_MUL0_4F4; // used by MUL (on targets that need it)
-  HWIV_INIT_MTMP_4F4; // used by MADD (on targets that need it)
-  HWIV_INIT_FILL;     // used by FILL
-  HWIV_INIT_RLTMP_4F4; // used by RAISE and LOWER
-  V4F4 v4_tmp;
-  V4F4 v4_Du;
-  V4F4 v4_Dv;
-  V4F4 v4_nabla_u;
-  V4F4 v4_nabla_v;
-  V4F4 v4_1;
-  V4F4 v4_4;
-  const float k_min=0.045f, k_max=0.07f, F_min=0.01f, F_max=0.09f;
-  float k_diff;
-  V4F4 v4_kdiff;
-  float * ubase;
-  float * ub_prev;
-  float * ub_next;
-  float * vbase; float * vb_prev; float * vb_next;
-  float * dubase; float * dvbase;
-
-  V4F4 v4_u_l;
-  V4F4 v4_u_r;
-  V4F4 v4_v_l;
-  V4F4 v4_v_r;
-
-  //F_diff = (F_max-F_min)/g_width;
-  k_diff = (k_max-k_min)/g_height;
-
-  // Initialize our vectorized scalars
-  HWIV_SPLAT_4F4(v4_speed, speed);
-  HWIV_SPLAT_4F4(v4_F, F);
-  HWIV_SPLAT_4F4(v4_k, k);
-  HWIV_SPLAT_4F4(v4_Du, D_u);
-  HWIV_SPLAT_4F4(v4_Dv, D_v);
-  HWIV_SPLAT_4F4(v4_1, 1.0);
-  HWIV_SPLAT_4F4(v4_4, 4.0);
-  HWIV_FILL_4F4(v4_kdiff, 0, -k_diff, -2*k_diff, -3*k_diff);
-
-  // Scan per row
-  for(int i = 0; i < g_height; i++) {
-    int iprev,inext;
-    int j2;
-
-    if (g_wrap) {
-      iprev = (i+g_height-1) % g_height;
-      inext = (i+1) % g_height;
-    } else {
-      iprev = max(i-1, 0);
-      inext = min(i+1, g_height-1);
-    }
-    /* Get pointers to beginning of rows for each of the grids. We access
-       3 rows each for u and v, and 1 row each for du and dv. */
-    ubase = &INDEX(u,i,0);
-    ub_prev = &INDEX(u,iprev,0);
-    ub_next = &INDEX(u,inext,0);
-    vbase = &INDEX(v,i,0);
-    vb_prev = &INDEX(v,iprev,0);
-    vb_next = &INDEX(v,inext,0);
-    dubase = &INDEX(du,i,0);
-    dvbase = &INDEX(dv,i,0);
-
-    if (parameter_space) {
-      // set F for this row (ignore the provided value)
-      F = F_min + (g_height-i-1) * (F_max-F_min)/g_width;
-      HWIV_SPLAT_4F4(v4_F, F);
-    }
-
-    /* Pre-load the first two blocks of data we need, which are the "center"
-       and "right" blocks from the end of the row (as if we have just wrapped
-       around from the end of the row back to the beginning) */
-    j2 = g_wrap ? (g_width-4) : 0;
-    HWIV_LOAD_4F4(v4_u, ubase+j2);
-    HWIV_LOAD_4F4(v4_u_r, ubase);
-    HWIV_LOAD_4F4(v4_v, vbase+j2);
-    HWIV_LOAD_4F4(v4_v_r, vbase);
-
-    // Scan per column in steps of vector width
-    for(int j = 0; j < g_width; j+=4) {
-      if (g_wrap) {
-        j2 = (j+4) % g_width;
-      } else {
-        j2 = min(j+4, g_height-4);
-      }
-
-      HWIV_COPY_4F4(v4_u_l, v4_u);
-      HWIV_COPY_4F4(v4_v_l, v4_v);
-      HWIV_COPY_4F4(v4_u, v4_u_r);
-      HWIV_COPY_4F4(v4_v, v4_v_r);
-      HWIV_LOAD_4F4(v4_u_r, ubase+j2);
-      HWIV_LOAD_4F4(v4_v_r, vbase+j2);
-
-      if (parameter_space) {
-        // set k for this column (ignore the provided value)
-        k = k_min + (g_width-j-1)*k_diff;
-        // k decreases by k_diff each time j increases by 1, so this vector
-        // needs to contain 4 different k values.
-        HWIV_SPLAT_4F4(v4_tmp, k);
-        HWIV_ADD_4F4(v4_k, v4_tmp, v4_kdiff);
-      }
-
-      // compute the Laplacians of u and v. "nabla" is the name of the
-      // "upside down delta" symbol used for the Laplacian in equations
-
-      /* Scalar code is:
-         nabla_u = u[i][jprev]+u[i][jnext]+u[iprev][j]+u[inext][j] - 4*uval; */
-      HWIV_RAISE_4F4(v4_nabla_u, v4_u, v4_u_l);
-
-      HWIV_LOWER_4F4(v4_tmp, v4_u, v4_u_r);
-      HWIV_ADD_4F4(v4_nabla_u, v4_nabla_u, v4_tmp);
-
-      // Now we add in the "up" and "down" neighbors
-      HWIV_LOAD_4F4(v4_tmp, ub_prev+j);
-      HWIV_ADD_4F4(v4_nabla_u, v4_nabla_u, v4_tmp);
-      HWIV_LOAD_4F4(v4_tmp, ub_next+j);
-      HWIV_ADD_4F4(v4_nabla_u, v4_nabla_u, v4_tmp);
-
-      // Now we compute -(4*u-neighbors)  = neighbors - 4*u
-      HWIV_NMSUB_4F4(v4_nabla_u, v4_4, v4_u, v4_nabla_u);
-
-      // Same thing all over again for the v's
-      HWIV_RAISE_4F4(v4_nabla_v, v4_v, v4_v_l);
-      HWIV_LOWER_4F4(v4_tmp, v4_v, v4_v_r);
-      HWIV_ADD_4F4(v4_nabla_v, v4_nabla_v, v4_tmp);
-      HWIV_LOAD_4F4(v4_tmp, vb_prev+j);
-      HWIV_ADD_4F4(v4_nabla_v, v4_nabla_v, v4_tmp);
-      HWIV_LOAD_4F4(v4_tmp, vb_next+j);
-      HWIV_ADD_4F4(v4_nabla_v, v4_nabla_v, v4_tmp);
-      HWIV_NMSUB_4F4(v4_nabla_v, v4_4, v4_v, v4_nabla_v);
-
-      // compute the new rate of change of u and v
-
-      /* Scalar code is:
-           du[i][j] = D_u * nabla_u - uval*vval*vval + F*(1-uval);
-         We treat it as:
-                      D_u * nabla_u - (uval*vval*vval - (-(F*uval-F)) ) */
-
-      HWIV_NMSUB_4F4(v4_tmp, v4_F, v4_u, v4_F);        // -(F*u-F) = F-F*u = F(1-u)
-      HWIV_MUL_4F4(v4_dv, v4_v, v4_v);                 // v^2
-      HWIV_MSUB_4F4(v4_tmp, v4_u, v4_dv, v4_tmp);      // u*v^2 - F(1-u)
-      HWIV_MSUB_4F4(v4_du, v4_Du, v4_nabla_u, v4_tmp); // D_u*nabla_u - (u*v^2 - F(1-u))
-                                                       // = D_u*nabla_u - u*v^2 + F(1-u)
-      HWIV_SAVE_4F4(dubase+j, v4_du);
-
-      /* dv formula is similar:
-           dv[i][j] = D_v * nabla_v + uval*vval*vval - (F+k)*vval;
-         We treat it as:
-                      D_v * nabla_v + uval*vval*vval - (F*vval + k*vval); */
-      HWIV_MUL_4F4(v4_tmp, v4_k, v4_v);                // k*v
-      HWIV_MADD_4F4(v4_tmp, v4_F, v4_v, v4_tmp);       // F*v+k*v = (F+k)v
-                                                       // v^2 is still in v4_dv
-      HWIV_MSUB_4F4(v4_tmp, v4_u, v4_dv, v4_tmp);      // u*v^2 - (F+k)v
-      HWIV_MADD_4F4(v4_dv, v4_Dv, v4_nabla_v, v4_tmp); // D_v*nabla_v + u*v^2 - (F+k)v
-      HWIV_SAVE_4F4(dvbase+j, v4_dv);
-    }
-  }
-
-  // effect change
-    for(int i = 0; i < g_height; i++) {
-      ubase = &INDEX(u,i,0);
-      vbase = &INDEX(v,i,0);
-      dubase = &INDEX(du,i,0);
-      dvbase = &INDEX(dv,i,0);
-      for(int j = 0; j < g_width; j+=4) {
-        // u[i][j] = u[i][j] + speed * du[i][j];
-        HWIV_LOAD_4F4(v4_u, ubase+j);               // get u
-        HWIV_LOAD_4F4(v4_du, dubase+j);             // get du
-        HWIV_MADD_4F4(v4_u, v4_speed, v4_du, v4_u); // speed*du + u
-        HWIV_SAVE_4F4(ubase+j, v4_u);               // write it back
-
-        // v[i][j] = v[i][j] + speed * dv[i][j];
-        HWIV_LOAD_4F4(v4_v, vbase+j);
-        HWIV_LOAD_4F4(v4_dv, dvbase+j);
-        HWIV_MADD_4F4(v4_v, v4_speed, v4_dv, v4_v);
-        HWIV_SAVE_4F4(vbase+j, v4_v);
-      }
-    }
-}
-
-#endif
diff --git a/SpeedComparisons/GrayScott_HWIVector/hwi_vector.h b/SpeedComparisons/GrayScott_HWIVector/hwi_vector.h
deleted file mode 100644
index b3751a096..000000000
--- a/SpeedComparisons/GrayScott_HWIVector/hwi_vector.h
+++ /dev/null
@@ -1,432 +0,0 @@
-/*
-  
-  hwi_vector.h                                                                                           ^u 140 ^x f
-
-This is a hardware-independent vector library. It enables vector SIMD code to be written, compiled and run on non-vector
-machines. Once the code is tested and working, a compile-time macro can be changed, and a recompile causes the program
-to use actual vector instructions.
-
-In the simplest case, that is all that is needed. In real applications there are usually several more steps:
-
-  * The application developer decides what base hardware platform to compile for. This might be the oldest CPU the
-    company will support, such as a Pentium 4 HT.
-
-  * The compiler defines certain flags, such as __i386__ and __SSE2__ that can be tested by an #ifdef
-
-  * The source code can set flags of its own, such as HWIV_EMULATE or HWIV_WANT_V4F4. This might be done in order to
-    create two versions of a calculation routine (one that uses vector instructions and one that does not)
-
-  * The program is built from one or more compilations of the same source. This might be done in order to create a
-    "universal binary" capable of being copied to and run on a variety of computer products. A typical example is a
-    program file that contains both a 32-bit and a 64-bit version, and the operating system loads whichever one is
-    appropriate when the program is launched.
-
-  * At run-time, the program tests for the presence of vector instructions using the CPUID instruction or its
-    equivalent on non Intel CPUs.
-
-  * At run-time, based on the CPUID test, the program transfers control to one or another of the calculation
-    subroutines depending on which vector instructions are actually available.
-
-*/
-
-// First, honor the user's request to use emulation
-#ifdef HWIV_EMULATE
-# define HWIV_V4F4_EMULATED
-#endif
-
-// Next, find out if the compiler will give us SSE2 intrinsics
-#ifndef HWIV_V4F4_EMULATED
-# if defined(__SSE2__)
-#   define HWIV_V4F4_SSE2
-# endif
-  // Workaround because Visual Studio doesn't seem to set its _M_IX86_FP flag, 
-  // or give us any indication what level of SSE support is available.
-  // So we just assume SSE2 is available
-# if defined(_M_X64) || defined(_M_IX86)
-#  define HWIV_V4F4_SSE2
-# endif
-#endif
-
-
-// Finally, fall back to emulated if no hardware option is available
-#ifndef HWIV_V4F4_SSE2
-# ifndef HWIV_V4F4_EMULATED
-#   define HWIV_V4F4_EMULATED
-# endif
-#endif
-
-
-// See is client wants V4F4
-#ifdef HWIV_WANT_V4F4
-
-// Okay, now see how we should create the macros for V4F4
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//  #include<stdio.h>                                                                                                //
-//  /*                                                     */                    /*                             */   //
-//  /*                                                     */                    /*                             */   //
-//  int                 main()  {char      a[]     ={     'T',    'h','i',     's',32,     'p','r'        ,'o','g'   //
-//  ,'r','a','m',      32,   'j',   'u',   's',    't'    ,32           ,'d',    'o',   'e'     ,'s'    ,32    ,'i'  //
-//  ,'t'              ,32    ,'t'    ,'h'  ,'e'   ,32,    'h'    ,97,'r','d'   ,32,    'w',97,121,33,   32,    40,   //
-//  68,               'o',   'n',    39,   't'    ,32     ,'y'   ,'o',   117,    32    ,'t'             ,'h'   ,'i'  //
-//  ,'n'               /*     Xy     =a     +3     +n      ++     ;a=     b-     (*     x/z              );     if   //
-//  (Xy-++n<(z+*x))z  =b;a   +b,     z+=     x*/,107 ,    63,63    ,63,41,'\n'   ,00};    puts(a);}      /*.RPM.*/   //
-//                                                                                                                   //
-//        Emulated versions of the V4F4 macros. (These also serve as documentation for what each macro does)         //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-# ifdef HWIV_V4F4_EMULATED
-
-/* Scalar code to emulate the V4F4 vector model */
-#define HWIV_HAVE_V4F4
-
-/* The vector register type */
-typedef float V4F4[4];
-
-/* Aligned memory suitable for load to / store from the vector register */
-typedef float HWIV_4F4_ALIGNED[4];
-
-/* LOAD_4F4: dst is a vector of 4 floats. src is a pointer to an array
-   of floats in memory. This opcode loads 4 consecutive floats from the
-   given location into the vector. The first float from memory will be loaded
-   into element 0 of the vector, the 2nd into element 1, and so on. */
-#define HWIV_LOAD_4F4(dst, src)  { (dst)[0]=(src)[0]; (dst)[1]=(src)[1]; \
-                                   (dst)[2]=(src)[2]; (dst)[3]=(src)[3]; }
-
-#define HWIV_LOADU_4F4(dst, src) { (dst)[0]=(src)[0]; (dst)[1]=(src)[1]; \
-                                   (dst)[2]=(src)[2]; (dst)[3]=(src)[3]; }
-
-/* LOADO_4F4: dst is a vector of 4 floats. src is a pointer to an array
-   of floats in memory. offset is a byte offset, which must be a multiple
-   of sizeof(float).
-     This opcode loads 4 consecutive floats from the given location plus
-   offset into the vector. The first float from memory will be loaded into
-   element 0 of the vector, the 2nd into element 1, and so on. */
-#define HWIV_LOADO_4F4(dst, src, offset)  { (dst)[0]=(src)[(offset)/4]; \
-                                            (dst)[1]=(src)[(offset)/4+1]; \
-                                            (dst)[2]=(src)[(offset)/4+2]; \
-                                            (dst)[3]=(src)[(offset)/4+3]; }
-
-/* COPY_4F4: dst and src are each a vector of 4 floats. This opcode copies
-   the contents of the source vector into the destination vector. */
-#define HWIV_COPY_4F4(dst, src)  HWIV_LOAD_4F4((dst), (src))
-
-/* FILL_4F4: dst is a vector of 4 floats. sc0, sc1, sc2, and sc3 ("scalars")
-   are each a float. tmp is a (float *) pointing to memory of sufficient
-   size to hold four floats, and aligned to a 16-byte boundary. Use the
-   HWIV_4F4_ALIGNED typedef to declare a suitable float[4] which can
-   be passed as tmp to this and other similar macros.
-     This opcode loads the 4 scalar values into the vector. The sc0 will be
-   loaded into element 0 of the vector, sc1 into element 1, and so on. */
-#define HWIV_FILL_4F4(dst, sc0, sc1, sc2, sc3) { \
-            (dst)[0]=(sc0); (dst)[1]=(sc1); (dst)[2]=(sc2); (dst)[3]=(sc3); }
-#define HWIV_INIT_FILL /* nop */
-
-#define HWIV_SPLAT_4F4(dst, s) HWIV_FILL_4F4((dst), (s), (s), (s), (s))
-
-/* SAVE_4F4: dst is a pointer to an array of floats in memory. src is a
-   vector of 4 floats. 
-     This opcode stores the 4 floats in the source vector into 4 consecutive
-   float-sized blocks of memory (i.e. 16 consecutive bytes, 4 bytes per
-   float) beginning at the given destination location. Element 0 of the
-   vector will be stored into the first 4 bytes in memory, element 1 into
-   the next 4 bytes of memory, and so on. */
-#define HWIV_SAVE_4F4(dst, src)  memcpy((void *) (dst), (void *) (src), 16);
-
-/* SAVEO_4F4: dst is a pointer to an array of floats in memory. src is a
-   vector of 4 floats. offset is a byte offset, which must be a multiple
-   of sizeof(float). 
-     This opcode stores the 4 floats in the source vector into 4 consecutive
-   float-sized blocks of memory (i.e. 16 consecutive bytes, 4 bytes per
-   float) beginning at the given destination location plus offset. Element
-   0 of the vector will be stored into the first 4 bytes in memory, element
-   1 into the next 4 bytes of memory, and so on. */
-#define HWIV_SAVEO_4F4(dst, offset, src) { (dst)[(offset)/4]=(src)[0]; \
-                                         (dst)[(offset)/4+1]=(src)[1]; \
-                                         (dst)[(offset)/4+2]=(src)[2]; \
-                                         (dst)[(offset)/4+3]=(src)[3]; }
-
-/* ADD_4F4: dst, a, and b are each a vector of 4 floats.
-     This opcode adds each of the components of a to the corresponding
-   component of b and puts the result into dst. */
-#define HWIV_ADD_4F4(dst, a, b) { (dst)[0]=a[0]+b[0]; (dst)[1]=a[1]+b[1]; \
-                                (dst)[2]=a[2]+b[2]; (dst)[3]=a[3]+b[3]; }
-
-/* SUB_4F4: dst, a, and b are each a vector of 4 floats.
-     This opcode adds each of the components of a to the corresponding
-   component of b and puts the result into dst. */
-#define HWIV_SUB_4F4(dst, a, b) { (dst)[0]=a[0]-b[0]; (dst)[1]=a[1]-b[1]; \
-                                (dst)[2]=a[2]-b[2]; (dst)[3]=a[3]-b[3]; }
-
-#define HWIV_INIT_MUL0_4F4 /* nop */
-
-/* MUL_4F4: dst, a, and b are each a vector of 4 floats. v0 is a variable
-   declared with the HWIV_INIT_MUL0_4F4 macro.
-     This opcode multiplies each of the components of a to the corresponding
-   component of b and puts the result into dst. The varible v0 is used on
-   hardware that has no 2-argument multiply operation. */
-#define HWIV_MUL_4F4(dst, a, b) \
-                             { (dst)[0]=a[0]*b[0]; (dst)[1]=a[1]*b[1]; \
-                                (dst)[2]=a[2]*b[2]; (dst)[3]=a[3]*b[3]; }
-
-#define HWIV_INIT_MTMP_4F4 /* nop */
-
-/* MADD_4F4: dst, a, b, and c are each a vector of 4 floats. t is a variable
-   declared with the HWIV_INIT_MTMP_4F4 macro.
-     This opcode multiplies each of the components of a to the corresponding
-   component of b, then adds the corresponding component of c, and puts the
-   result into dst. The varible t is used on hardware that has no 3-argument
-   multiply-add operation. */
-#define HWIV_MADD_4F4(dst, a, b, c)  { (dst)[0]=a[0]*b[0] + c[0]; \
-                                        (dst)[1]=a[1]*b[1] + c[1]; \
-                                        (dst)[2]=a[2]*b[2] + c[2]; \
-                                        (dst)[3]=a[3]*b[3] + c[3]; }
-
-#define HWIV_MSUB_4F4(dst, a, b, c)  { (dst)[0]=a[0]*b[0] - c[0]; \
-                                           (dst)[1]=a[1]*b[1] - c[1]; \
-                                           (dst)[2]=a[2]*b[2] - c[2]; \
-                                           (dst)[3]=a[3]*b[3] - c[3]; }
-
-/* NMSUB_4F4: dst, a, b, and c are each a vector of 4 floats. t is a variable
-   declared with the HWIV_INIT_MTMP_4F4 macro.
-     This opcode multiplies each of the components of a to the corresponding
-   component of b, then subtracts that product from the corresponding
-   component of c, then puts the result into dst. The varible t is used on
-   hardware that has no 3-argument multiply-add operation. */
-#define HWIV_NMSUB_4F4(dst, a, b, c) { (dst)[0]=c[0] - a[0]*b[0]; \
-                                        (dst)[1]=c[1] - a[1]*b[1]; \
-                                        (dst)[2]=c[2] - a[2]*b[2]; \
-                                        (dst)[3]=c[3] - a[3]*b[3]; }
-
-// Declare this if you are doing any raise or lower operation
-#define HWIV_INIT_RLTMP_4F4 /* nop */
-
-/* RAISE_4F4: dst, src, extra, and tmp are each a vector of 4 floats.
-     This opcode "raises" three of the values from src to the next-higher
-   element of dst. Element 0 of dst is filled with the value from element
-   3 of "extra". The varible t is used on hardware that requires the
-   result to be computed in two pieces and then assembled via a blend
-   operation. ("VSHR_4F4" in old macros) */
-#define HWIV_RAISE_4F4(dst, src, extra) \
-                                      { dst[3]=src[2]; dst[2]=src[1]; \
-                                         dst[1]=src[0]; dst[0]=extra[3]; }
-
-/* LOWER_4F4: dst, src, extra, and tmp are each a vector of 4 floats.
-     This opcode "lowers" three of the values from src to the next-lower
-   element of dst. Element 3 of dst is filled with the value from element
-   0 of "extra". The varible t is used on hardware that requires the
-   result to be computed in two pieces and then assembled via a blend
-   operation. ("VSHL_4F4" in old macros) */
-#define HWIV_LOWER_4F4(dst, src, extra) \
-                                      { dst[0]=src[1]; dst[1]=src[2]; \
-                                         dst[2]=src[3]; dst[3]=extra[0]; }
-
-
-// We also define a small set of macros for FORTRAN-style code. Using these you can build up expressions like
-//
-//    a = v4ADD(v4MUL(b,c),d);  /*  a = b*c + d;  */
-//
-// These macros do not form a complete solution, you still need things like LOAD and SAVE to do any real work.
-
-#define v4LOADU(a) {(a)[0],(a)[1],(a)[2],(a)[3]}
-#define v4ADD(a,b) {(a)[0]+(b)[0],(a)[1]+(b)[1],(a)[2]+(b)[2],(a)[3]+(b)[3]}
-#define v4SUB(a,b) {(a)[0]-(b)[0],(a)[1]-(b)[1],(a)[2]-(b)[2],(a)[3]-(b)[3]}
-#define v4MUL(a,b) {(a)[0]*(b)[0],(a)[1]*(b)[1],(a)[2]*(b)[2],(a)[3]*(b)[3]}
-#define v4SET(v0,v1,v2,v3) {(v0),(v1),(v2),(v3)}
-#define v4SPLAT(a) {(a),(a),(a),(a)}
-#define v4ROUP(a) {(a)[3],(a)[0],(a)[1],(a)[2]}
-#define v4RODN(a) {(a)[1],(a)[2],(a)[3],(a)[0]}
-#define v4RAISE(a, new) {(new)[3],(a)[0],(a)[1],(a)[2]}
-#define v4LOWER(a, new) {(a)[1],(a)[2],(a)[3],(new)[0]}
-
-# endif
-/* - - - - - - - - - - - - - - - - - - - - - End of the EMULATED section  - - - - - - - - - - - - - - - - - - - - - - */
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//                             @@@@                       @@@@                     @@@@                              //
-//                             @@@@                       @@@@                     @@@@                              //
-//                             """"                       @@@@                     @@@@                              //
-//                             eeee    eeee  ,e@@e..   eee@@@@eee                  @@@@                              //
-//                             @@@@    @@@@@@@@@@@@@@. @@@@@@@@@@                  @@@@                              //
-//                             @@@@    @@@@f'    `@@@@    @@@@                     @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@        ,e@@@e.      @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@     e@@@@@@@@@@@e   @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@   .@@@@'     `@@@@i @@@@                              //
-//                             @@@@    @@@@       @@@@    @@@@kee@@@@eeeeeeeee@@@@ @@@@                              //
-//                             @@@@    @@@@       @@@@    `@@@@@@@@@@@@@@@@@@@@@@@@@@@@  (R)                         //
-//                                                               @@@@.                                               //
-//                                                               `@@@@e.    .eeee-                                   //
-//                                                                 *@@@@@@@@@@@*                                     //
-//                                                                    "*@@@@*"                                       //
-//                                                                                                                   //
-//             Versions of the V4F4 macros for the Intel SSE2 (or later) 128-bit vector instruction set              //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-# ifdef HWIV_V4F4_SSE2
-
-/* SSE3 implementation of the V4F4 vector model */
-#define HWIV_HAVE_V4F4
-
-# if (defined(HWIV_USE_IMMINTRIN) || defined (__AVX__))
-#   include <immintrin.h>
-# else
-#   include <xmmintrin.h>
-#   include <emmintrin.h>
-# endif
-
-
-typedef __m128 V4F4;
-
-#ifdef _WIN32
-# define ALIGNED_16 __declspec( align( 16 ) )
-#else
-# define ALIGNED_16 __attribute__((aligned (16)))
-#endif
-
-typedef float ALIGNED_16 HWIV_4F4_ALIGNED[4];
-
-#define HWIV_LOAD_4F4(dst, src)        (dst) = _mm_load_ps(src)
-#define HWIV_LOADU_4F4(dst, src)       (dst) = _mm_loadu_ps(src)
-#define HWIV_LOADO_4F4(dst, src, offset) \
-                                       (dst) = _mm_load_ps((src)+(offset)/4)
-#define HWIV_COPY_4F4(dst, src)        (dst) = (src)
-
-#define HWIV_FILL_4F4(dst, sc0, sc1, sc2, sc3) \
-            { HWIV_fill_4F4[0]=(sc0); HWIV_fill_4F4[1]=(sc1); \
-               HWIV_fill_4F4[2]=(sc2); HWIV_fill_4F4[3]=(sc3); \
-               HWIV_LOAD_4F4(dst, HWIV_fill_4F4); }
-
-#define HWIV_INIT_FILL float ALIGNED_16 HWIV_fill_4F4[4];
-
-
-#define HWIV_SPLAT_4F4(dst, s)         (dst) = _mm_set1_ps(s)
-
-#define HWIV_SAVE_4F4(dst, src)        _mm_store_ps((dst), (src))
-
-#define HWIV_SAVEO_4F4(dst, offset, src) \
-                                        _mm_store_ps((dst)+(offset)/4, (src))
-
-#define HWIV_ADD_4F4(dst, a, b) (dst) = _mm_add_ps((a), (b))
-
-#define HWIV_SUB_4F4(dst, a, b) (dst) = _mm_sub_ps((a), (b))
-
-// For INIT_MUL0, on Intel we do nothing because Intel actually has a
-// 2-operand multiply operation.
-#define HWIV_INIT_MUL0_4F4 /* nop */
-
-// For INIT_MTMP, on Intel SSE2 we need to declare a variable because
-// Intel SSE2 has no FMA (fused multiply-add) operations (this is
-// expected to come wth AVX2 on Haskell in 2013)
-#define HWIV_INIT_MTMP_4F4     V4F4 HWIV_mtmp_4F4 = _mm_setzero_ps()
-
-#define HWIV_MUL_4F4(dst, a, b)    (dst) = _mm_mul_ps((a), (b))
-
-#define HWIV_MADD_4F4(dst, a, b, c) { HWIV_mtmp_4F4 = _mm_mul_ps((a), (b)); \
-                                    (dst) = _mm_add_ps(HWIV_mtmp_4F4, (c)); }
-#define HWIV_MSUB_4F4(dst, a, b, c) { HWIV_mtmp_4F4 = _mm_mul_ps((a), (b)); \
-                                    (dst) = _mm_sub_ps(HWIV_mtmp_4F4, (c)); }
-#define HWIV_NMSUB_4F4(dst, a, b, c) { HWIV_mtmp_4F4 = _mm_mul_ps((a), (b)); \
-                                    (dst) = _mm_sub_ps((c), HWIV_mtmp_4F4); }
-
-#define HWIV_INIT_RLTMP_4F4 /* nop */
-
-
-/*
-HWIV_RODN_4F4 (ROtate DOwn) does a "downwards rotate" of a 4-element vector using the Intel VSHUFPS instruction
-(intrinsic _mm_shuffle_ps). If the input is {a,b,c,d} (with a being element 0) the result of the downwards rotate is
-{b,c,d,a} (with each element moving down tot he next-lower slot, except for a which rotates into the top position).
-
-  SRC1  {  x3 ,  x2 ,  x1 ,  x0  }
-  SRC2  {  y3 ,  y2 ,  y1 ,  y0  }
-  DEST  {  y0 ,  y3 ,  x2 ,  x1  }
-  imm8:    00    11    10    01      = 0x39
-
- */
-#define HWIV_RODN_4F4(dest, src)  (dest) = _mm_shuffle_ps((src), (src), 0x39)
-
-/*
-HWIV_ROUP_4F4 is an "upwards rotate": if the input is {a,b,c,d} (with a being element 0) the result is {d,a,b,c}.
-
-  SRC1  {  x3 ,  x2 ,  x1 ,  x0  }
-  SRC2  {  y3 ,  y2 ,  y1 ,  y0  }
-  DEST  {  y2 ,  y1 ,  x0 ,  x3  }
-  imm8:    10    01    00    11      = 0x93
-
- */
-#define HWIV_ROUP_4F4(dest, src)  (dest) = _mm_shuffle_ps((src), (src), 0x93)
-
-/*
-HWIV_RAISE_4F4 is an "upwards shift": if the input is {a,b,c,d} (with a being element 0) the result is {X,a,b,c} with
-the new element X coming from element 3 of the "new" argument.
-
-  new   {  x3 ,  x2 ,  x1 ,  x0  }
-  src   {  y3 ,  y2 ,  y1 ,  y0  }
-  dest  {  y0 ,  y0 ,  x3 ,  x3  }
-  imm8:    00    00    11    11      = 0x0F
-          src0  src0  new3  new3
-
-  dest  {  x3 ,  x2 ,  x1 ,  x0  }
-   src  {  y3 ,  y2 ,  y1 ,  y0  }
-  dest  {  y2 ,  y1 ,  x2 ,  x0  }
-  imm8:    10    01    10    00      = 0x98
-          src2  src1  src0  new3
-*/
-#define HWIV_RAISE_4F4(dest, src, new) { (dest) = _mm_shuffle_ps((new), (src), 0x0f); \
-                                         (dest) = _mm_shuffle_ps((dest), (src), 0x98);  }
-
-/*
-HWIV_LOWER_4F4 is an "downwards shift": if the input is {a,b,c,d} (with a being element 0) the result is {b,c,d,X} with
-the new element X coming from element 0 of the "new" argument.
-
-To accomplish a downwards shift we can just use _mm_move_ss to move a single scalar into the bottom position and then do
-a RODN (downwards rotate)
- */
-#define HWIV_LOWER_4F4(dest, src, new) {  (dest) = _mm_move_ss((src), (new));  \
-                                           HWIV_RODN_4F4(dest, dest); }
-
-// Here is the subset for FORTRAN-style code
-#define v4LOADU(a) _mm_loadu_ps(a)
-#define v4ADD(a,b) _mm_add_ps((a), (b))
-#define v4SUB(a,b) _mm_sub_ps((a), (b))
-#define v4MUL(a,b) _mm_mul_ps((a), (b))
-// in v4SET, note the reversal of argument order
-#define v4SET(v0,v1,v2,v3) _mm_set_ps((v3),(v2),(v1),(v0))
-#define v4SPLAT(a) _mm_set1_ps(a)
-#define v4ROUP(src) _mm_shuffle_ps((src), (src), 0x93)
-#define v4RODN(src) _mm_shuffle_ps((src), (src), 0x39)
-#define v4RAISE(src, new) _mm_shuffle_ps(_mm_shuffle_ps((new), (src), 0x0f), (src), 0x98)
-#define v4LOWER(src, new) _mm_shuffle_ps(_mm_move_ss((src),(new)), _mm_move_ss((src),(new)), 0x39)
-
-# endif
-/* - - - - - - - - - - - - - - - - - - - - - - End of the INTEL section - - - - - - - - - - - - - - - - - - - - - - - */
-
-
-
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-//                                                                                                                   //
-//                                                                                                                   //
-//                         /^^^^^\                                       /^^^^^\  _-^^^^^^/ TM                       //
-//                        /  ,-   )_-----_ --- ---- ----..----.  ---.---/  ,-   )/   ,---/                           //
-//                       /  /_)  //  __   )| |/   |/  _// .-   )/    _//  /_)  //   /                                //
-//                      /  //   //  / /  /|          / / (/___//   .^ /  //   /(   |                                 //
-//                     /  / ^^^'(   (/  / |   /|   _/ (  `----/   /  /  / ^^^' |   `---/                             //
-//                    /__/       \____-'  |__/ |__/    \____//___/  /__/        \_____/                              //
-//                                                                                                                   //
-//                                                                                                                   //
-//                Versions of the V4F4 macros for the PowerPC AltiVec 128-bit vector instruction set                 //
-//                                                                                                                   //
-///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
-/*
-
-The AltiVec instruction set first came with the "G4" (744x and 745x) processors from Motorola, then the "G5" (97x)
-series from IBM, the "Cell" 8-core CPU used in the Sony Playstation 3, and in IBM's POWER6 (and later) server CPUs.
-
-  Not yet implemented -- do we care about AltiVec? */
-
-#endif
diff --git a/SpeedComparisons/GrayScott_OpenCL/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenCL/CMakeLists.txt
deleted file mode 100644
index c62df02e1..000000000
--- a/SpeedComparisons/GrayScott_OpenCL/CMakeLists.txt
+++ /dev/null
@@ -1,29 +0,0 @@
-project(GrayScott_OpenCL)
-
-set(CMAKE_MODULE_PATH ${GrayScott_OpenCL_SOURCE_DIR})
-# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
-
-find_package(OpenCV REQUIRED)
-include_directories( ${OPENCV_INCLUDE_DIR})
-link_libraries( ${OpenCV_LIBS} )
-
-# only build the OpenCL version if OpenCL was found
-find_package ( OpenCL )
-if(OPENCL_FOUND)
-  include_directories ( ${OPENCL_INCLUDE_DIRS} )
-  link_libraries ( ${OPENCL_LIBRARIES} )
-
-  # tell the code where the .cl file will live
-  add_definitions(-DCL_SOURCE_DIR="${GrayScott_OpenCL_SOURCE_DIR}")
-
-  INCLUDE_DIRECTORIES( "../Display" )
-
-  add_executable(GrayScott_OpenCL
-    gray_scott_opencl.cpp 
-    grayscott_kernel.cl 
-    ../Display/display.cpp
-    ../Display/display.h
-    ../Display/defs.h
-  )
-endif()
-
diff --git a/SpeedComparisons/GrayScott_OpenCL/FindOpenCL.cmake b/SpeedComparisons/GrayScott_OpenCL/FindOpenCL.cmake
deleted file mode 100644
index fde90efae..000000000
--- a/SpeedComparisons/GrayScott_OpenCL/FindOpenCL.cmake
+++ /dev/null
@@ -1,79 +0,0 @@
-# - Try to find OpenCL
-# This module tries to find an OpenCL implementation on your system. It supports
-# AMD / ATI, Apple and NVIDIA implementations, but shoudl work, too.
-#
-# Once done this will define
-#  OPENCL_FOUND        - system has OpenCL
-#  OPENCL_INCLUDE_DIRS  - the OpenCL include directory
-#  OPENCL_LIBRARIES    - link these to use OpenCL
-#
-# WIN32 should work, but is untested
-
-FIND_PACKAGE( PackageHandleStandardArgs )
-
-SET (OPENCL_VERSION_STRING "0.1.0")
-SET (OPENCL_VERSION_MAJOR 0)
-SET (OPENCL_VERSION_MINOR 1)
-SET (OPENCL_VERSION_PATCH 0)
-
-IF (APPLE)
-
-  FIND_LIBRARY(OPENCL_LIBRARIES OpenCL DOC "OpenCL lib for OSX")
-  FIND_PATH(OPENCL_INCLUDE_DIRS OpenCL/cl.h DOC "Include for OpenCL on OSX")
-  FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS OpenCL/cl.hpp DOC "Include for OpenCL CPP bindings on OSX")
-
-ELSE (APPLE)
-
-	IF (WIN32)
-	
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h)
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp)
-	
-	    # The AMD SDK currently installs both x86 and x86_64 libraries
-	    # This is only a hack to find out architecture
-	    IF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86_64")
-	    ELSE (${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64")
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86")
-	    ENDIF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    FIND_LIBRARY(OPENCL_LIBRARIES OpenCL.lib ${OPENCL_LIB_DIR})
-	    
-	    GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-	    
-	    # On Win32 search relative to the library
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS "${_OPENCL_INC_CAND}")
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS "${_OPENCL_INC_CAND}")
-	
-	ELSE (WIN32)
-
-            # Unix style platforms
-            FIND_LIBRARY(OPENCL_LIBRARIES OpenCL
-              ENV LD_LIBRARY_PATH
-            )
-
-            GET_FILENAME_COMPONENT(OPENCL_LIB_DIR ${OPENCL_LIBRARIES} PATH)
-            GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-
-            # The AMD SDK currently does not place its headers
-            # in /usr/include, therefore also search relative
-            # to the library
-            FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-            FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-
-	ENDIF (WIN32)
-
-ENDIF (APPLE)
-
-FIND_PACKAGE_HANDLE_STANDARD_ARGS( OpenCL DEFAULT_MSG OPENCL_LIBRARIES OPENCL_INCLUDE_DIRS )
-
-IF( _OPENCL_CPP_INCLUDE_DIRS )
-	SET( OPENCL_HAS_CPP_BINDINGS TRUE )
-	LIST( APPEND OPENCL_INCLUDE_DIRS ${_OPENCL_CPP_INCLUDE_DIRS} )
-	# This is often the same, so clean up
-	LIST( REMOVE_DUPLICATES OPENCL_INCLUDE_DIRS )
-ENDIF( _OPENCL_CPP_INCLUDE_DIRS )
-
-MARK_AS_ADVANCED(
-  OPENCL_INCLUDE_DIRS
-)
-
diff --git a/SpeedComparisons/GrayScott_OpenCL/cl.hpp b/SpeedComparisons/GrayScott_OpenCL/cl.hpp
deleted file mode 100644
index 99b86a665..000000000
--- a/SpeedComparisons/GrayScott_OpenCL/cl.hpp
+++ /dev/null
@@ -1,4011 +0,0 @@
-/*******************************************************************************
- * Copyright (c) 2008-2010 The Khronos Group Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and/or associated documentation files (the
- * "Materials"), to deal in the Materials without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sublicense, and/or sell copies of the Materials, and to
- * permit persons to whom the Materials are furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice shall be included
- * in all copies or substantial portions of the Materials.
- *
- * THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
- * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
- * CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
- * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
- * MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
- ******************************************************************************/
-
-/*! \file
- *
- *   \brief C++ bindings for OpenCL 1.0 (rev 48) and OpenCL 1.1 (rev 33)    
- *   \author Benedict R. Gaster and Laurent Morichetti
- *   
- *   Additions and fixes from Brian Cole, March 3rd 2010.
- *   
- *   \version 1.1
- *   \date June 2010
- *
- *   Optional extension support
- *
- *         cl
- *         cl_ext_device_fission
- *				#define USE_CL_DEVICE_FISSION
- */
-
-/*! \mainpage
- * \section intro Introduction
- * For many large applications C++ is the language of choice and so it seems
- * reasonable to define C++ bindings for OpenCL.
- *
- *
- * The interface is contained with a single C++ header file \em cl.hpp and all
- * definitions are contained within the namespace \em cl. There is no additional
- * requirement to include \em cl.h and to use either the C++ or original C
- * bindings it is enough to simply include \em cl.hpp.
- *
- * The bindings themselves are lightweight and correspond closely to the
- * underlying C API. Using the C++ bindings introduces no additional execution
- * overhead.
- *
- * For detail documentation on the bindings see:
- *
- * The OpenCL C++ Wrapper API 1.1 (revision 04)
- *  http://www.khronos.org/registry/cl/specs/opencl-cplusplus-1.1.pdf
- *
- * \section example Example
- *
- * The following example shows a general use case for the C++
- * bindings, including support for the optional exception feature and
- * also the supplied vector and string classes, see following sections for
- * decriptions of these features.
- *
- * \code
- * #define __CL_ENABLE_EXCEPTIONS
- * 
- * #if defined(__APPLE__) || defined(__MACOSX)
- * #include <OpenCL/cl.hpp>
- * #else
- * #include <CL/cl.hpp>
- * #endif
- * #include <cstdio>
- * #include <cstdlib>
- * #include <iostream>
- * 
- *  const char * helloStr  = "__kernel void "
- *                           "hello(void) "
- *                           "{ "
- *                           "  "
- *                           "} ";
- * 
- *  int
- *  main(void)
- *  {
- *     cl_int err = CL_SUCCESS;
- *     try {
- *
- *       std::vector<cl::Platform> platforms;
- *       cl::Platform::get(&platforms);
- *       if (platforms.size() == 0) {
- *           std::cout << "Platform size 0\n";
- *           return -1;
- *       }
- *
- *       cl_context_properties properties[] = 
- *          { CL_CONTEXT_PLATFORM, (cl_context_properties)(platforms[0])(), 0};
- *       cl::Context context(CL_DEVICE_TYPE_CPU, properties); 
- * 
- *       std::vector<cl::Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
- * 
- *       cl::Program::Sources source(1,
- *           std::make_pair(helloStr,strlen(helloStr)));
- *       cl::Program program_ = cl::Program(context, source);
- *       program_.build(devices);
- * 
- *       cl::Kernel kernel(program_, "hello", &err);
- * 
- *       cl::Event event;
- *       cl::CommandQueue queue(context, devices[0], 0, &err);
- *       queue.enqueueNDRangeKernel(
- *           kernel, 
- *           cl::NullRange, 
- *           cl::NDRange(4,4),
- *           cl::NullRange,
- *           NULL,
- *           &event); 
- * 
- *       event.wait();
- *     }
- *     catch (cl::Error err) {
- *        std::cerr 
- *           << "ERROR: "
- *           << err.what()
- *           << "("
- *           << err.err()
- *           << ")"
- *           << std::endl;
- *     }
- * 
- *    return EXIT_SUCCESS;
- *  }
- * 
- * \endcode
- *
- */
-#ifndef CL_HPP_
-#define CL_HPP_
-
-#ifdef _WIN32
-#include <windows.h>
-#include <malloc.h>
-#if defined(USE_DX_INTEROP)
-#include <CL/cl_d3d10.h>
-#endif
-#endif // _WIN32
-
-// 
-#if defined(USE_CL_DEVICE_FISSION)
-#include <CL/cl_ext.h>
-#endif
-
-#if defined(__APPLE__) || defined(__MACOSX)
-#include <OpenGL/OpenGL.h>
-#include <OpenCL/opencl.h>
-#else
-#include <GL/gl.h>
-#include <CL/opencl.h>
-#endif // !__APPLE__
-
-#if !defined(CL_CALLBACK)
-#define CL_CALLBACK
-#endif //CL_CALLBACK
-
-#include <utility>
-
-#if !defined(__NO_STD_VECTOR)
-#include <vector>
-#endif
-
-#if !defined(__NO_STD_STRING)
-#include <string>
-#endif 
-
-#if defined(linux) || defined(__APPLE__) || defined(__MACOSX)
-# include <alloca.h>
-#endif // linux
-
-#include <cstring>
-
-/*! \namespace cl
- *
- * \brief The OpenCL C++ bindings are defined within this namespace.
- *
- */
-namespace cl {
-
-#define __INIT_CL_EXT_FCN_PTR(name) \
-    if(!pfn_##name) { \
-        pfn_##name = (PFN_##name) \
-            clGetExtensionFunctionAddress(#name); \
-        if(!pfn_##name) { \
-        } \
-    }
-
-class Program;
-class Device;
-class Context;
-class CommandQueue;
-class Memory;
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-#include <exception>
-/*! \class Error
- * \brief Exception class
- */
-class Error : public std::exception
-{
-private:
-    cl_int err_;
-    const char * errStr_;
-public:
-    /*! Create a new CL error exception for a given error code
-     *  and corresponding message.
-     */
-    Error(cl_int err, const char * errStr = NULL) : err_(err), errStr_(errStr)
-    {}
-
-    ~Error() throw() {}
-
-    /*! \brief Get error string associated with exception
-     *
-     * \return A memory pointer to the error message string.
-     */
-    virtual const char * what() const throw ()
-    {
-        if (errStr_ == NULL) {
-            return "empty";
-        }
-        else {
-            return errStr_;
-        }
-    }
-
-    /*! \brief Get error code associated with exception
-     *
-     *  \return The error code.
-     */
-    const cl_int err(void) const { return err_; }
-};
-
-#define __ERR_STR(x) #x
-#else
-#define __ERR_STR(x) NULL
-#endif // __CL_ENABLE_EXCEPTIONS
-
-//! \cond DOXYGEN_DETAIL
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#define __GET_DEVICE_INFO_ERR               __ERR_STR(clgetDeviceInfo)
-#define __GET_PLATFORM_INFO_ERR             __ERR_STR(clGetPlatformInfo)
-#define __GET_DEVICE_IDS_ERR                __ERR_STR(clGetDeviceIDs)
-#define __GET_PLATFORM_IDS_ERR              __ERR_STR(clGetPlatformIDs)
-#define __GET_CONTEXT_INFO_ERR              __ERR_STR(clGetContextInfo)
-#define __GET_EVENT_INFO_ERR                __ERR_STR(clGetEventInfo)
-#define __GET_EVENT_PROFILE_INFO_ERR        __ERR_STR(clGetEventProfileInfo)
-#define __GET_MEM_OBJECT_INFO_ERR           __ERR_STR(clGetMemObjectInfo)
-#define __GET_IMAGE_INFO_ERR                __ERR_STR(clGetImageInfo)
-#define __GET_SAMPLER_INFO_ERR              __ERR_STR(clGetSamplerInfo)
-#define __GET_KERNEL_INFO_ERR               __ERR_STR(clGetKernelInfo)
-#define __GET_KERNEL_WORK_GROUP_INFO_ERR    __ERR_STR(clGetKernelWorkGroupInfo)
-#define __GET_PROGRAM_INFO_ERR              __ERR_STR(clGetProgramInfo)
-#define __GET_PROGRAM_BUILD_INFO_ERR        __ERR_STR(clGetProgramBuildInfo)
-#define __GET_COMMAND_QUEUE_INFO_ERR        __ERR_STR(clGetCommandQueueInfo)
-
-#define __CREATE_CONTEXT_FROM_TYPE_ERR      __ERR_STR(clCreateContextFromType)
-#define __GET_SUPPORTED_IMAGE_FORMATS_ERR   __ERR_STR(clGetSupportedImageFormats)
-
-#define __CREATE_BUFFER_ERR                 __ERR_STR(clCreateBuffer)
-#define __CREATE_SUBBUFFER_ERR              __ERR_STR(clCreateSubBuffer)
-#define __CREATE_GL_BUFFER_ERR              __ERR_STR(clCreateFromGLBuffer)
-#define __GET_GL_OBJECT_INFO_ERR            __ERR_STR(clGetGLObjectInfo)
-#define __CREATE_IMAGE2D_ERR                __ERR_STR(clCreateImage2D)
-#define __CREATE_IMAGE3D_ERR                __ERR_STR(clCreateImage3D)
-#define __CREATE_SAMPLER_ERR                __ERR_STR(clCreateSampler)
-#define __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR __ERR_STR(clSetMemObjectDestructorCallback)
-
-#define __CREATE_USER_EVENT_ERR             __ERR_STR(clCreateUserEvent)
-#define __SET_USER_EVENT_STATUS_ERR         __ERR_STR(clSetUserEventStatus)
-#define __SET_EVENT_CALLBACK_ERR            __ERR_STR(clSetEventCallback)
-#define __WAIT_FOR_EVENTS_ERR               __ERR_STR(clWaitForEvents)
-
-#define __CREATE_KERNEL_ERR                 __ERR_STR(clCreateKernel)
-#define __SET_KERNEL_ARGS_ERR               __ERR_STR(clSetKernelArg)
-#define __CREATE_PROGRAM_WITH_SOURCE_ERR    __ERR_STR(clCreateProgramWithSource)
-#define __CREATE_PROGRAM_WITH_BINARY_ERR    __ERR_STR(clCreateProgramWithBinary)
-#define __BUILD_PROGRAM_ERR                 __ERR_STR(clBuildProgram)
-#define __CREATE_KERNELS_IN_PROGRAM_ERR     __ERR_STR(clCreateKernelsInProgram)
-
-#define __CREATE_COMMAND_QUEUE_ERR          __ERR_STR(clCreateCommandQueue)
-#define __SET_COMMAND_QUEUE_PROPERTY_ERR    __ERR_STR(clSetCommandQueueProperty)
-#define __ENQUEUE_READ_BUFFER_ERR           __ERR_STR(clEnqueueReadBuffer)
-#define __ENQUEUE_READ_BUFFER_RECT_ERR      __ERR_STR(clEnqueueReadBufferRect)
-#define __ENQUEUE_WRITE_BUFFER_ERR          __ERR_STR(clEnqueueWriteBuffer)
-#define __ENQUEUE_WRITE_BUFFER_RECT_ERR     __ERR_STR(clEnqueueWriteBufferRect)
-#define __ENQEUE_COPY_BUFFER_ERR            __ERR_STR(clEnqueueCopyBuffer)
-#define __ENQEUE_COPY_BUFFER_RECT_ERR       __ERR_STR(clEnqueueCopyBufferRect)
-#define __ENQUEUE_READ_IMAGE_ERR            __ERR_STR(clEnqueueReadImage)
-#define __ENQUEUE_WRITE_IMAGE_ERR           __ERR_STR(clEnqueueWriteImage)
-#define __ENQUEUE_COPY_IMAGE_ERR            __ERR_STR(clEnqueueCopyImage)
-#define __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR  __ERR_STR(clEnqueueCopyImageToBuffer)
-#define __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR  __ERR_STR(clEnqueueCopyBufferToImage)
-#define __ENQUEUE_MAP_BUFFER_ERR            __ERR_STR(clEnqueueMapBuffer)
-#define __ENQUEUE_MAP_IMAGE_ERR             __ERR_STR(clEnqueueMapImage)
-#define __ENQUEUE_UNMAP_MEM_OBJECT_ERR      __ERR_STR(clEnqueueUnMapMemObject)
-#define __ENQUEUE_NDRANGE_KERNEL_ERR        __ERR_STR(clEnqueueNDRangeKernel)
-#define __ENQUEUE_TASK_ERR                  __ERR_STR(clEnqueueTask)
-#define __ENQUEUE_NATIVE_KERNEL             __ERR_STR(clEnqueueNativeKernel)
-#define __ENQUEUE_MARKER_ERR                __ERR_STR(clEnqueueMarker)
-#define __ENQUEUE_WAIT_FOR_EVENTS_ERR       __ERR_STR(clEnqueueWaitForEvents)
-#define __ENQUEUE_BARRIER_ERR               __ERR_STR(clEnqueueBarrier)
-
-#define __ENQUEUE_ACQUIRE_GL_ERR            __ERR_STR(clEnqueueAcquireGLObjects)
-#define __ENQUEUE_RELEASE_GL_ERR            __ERR_STR(clEnqueueReleaseGLObjects)
-
-#define __UNLOAD_COMPILER_ERR               __ERR_STR(clUnloadCompiler)
-
-#define __FLUSH_ERR                         __ERR_STR(clFlush)
-#define __FINISH_ERR                        __ERR_STR(clFinish)
-
-#define __CREATE_SUB_DEVICES                __ERR_STR(clCreateSubDevicesEXT)
-#endif // __CL_USER_OVERRIDE_ERROR_STRINGS
-//! \endcond
-
-/*! \class string
- * \brief Simple string class, that provides a limited subset of std::string
- * functionality but avoids many of the issues that come with that class.
- */
-class string
-{
-private:
-    ::size_t size_;
-    char * str_;
-public:
-    string(void) : size_(0), str_(NULL)
-    {
-    }
-
-    string(char * str, ::size_t size) :
-        size_(size),
-        str_(NULL)
-    {
-        str_ = new char[size_+1];
-        if (str_ != NULL) {
-            memcpy(str_, str, size_  * sizeof(char));
-            str_[size_] = '\0';
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string(char * str) :
-        str_(NULL)
-    {
-        size_= ::strlen(str);
-        str_ = new char[size_ + 1];
-        if (str_ != NULL) {
-            memcpy(str_, str, (size_ + 1) * sizeof(char));
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string& operator=(const string& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        if (rhs.size_ == 0 || rhs.str_ == NULL) {
-            size_ = 0;
-            str_  = NULL;
-        } 
-        else {
-            size_ = rhs.size_;
-            str_ = new char[size_ + 1];
-            if (str_ != NULL) {
-                memcpy(str_, rhs.str_, (size_ + 1) * sizeof(char));
-            }
-            else {
-                size_ = 0;
-            }
-        }
-
-        return *this;
-    }
-
-    string(const string& rhs)
-    {
-        *this = rhs;
-    }
-
-    ~string()
-    {
-        if (str_ != NULL) {
-            delete[] str_;
-        }
-    }
-
-    ::size_t size(void) const   { return size_; }
-    ::size_t length(void) const { return size(); }
-
-    const char * c_str(void) const { return (str_) ? str_ : "";}
-};
-
-#if !defined(__USE_DEV_STRING) && !defined(__NO_STD_STRING)
-#include <string>
-typedef std::string STRING_CLASS;
-#elif !defined(__USE_DEV_STRING) 
-typedef cl::string STRING_CLASS;
-#endif
-
-#if !defined(__USE_DEV_VECTOR) && !defined(__NO_STD_VECTOR)
-#include <vector>
-#define VECTOR_CLASS std::vector
-#elif !defined(__USE_DEV_VECTOR) 
-#define VECTOR_CLASS cl::vector 
-#endif
-
-#if !defined(__MAX_DEFAULT_VECTOR_SIZE)
-#define __MAX_DEFAULT_VECTOR_SIZE 10
-#endif
-
-/*! \class vector
- * \brief Fixed sized vector implementation that mirroring 
- * std::vector functionality.
- */
-template <typename T, unsigned int N = __MAX_DEFAULT_VECTOR_SIZE>
-class vector
-{
-private:
-    T data_[N];
-    unsigned int size_;
-    bool empty_;
-public:
-    vector() : 
-        size_(-1),
-        empty_(true)
-    {}
-
-    ~vector() {}
-
-    unsigned int size(void) const
-    {
-        return size_ + 1;
-    }
-
-    void clear()
-    {
-        size_ = -1;
-        empty_ = true;
-    }
-
-    void push_back (const T& x)
-    { 
-        if (size() < N) {
-            size_++;  
-            data_[size_] = x;
-            empty_ = false;
-        }
-    }
-
-    void pop_back(void)
-    {
-        if (!empty_) {
-            data_[size_].~T();
-            size_--;
-            if (size_ == -1) {
-                empty_ = true;
-            }
-        }
-    }
-  
-    vector(const vector<T, N>& vec) : 
-        size_(vec.size_),
-        empty_(vec.empty_)
-    {
-        if (!empty_) {
-            memcpy(&data_[0], &vec.data_[0], size() * sizeof(T));
-        }
-    } 
-
-    vector(unsigned int size, const T& val = T()) :
-        size_(-1),
-        empty_(true)
-    {
-        for (unsigned int i = 0; i < size; i++) {
-            push_back(val);
-        }
-    }
-
-    vector<T, N>& operator=(const vector<T, N>& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        size_  = rhs.size_;
-        empty_ = rhs.empty_;
-
-        if (!empty_) {	
-            memcpy(&data_[0], &rhs.data_[0], size() * sizeof(T));
-        }
-    
-        return *this;
-    }
-
-    bool operator==(vector<T,N> &vec)
-    {
-        if (empty_ && vec.empty_) {
-            return true;
-        }
-
-        if (size() != vec.size()) {
-            return false;
-        }
-
-        return memcmp(&data_[0], &vec.data_[0], size() * sizeof(T)) == 0 ? true : false;
-    }
-  
-    operator T* ()             { return data_; }
-    operator const T* () const { return data_; }
-   
-    bool empty (void) const
-    {
-        return empty_;
-    }
-  
-    unsigned int max_size (void) const
-    {
-        return N;
-    }
-
-    unsigned int capacity () const
-    {
-        return sizeof(T) * N;
-    }
-
-    T& operator[](int index)
-    {
-        return data_[index];
-    }
-  
-    T operator[](int index) const
-    {
-        return data_[index];
-    }
-  
-    template<class I>
-    void assign(I start, I end)
-    {
-        clear();   
-        while(start < end) {
-            push_back(*start);
-            start++;
-        }
-    }
-
-    /*! \class iterator
-     * \brief Iterator class for vectors
-     */
-    class iterator
-    {
-    private:
-        vector<T,N> vec_;
-        int index_;
-        bool initialized_;
-    public:
-        iterator(void) : 
-            index_(-1),
-            initialized_(false)
-        {
-            index_ = -1;
-            initialized_ = false;
-        }
-
-        ~iterator(void) {}
-
-        static iterator begin(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = 0;
-            }
-
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-
-        static iterator end(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = vec.size();
-            }
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-    
-        bool operator==(iterator i)
-        {
-            return ((vec_ == i.vec_) && 
-                    (index_ == i.index_) && 
-                    (initialized_ == i.initialized_));
-        }
-
-        bool operator!=(iterator i)
-        {
-            return (!(*this==i));
-        }
-
-        void operator++()
-        {
-            index_++;
-        }
-
-        void operator++(int x)
-        {
-            index_ += x;
-        }
-
-        void operator--()
-        {
-            index_--;
-        }
-
-        void operator--(int x)
-        {
-            index_ -= x;
-        }
-
-        T operator *()
-        {
-            return vec_[index_];
-        }
-    };
-
-    iterator begin(void)
-    {
-        return iterator::begin(*this);
-    }
-
-    iterator end(void)
-    {
-        return iterator::end(*this);
-    }
-
-    T& front(void)
-    {
-        return data_[0];
-    }
-
-    T& back(void)
-    {
-        return data_[size_];
-    }
-
-    const T& front(void) const
-    {
-        return data_[0];
-    }
-
-    const T& back(void) const
-    {
-        return data_[size_];
-    }
-};  
-    
-/*!
- * \brief size_t class used to interface between C++ and
- * OpenCL C calls that require arrays of size_t values, who's
- * size is known statically.
- */
-template <int N>
-struct size_t : public cl::vector< ::size_t, N> { };
-
-namespace detail {
-
-// GetInfo help struct
-template <typename Functor, typename T>
-struct GetInfoHelper
-{
-    static cl_int
-    get(Functor f, cl_uint name, T* param)
-    {
-        return f(name, sizeof(T), param, NULL);
-    }
-};
-
-// Specialized GetInfoHelper for VECTOR_CLASS params
-template <typename Func, typename T>
-struct GetInfoHelper<Func, VECTOR_CLASS<T> >
-{
-    static cl_int get(Func f, cl_uint name, VECTOR_CLASS<T>* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        T* value = (T*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        param->assign(&value[0], &value[required/sizeof(T)]);
-        return CL_SUCCESS;
-    }
-};
-
-// Specialized for getInfo<CL_PROGRAM_BINARIES>
-template <typename Func>
-struct GetInfoHelper<Func, VECTOR_CLASS<char *> >
-{
-    static cl_int
-    get(Func f, cl_uint name, VECTOR_CLASS<char *>* param)
-    {
-      cl_uint err = f(name, param->size() * sizeof(char *), &(*param)[0], NULL);
-      if (err != CL_SUCCESS) {
-        return err;
-      }
-      
-      return CL_SUCCESS;
-    }
-};
-
-// Specialized GetInfoHelper for STRING_CLASS params
-template <typename Func>
-struct GetInfoHelper<Func, STRING_CLASS>
-{
-    static cl_int get(Func f, cl_uint name, STRING_CLASS* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        char* value = (char*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        *param = value;
-        return CL_SUCCESS;
-    }
-};
-
-#define __GET_INFO_HELPER_WITH_RETAIN(CPP_TYPE) \
-namespace detail { \
-template <typename Func> \
-struct GetInfoHelper<Func, CPP_TYPE> \
-{ \
-    static cl_int get(Func f, cl_uint name, CPP_TYPE* param) \
-    { \
-      cl_uint err = f(name, sizeof(CPP_TYPE), param, NULL); \
-      if (err != CL_SUCCESS) { \
-        return err; \
-      } \
-      \
-      return ReferenceHandler<CPP_TYPE::cl_type>::retain((*param)()); \
-    } \
-}; \
-} 
-
-
-#define __PARAM_NAME_INFO_1_0(F) \
-    F(cl_platform_info, CL_PLATFORM_PROFILE, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VERSION, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_NAME, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VENDOR, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_device_info, CL_DEVICE_TYPE, cl_device_type) \
-    F(cl_device_info, CL_DEVICE_VENDOR_ID, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_COMPUTE_UNITS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_CLOCK_FREQUENCY, cl_uint) \
-    F(cl_device_info, CL_DEVICE_ADDRESS_BITS, cl_bitfield) \
-    F(cl_device_info, CL_DEVICE_MAX_READ_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WRITE_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_MEM_ALLOC_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_DEPTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE_SUPPORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_PARAMETER_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_SAMPLERS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MEM_BASE_ADDR_ALIGN, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_SINGLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_TYPE, cl_device_mem_cache_type) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE, cl_uint)\
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_TYPE, cl_device_local_mem_type) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_ERROR_CORRECTION_SUPPORT, cl_bool) \
-    F(cl_device_info, CL_DEVICE_PROFILING_TIMER_RESOLUTION, ::size_t) \
-    F(cl_device_info, CL_DEVICE_ENDIAN_LITTLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_COMPILER_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_EXECUTION_CAPABILITIES, cl_device_exec_capabilities) \
-    F(cl_device_info, CL_DEVICE_QUEUE_PROPERTIES, cl_command_queue_properties) \
-    F(cl_device_info, CL_DEVICE_PLATFORM, cl_platform_id) \
-    F(cl_device_info, CL_DEVICE_NAME, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VENDOR, STRING_CLASS) \
-    F(cl_device_info, CL_DRIVER_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_PROFILE, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_context_info, CL_CONTEXT_REFERENCE_COUNT, cl_uint) \
-    F(cl_context_info, CL_CONTEXT_DEVICES, VECTOR_CLASS<Device>) \
-    F(cl_context_info, CL_CONTEXT_PROPERTIES, VECTOR_CLASS<cl_context_properties>) \
-    \
-    F(cl_event_info, CL_EVENT_COMMAND_QUEUE, cl::CommandQueue) \
-    F(cl_event_info, CL_EVENT_COMMAND_TYPE, cl_command_type) \
-    F(cl_event_info, CL_EVENT_REFERENCE_COUNT, cl_uint) \
-    F(cl_event_info, CL_EVENT_COMMAND_EXECUTION_STATUS, cl_uint) \
-    \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_QUEUED, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_SUBMIT, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_START, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_END, cl_ulong) \
-    \
-    F(cl_mem_info, CL_MEM_TYPE, cl_mem_object_type) \
-    F(cl_mem_info, CL_MEM_FLAGS, cl_mem_flags) \
-    F(cl_mem_info, CL_MEM_SIZE, ::size_t) \
-    F(cl_mem_info, CL_MEM_HOST_PTR, void*) \
-    F(cl_mem_info, CL_MEM_MAP_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_REFERENCE_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_CONTEXT, cl::Context) \
-    \
-    F(cl_image_info, CL_IMAGE_FORMAT, cl_image_format) \
-    F(cl_image_info, CL_IMAGE_ELEMENT_SIZE, ::size_t) \
-    F(cl_image_info, CL_IMAGE_ROW_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_SLICE_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_WIDTH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_HEIGHT, ::size_t) \
-    F(cl_image_info, CL_IMAGE_DEPTH, ::size_t) \
-    \
-    F(cl_sampler_info, CL_SAMPLER_REFERENCE_COUNT, cl_uint) \
-    F(cl_sampler_info, CL_SAMPLER_CONTEXT, cl::Context) \
-    F(cl_sampler_info, CL_SAMPLER_NORMALIZED_COORDS, cl_addressing_mode) \
-    F(cl_sampler_info, CL_SAMPLER_ADDRESSING_MODE, cl_filter_mode) \
-    F(cl_sampler_info, CL_SAMPLER_FILTER_MODE, cl_bool) \
-    \
-    F(cl_program_info, CL_PROGRAM_REFERENCE_COUNT, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_CONTEXT, cl::Context) \
-    F(cl_program_info, CL_PROGRAM_NUM_DEVICES, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_DEVICES, VECTOR_CLASS<cl_device_id>) \
-    F(cl_program_info, CL_PROGRAM_SOURCE, STRING_CLASS) \
-    F(cl_program_info, CL_PROGRAM_BINARY_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_program_info, CL_PROGRAM_BINARIES, VECTOR_CLASS<char *>) \
-    \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_STATUS, cl_build_status) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_OPTIONS, STRING_CLASS) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_LOG, STRING_CLASS) \
-    \
-    F(cl_kernel_info, CL_KERNEL_FUNCTION_NAME, STRING_CLASS) \
-    F(cl_kernel_info, CL_KERNEL_NUM_ARGS, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_REFERENCE_COUNT, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_CONTEXT, cl::Context) \
-    F(cl_kernel_info, CL_KERNEL_PROGRAM, cl::Program) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_COMPILE_WORK_GROUP_SIZE, cl::size_t<3>) \
-    F(cl_kernel_work_group_info, CL_KERNEL_LOCAL_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_command_queue_info, CL_QUEUE_CONTEXT, cl::Context) \
-    F(cl_command_queue_info, CL_QUEUE_DEVICE, cl::Device) \
-    F(cl_command_queue_info, CL_QUEUE_REFERENCE_COUNT, cl_uint) \
-    F(cl_command_queue_info, CL_QUEUE_PROPERTIES, cl_command_queue_properties)
-
-#if defined(CL_VERSION_1_1)
-#define __PARAM_NAME_INFO_1_1(F) \
-    F(cl_context_info, CL_CONTEXT_NUM_DEVICES, cl_uint)\
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_DOUBLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HALF_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HOST_UNIFIED_MEMORY, cl_bool) \
-    \
-    F(cl_mem_info, CL_MEM_ASSOCIATED_MEMOBJECT, cl::Memory) \
-    F(cl_mem_info, CL_MEM_OFFSET, ::size_t) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_PRIVATE_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_event_info, CL_EVENT_CONTEXT, cl::Context)
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-#define __PARAM_NAME_DEVICE_FISSION(F) \
-    F(cl_device_info, CL_DEVICE_PARENT_DEVICE_EXT, cl_device_id) \
-	F(cl_device_info, CL_DEVICE_PARTITION_TYPES_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_AFFINITY_DOMAINS_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_REFERENCE_COUNT_EXT , cl_uint) \
-	F(cl_device_info, CL_DEVICE_PARTITION_STYLE_EXT, VECTOR_CLASS<cl_device_partition_property_ext>)
-#endif // USE_CL_DEVICE_FISSION
-
-template <typename enum_type, cl_int Name>
-struct param_traits {};
-
-#define __DECLARE_PARAM_TRAITS(token, param_name, T) \
-struct token;                                        \
-template<>                                           \
-struct param_traits<detail:: token,param_name>       \
-{                                                    \
-    enum { value = param_name };                     \
-    typedef T param_type;                            \
-};
-
-__PARAM_NAME_INFO_1_0(__DECLARE_PARAM_TRAITS);
-#if defined(CL_VERSION_1_1)
-__PARAM_NAME_INFO_1_1(__DECLARE_PARAM_TRAITS);
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-__PARAM_NAME_DEVICE_FISSION(__DECLARE_PARAM_TRAITS);
-#endif // USE_CL_DEVICE_FISSION
-
-#undef __DECLARE_PARAM_TRAITS
-
-// Convenience functions
-
-template <typename Func, typename T>
-inline cl_int
-getInfo(Func f, cl_uint name, T* param)
-{
-    return GetInfoHelper<Func, T>::get(f, name, param);
-}
-
-template <typename Func, typename Arg0>
-struct GetInfoFunctor0
-{
-    Func f_; const Arg0& arg0_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename Arg1>
-struct GetInfoFunctor1
-{
-    Func f_; const Arg0& arg0_; const Arg1& arg1_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, arg1_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, cl_uint name, T* param)
-{
-    GetInfoFunctor0<Func, Arg0> f0 = { f, arg0 };
-    return GetInfoHelper<GetInfoFunctor0<Func, Arg0>, T>
-        ::get(f0, name, param);
-}
-
-template <typename Func, typename Arg0, typename Arg1, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, const Arg1& arg1, cl_uint name, T* param)
-{
-    GetInfoFunctor1<Func, Arg0, Arg1> f0 = { f, arg0, arg1 };
-    return GetInfoHelper<GetInfoFunctor1<Func, Arg0, Arg1>, T>
-        ::get(f0, name, param);
-}
-
-template<typename T>
-struct ReferenceHandler
-{ };
-
-template <>
-struct ReferenceHandler<cl_device_id>
-{
-    // cl_device_id does not have retain().
-    static cl_int retain(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-    // cl_device_id does not have release().
-    static cl_int release(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-};
-
-template <>
-struct ReferenceHandler<cl_platform_id>
-{
-    // cl_platform_id does not have retain().
-    static cl_int retain(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-    // cl_platform_id does not have release().
-    static cl_int release(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-};
-
-template <>
-struct ReferenceHandler<cl_context>
-{
-    static cl_int retain(cl_context context)
-    { return ::clRetainContext(context); }
-    static cl_int release(cl_context context)
-    { return ::clReleaseContext(context); }
-};
-
-template <>
-struct ReferenceHandler<cl_command_queue>
-{
-    static cl_int retain(cl_command_queue queue)
-    { return ::clRetainCommandQueue(queue); }
-    static cl_int release(cl_command_queue queue)
-    { return ::clReleaseCommandQueue(queue); }
-};
-
-template <>
-struct ReferenceHandler<cl_mem>
-{
-    static cl_int retain(cl_mem memory)
-    { return ::clRetainMemObject(memory); }
-    static cl_int release(cl_mem memory)
-    { return ::clReleaseMemObject(memory); }
-};
-
-template <>
-struct ReferenceHandler<cl_sampler>
-{
-    static cl_int retain(cl_sampler sampler)
-    { return ::clRetainSampler(sampler); }
-    static cl_int release(cl_sampler sampler)
-    { return ::clReleaseSampler(sampler); }
-};
-
-template <>
-struct ReferenceHandler<cl_program>
-{
-    static cl_int retain(cl_program program)
-    { return ::clRetainProgram(program); }
-    static cl_int release(cl_program program)
-    { return ::clReleaseProgram(program); }
-};
-
-template <>
-struct ReferenceHandler<cl_kernel>
-{
-    static cl_int retain(cl_kernel kernel)
-    { return ::clRetainKernel(kernel); }
-    static cl_int release(cl_kernel kernel)
-    { return ::clReleaseKernel(kernel); }
-};
-
-template <>
-struct ReferenceHandler<cl_event>
-{
-    static cl_int retain(cl_event event)
-    { return ::clRetainEvent(event); }
-    static cl_int release(cl_event event)
-    { return ::clReleaseEvent(event); }
-};
-
-template <typename T>
-class Wrapper
-{
-public:
-    typedef T cl_type;
-
-protected:
-    cl_type object_;
-
-public:
-    Wrapper() : object_(NULL) { }
-
-    ~Wrapper()
-    {
-        if (object_ != NULL) { release(); }
-    }
-
-    Wrapper(const Wrapper<cl_type>& rhs)
-    {
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-    }
-
-    Wrapper<cl_type>& operator = (const Wrapper<cl_type>& rhs)
-    {
-        if (object_ != NULL) { release(); }
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-        return *this;
-    }
-
-    cl_type operator ()() const { return object_; }
-
-    cl_type& operator ()() { return object_; }
-
-protected:
-
-    cl_int retain() const
-    {
-        return ReferenceHandler<cl_type>::retain(object_);
-    }
-
-    cl_int release() const
-    {
-        return ReferenceHandler<cl_type>::release(object_);
-    }
-};
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-static inline cl_int errHandler (
-    cl_int err,
-    const char * errStr = NULL) throw(Error)
-{
-    if (err != CL_SUCCESS) {
-        throw Error(err, errStr);
-    }
-    return err;
-}
-#else
-static inline cl_int errHandler (cl_int err, const char * errStr = NULL)
-{
-    return err;
-}
-#endif // __CL_ENABLE_EXCEPTIONS
-
-} // namespace detail
-//! \endcond
-
-/*! \stuct ImageFormat
- * \brief ImageFormat interface fro cl_image_format.
- */
-struct ImageFormat : public cl_image_format
-{
-    ImageFormat(){}
-
-    ImageFormat(cl_channel_order order, cl_channel_type type)
-    {
-        image_channel_order = order;
-        image_channel_data_type = type;
-    }
-
-    ImageFormat& operator = (const ImageFormat& rhs)
-    {
-        if (this != &rhs) {
-            this->image_channel_data_type = rhs.image_channel_data_type;
-            this->image_channel_order     = rhs.image_channel_order;
-        }
-        return *this;
-    }
-};
-
-/*! \class Device
- * \brief Device interface for cl_device_id.
- */
-class Device : public detail::Wrapper<cl_device_id>
-{
-public:
-    Device(cl_device_id device) { object_ = device; }
-
-    Device() : detail::Wrapper<cl_type>() { }
-
-    Device(const Device& device) : detail::Wrapper<cl_type>(device) { }
-
-    Device& operator = (const Device& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_device_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetDeviceInfo, object_, name, param),
-            __GET_DEVICE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_device_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_device_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(USE_CL_DEVICE_FISSION)
-	cl_int createSubDevices(
-		const cl_device_partition_property_ext * properties,
-		VECTOR_CLASS<Device>* devices)
-	{
-		typedef CL_API_ENTRY cl_int 
-			( CL_API_CALL * PFN_clCreateSubDevicesEXT)(
-				cl_device_id /*in_device*/,
-                const cl_device_partition_property_ext * /* properties */,
-                cl_uint /*num_entries*/,
-                cl_device_id * /*out_devices*/,
-                cl_uint * /*num_devices*/ ) CL_EXT_SUFFIX__VERSION_1_1;
-
-		static PFN_clCreateSubDevicesEXT pfn_clCreateSubDevicesEXT = NULL;
-		__INIT_CL_EXT_FCN_PTR(clCreateSubDevicesEXT);
-
-		cl_uint n = 0;
-        cl_int err = pfn_clCreateSubDevicesEXT(object_, properties, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clCreateSubDevicesEXT(object_, properties, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
- 	}
-#endif
-};
-
-/*! \class Platform
- *  \brief Platform interface.
- */
-class Platform : public detail::Wrapper<cl_platform_id>
-{
-public:
-    static const Platform null();
-
-    Platform(cl_platform_id platform) { object_ = platform; }
-
-    Platform() : detail::Wrapper<cl_type>()  { }
-
-    Platform(const Platform& platform) : detail::Wrapper<cl_type>(platform) { }
-
-    Platform& operator = (const Platform& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getInfo(cl_platform_info name, STRING_CLASS* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetPlatformInfo, object_, name, param),
-            __GET_PLATFORM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_platform_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_platform_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getDevices(
-        cl_device_type type,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetDeviceIDs(object_, type, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = ::clGetDeviceIDs(object_, type, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-
-#if defined(USE_DX_INTEROP)
-   /*! \brief Get the list of available D3D10 devices.
-     *
-     *  \param d3d_device_source.
-     *
-     *  \param d3d_object.
-     *
-     *  \param d3d_device_set.
-     *
-     *  \param devices returns a vector of OpenCL D3D10 devices found. The cl::Device
-     *  values returned in devices can be used to identify a specific OpenCL
-     *  device. If \a devices argument is NULL, this argument is ignored.
-     *
-     *  \return One of the following values:
-     *    - CL_SUCCESS if the function is executed successfully.
-     *
-     *  The application can query specific capabilities of the OpenCL device(s)
-     *  returned by cl::getDevices. This can be used by the application to
-     *  determine which device(s) to use.
-     *
-     * \note In the case that exceptions are enabled and a return value
-     * other than CL_SUCCESS is generated, then cl::Error exception is
-     * generated.
-     */
-    cl_int getDevices(
-        cl_d3d10_device_source_khr d3d_device_source,
-        void *                     d3d_object,
-        cl_d3d10_device_set_khr    d3d_device_set,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clGetDeviceIDsFromD3D10KHR)(
-            cl_platform_id platform, 
-            cl_d3d10_device_source_khr d3d_device_source, 
-            void * d3d_object,
-            cl_d3d10_device_set_khr d3d_device_set,
-            cl_uint num_entries,
-            cl_device_id * devices,
-            cl_uint* num_devices);
-
-        static PFN_clGetDeviceIDsFromD3D10KHR pfn_clGetDeviceIDsFromD3D10KHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clGetDeviceIDsFromD3D10KHR);
-
-        cl_uint n = 0;
-        cl_int err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set, 
-            0, 
-            NULL, 
-            &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set,
-            n, 
-            ids, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-#endif
-
-    static cl_int get(
-        VECTOR_CLASS<Platform>* platforms)
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetPlatformIDs(0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        cl_platform_id* ids = (cl_platform_id*) alloca(
-            n * sizeof(cl_platform_id));
-        err = ::clGetPlatformIDs(n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        platforms->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-};
-
-static inline cl_int
-UnloadCompiler()
-{
-    return ::clUnloadCompiler();
-}
-
-class Context : public detail::Wrapper<cl_context>
-{
-public:
-    Context(
-        const VECTOR_CLASS<Device>& devices,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContext(
-            properties, (cl_uint) devices.size(),
-            (cl_device_id*) &devices.front(),
-            notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context(
-        cl_device_type type,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContextFromType(
-            properties, type, notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context() : detail::Wrapper<cl_type>() { }
-
-    Context(const Context& context) : detail::Wrapper<cl_type>(context) { }
-
-    Context& operator = (const Context& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_context_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetContextInfo, object_, name, param),
-            __GET_CONTEXT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_context_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_context_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getSupportedImageFormats(
-        cl_mem_flags flags,
-        cl_mem_object_type type,
-        VECTOR_CLASS<ImageFormat>* formats) const
-    {
-        cl_uint numEntries;
-        cl_int err = ::clGetSupportedImageFormats(
-           object_, 
-           flags,
-           type, 
-           0, 
-           NULL, 
-           &numEntries);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        ImageFormat* value = (ImageFormat*)
-            alloca(numEntries * sizeof(ImageFormat));
-        err = ::clGetSupportedImageFormats(
-            object_, 
-            flags, 
-            type, 
-            numEntries,
-            (cl_image_format*) value, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        formats->assign(&value[0], &value[numEntries]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Context)
-
-/*! \class Event
- * \brief Event interface for cl_event.
- */
-class Event : public detail::Wrapper<cl_event>
-{
-public:
-    Event() : detail::Wrapper<cl_type>() { }
-
-    Event(const Event& event) : detail::Wrapper<cl_type>(event) { }
-
-    Event& operator = (const Event& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_event_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetEventInfo, object_, name, param),
-            __GET_EVENT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_event_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_event_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getProfilingInfo(cl_profiling_info name, T* param) const
-    {
-        return detail::errHandler(detail::getInfo(
-            &::clGetEventProfilingInfo, object_, name, param),
-            __GET_EVENT_PROFILE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_profiling_info, name>::param_type
-    getProfilingInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_profiling_info, name>::param_type param;
-        cl_int result = getProfilingInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int wait() const
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(1, &object_),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setCallback(
-        cl_int type,
-        void (CL_CALLBACK * pfn_notify)(cl_event, cl_int, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetEventCallback(
-                object_,
-                type,
-                pfn_notify,
-                user_data), 
-            __SET_EVENT_CALLBACK_ERR);
-    }
-#endif
-
-    static cl_int
-    waitForEvents(const VECTOR_CLASS<Event>& events)
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(
-                (cl_uint) events.size(), (cl_event*)&events.front()),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Event)
-
-#if defined(CL_VERSION_1_1)
-/*! \class UserEvent
- * \brief User event interface for cl_event.
- */
-class UserEvent : public Event
-{
-public:
-    UserEvent(
-        const Context& context,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateUserEvent(
-            context(),
-            &error);
-
-        detail::errHandler(error, __CREATE_USER_EVENT_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    UserEvent() : Event() { }
-
-    UserEvent(const UserEvent& event) : Event(event) { }
-
-    UserEvent& operator = (const UserEvent& rhs)
-    {
-        if (this != &rhs) {
-            Event::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int setStatus(cl_int status)
-    {
-        return detail::errHandler(
-            ::clSetUserEventStatus(object_,status), 
-            __SET_USER_EVENT_STATUS_ERR);
-    }
-};
-#endif
-
-inline static cl_int
-WaitForEvents(const VECTOR_CLASS<Event>& events)
-{
-    return detail::errHandler(
-        ::clWaitForEvents(
-            (cl_uint) events.size(), (cl_event*)&events.front()),
-        __WAIT_FOR_EVENTS_ERR);
-}
-
-/*! \class Memory
- * \brief Memory interface for cl_mem.
- */
-class Memory : public detail::Wrapper<cl_mem>
-{
-public:
-    Memory() : detail::Wrapper<cl_type>() { }
-
-    Memory(const Memory& memory) : detail::Wrapper<cl_type>(memory) { }
-
-    Memory& operator = (const Memory& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_mem_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetMemObjectInfo, object_, name, param),
-            __GET_MEM_OBJECT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_mem_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_mem_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setDestructorCallback(
-        void (CL_CALLBACK * pfn_notify)(cl_mem, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetMemObjectDestructorCallback(
-                object_,
-                pfn_notify,
-                user_data), 
-            __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR);
-    }
-#endif
-
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Memory)
-
-/*! \class Buffer
- * \brief Memory buffer interface.
- */
-class Buffer : public Memory
-{
-public:
-    Buffer(
-        const Context& context,
-        cl_mem_flags flags,
-        ::size_t size,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateBuffer(context(), flags, size, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Buffer() : Memory() { }
-
-    Buffer(const Buffer& buffer) : Memory(buffer) { }
-
-    Buffer& operator = (const Buffer& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-
-#if defined(CL_VERSION_1_1)
-    Buffer createSubBuffer(
-        cl_mem_flags flags,
-        cl_buffer_create_type buffer_create_type,
-        const void * buffer_create_info,
-        cl_int * err = NULL)
-    {
-        Buffer result;
-        cl_int error;
-        result.object_ = ::clCreateSubBuffer(
-            object_, 
-            flags, 
-            buffer_create_type, 
-            buffer_create_info, 
-            &error);
-
-        detail::errHandler(error, __CREATE_SUBBUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-
-        return result;
-	}		
-#endif
-};
-
-#if defined (USE_DX_INTEROP)
-class BufferD3D10 : public Buffer
-{
-public:
-    typedef CL_API_ENTRY cl_mem (CL_API_CALL *PFN_clCreateFromD3D10BufferKHR)(
-    cl_context context, cl_mem_flags flags, ID3D10Buffer*  buffer,
-    cl_int* errcode_ret);
-
-    BufferD3D10(
-        const Context& context,
-        cl_mem_flags flags,
-        ID3D10Buffer* bufobj,
-        cl_int * err = NULL)
-    {
-        static PFN_clCreateFromD3D10BufferKHR pfn_clCreateFromD3D10BufferKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clCreateFromD3D10BufferKHR);
-
-        cl_int error;
-        object_ = pfn_clCreateFromD3D10BufferKHR(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferD3D10() : Buffer() { }
-
-    BufferD3D10(const BufferD3D10& buffer) : Buffer(buffer) { }
-
-    BufferD3D10& operator = (const BufferD3D10& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-};
-#endif
-
-/*! \class BufferGL
- * \brief Memory buffer interface for GL interop.
- */
-class BufferGL : public Buffer
-{
-public:
-    BufferGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLBuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferGL() : Buffer() { }
-
-    BufferGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferGL& operator = (const BufferGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class BufferRenderGL
- * \brief Memory buffer interface for GL interop with renderbuffer.
- */
-class BufferRenderGL : public Buffer
-{
-public:
-    BufferRenderGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLRenderbuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferRenderGL() : Buffer() { }
-
-    BufferRenderGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferRenderGL& operator = (const BufferRenderGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class Image
- * \brief Base class  interface for all images.
- */
-class Image : public Memory
-{
-protected:
-    Image() : Memory() { }
-
-    Image(const Image& image) : Memory(image) { }
-
-    Image& operator = (const Image& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-public:
-    template <typename T>
-    cl_int getImageInfo(cl_image_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetImageInfo, object_, name, param),
-            __GET_IMAGE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_image_info, name>::param_type
-    getImageInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_image_info, name>::param_type param;
-        cl_int result = getImageInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-/*! \class Image2D
- * \brief Image interface for 2D images.
- */
-class Image2D : public Image
-{
-public:
-    Image2D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t row_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage2D(
-            context(), flags,&format, width, height, row_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE2D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2D() { }
-
-    Image2D(const Image2D& image2D) : Image(image2D) { }
-
-    Image2D& operator = (const Image2D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image2DGL : public Image2D
-{
-public:
-    Image2DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture2D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2DGL() : Image2D() { }
-
-    Image2DGL(const Image2DGL& image) : Image2D(image) { }
-
-    Image2DGL& operator = (const Image2DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image2D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image3D
- * \brief Image interface for 3D images.
- */
-class Image3D : public Image
-{
-public:
-    Image3D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t depth,
-        ::size_t row_pitch = 0,
-        ::size_t slice_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage3D(
-            context(), flags, &format, width, height, depth, row_pitch,
-            slice_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE3D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3D() { }
-
-    Image3D(const Image3D& image3D) : Image(image3D) { }
-
-    Image3D& operator = (const Image3D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image3DGL : public Image3D
-{
-public:
-    Image3DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture3D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3DGL() : Image3D() { }
-
-    Image3DGL(const Image3DGL& image) : Image3D(image) { }
-
-    Image3DGL& operator = (const Image3DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image3D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Sampler
- * \brief Sampler interface for cl_sampler.
- */
-class Sampler : public detail::Wrapper<cl_sampler>
-{
-public:
-    Sampler() { }
-
-    Sampler(
-        const Context& context,
-        cl_bool normalized_coords,
-        cl_addressing_mode addressing_mode,
-        cl_filter_mode filter_mode,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateSampler(
-            context(), 
-            normalized_coords,
-            addressing_mode,
-            filter_mode,
-            &error);
-
-        detail::errHandler(error, __CREATE_SAMPLER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Sampler(const Sampler& sampler) : detail::Wrapper<cl_type>(sampler) { }
-
-    Sampler& operator = (const Sampler& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_sampler_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetSamplerInfo, object_, name, param),
-            __GET_SAMPLER_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_sampler_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_sampler_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Sampler)
-
-class Program;
-class CommandQueue;
-class Kernel;
-
-/*! \class NDRange
- * \brief NDRange interface
- */
-class NDRange
-{
-private:
-    size_t<3> sizes_;
-    cl_uint dimensions_;
-
-public:
-    NDRange()
-        : dimensions_(0)
-    { }
-
-    NDRange(::size_t size0)
-        : dimensions_(1)
-    {
-        sizes_.push_back(size0);
-    }
-
-    NDRange(::size_t size0, ::size_t size1)
-        : dimensions_(2)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-    }
-
-    NDRange(::size_t size0, ::size_t size1, ::size_t size2)
-        : dimensions_(3)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-        sizes_.push_back(size2);
-    }
-
-    operator const ::size_t*() const { return (const ::size_t*) sizes_; }
-    ::size_t dimensions() const { return dimensions_; }
-};
-
-static const NDRange NullRange;
-
-/*!
- * \struct LocalSpaceArg
- * \brief Local address raper for use with Kernel::setArg
- */
-struct LocalSpaceArg
-{
-    ::size_t size_;
-};
-
-namespace detail {
-
-template <typename T>
-struct KernelArgumentHandler
-{
-    static ::size_t size(const T&) { return sizeof(T); }
-    static T* ptr(T& value) { return &value; }
-};
-
-template <>
-struct KernelArgumentHandler<LocalSpaceArg>
-{
-    static ::size_t size(const LocalSpaceArg& value) { return value.size_; }
-    static void* ptr(LocalSpaceArg&) { return NULL; }
-};
-
-} 
-//! \endcond
-
-inline LocalSpaceArg
-__local(::size_t size)
-{
-    LocalSpaceArg ret = { size };
-    return ret;
-}
-
-class KernelFunctor;
-
-/*! \class Kernel
- * \brief Kernel interface that implements cl_kernel
- */
-class Kernel : public detail::Wrapper<cl_kernel>
-{
-public:
-    inline Kernel(const Program& program, const char* name, cl_int* err = NULL);
-
-    Kernel() { }
-
-    Kernel(const Kernel& kernel) : detail::Wrapper<cl_type>(kernel) { }
-
-    Kernel& operator = (const Kernel& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_kernel_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetKernelInfo, object_, name, param),
-            __GET_KERNEL_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_kernel_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getWorkGroupInfo(
-        const Device& device, cl_kernel_work_group_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetKernelWorkGroupInfo, object_, device(), name, param),
-                __GET_KERNEL_WORK_GROUP_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_work_group_info, name>::param_type
-        getWorkGroupInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-        detail::cl_kernel_work_group_info, name>::param_type param;
-        cl_int result = getWorkGroupInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int setArg(cl_uint index, T value)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(
-                object_,
-                index,
-                detail::KernelArgumentHandler<T>::size(value),
-                detail::KernelArgumentHandler<T>::ptr(value)),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    cl_int setArg(cl_uint index, ::size_t size, void* argPtr)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(object_, index, size, argPtr),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local);
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& global,
-        const NDRange& local);
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Kernel)
-
-/*! \class Program
- * \brief Program interface that implements cl_program.
- */
-class Program : public detail::Wrapper<cl_program>
-{
-public:
-    typedef VECTOR_CLASS<std::pair<const void*, ::size_t> > Binaries;
-    typedef VECTOR_CLASS<std::pair<const char*, ::size_t> > Sources;
-
-    Program(
-        const Context& context,
-        const Sources& sources,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-
-        const ::size_t n = (::size_t)sources.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const char** strings = (const char**) alloca(n * sizeof(const char*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            strings[i] = sources[(int)i].first;
-            lengths[i] = sources[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithSource(
-            context(), (cl_uint)n, strings, lengths, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_SOURCE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program(
-        const Context& context,
-        const VECTOR_CLASS<Device>& devices,
-        const Binaries& binaries,
-        VECTOR_CLASS<cl_int>* binaryStatus = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        const ::size_t n = binaries.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const unsigned char** images = (const unsigned char**) alloca(n * sizeof(const void*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            images[i] = (const unsigned char*)binaries[(int)i].first;
-            lengths[i] = binaries[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithBinary(
-            context(), (cl_uint) devices.size(),
-            (cl_device_id*)&devices.front(),
-            lengths, images, binaryStatus != NULL
-               ? (cl_int*) &binaryStatus->front()
-               : NULL, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_BINARY_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program() { }
-
-    Program(const Program& program) : detail::Wrapper<cl_type>(program) { }
-
-    Program& operator = (const Program& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int build(
-        const VECTOR_CLASS<Device>& devices,
-        const char* options = NULL,
-        void (CL_CALLBACK * notifyFptr)(cl_program, void *) = NULL,
-        void* data = NULL) const
-    {
-        return detail::errHandler(
-            ::clBuildProgram(
-                object_,
-                (cl_uint)
-                devices.size(),
-                (cl_device_id*)&devices.front(),
-                options,
-                notifyFptr,
-                data),
-                __BUILD_PROGRAM_ERR);
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_program_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetProgramInfo, object_, name, param),
-            __GET_PROGRAM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getBuildInfo(
-        const Device& device, cl_program_build_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetProgramBuildInfo, object_, device(), name, param),
-                __GET_PROGRAM_BUILD_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_build_info, name>::param_type
-    getBuildInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_build_info, name>::param_type param;
-        cl_int result = getBuildInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int createKernels(VECTOR_CLASS<Kernel>* kernels)
-    {
-        cl_uint numKernels;
-        cl_int err = ::clCreateKernelsInProgram(object_, 0, NULL, &numKernels);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        Kernel* value = (Kernel*) alloca(numKernels * sizeof(Kernel));
-        err = ::clCreateKernelsInProgram(
-            object_, numKernels, (cl_kernel*) value, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        kernels->assign(&value[0], &value[numKernels]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Program)
-
-inline Kernel::Kernel(const Program& program, const char* name, cl_int* err)
-{
-    cl_int error;
-
-    object_ = ::clCreateKernel(program(), name, &error);
-    detail::errHandler(error, __CREATE_KERNEL_ERR);
-
-    if (err != NULL) {
-        *err = error;
-    }
-
-}
-
-/*! \class CommandQueue
- * \brief CommandQueue interface for cl_command_queue.
- */
-class CommandQueue : public detail::Wrapper<cl_command_queue>
-{
-public:
-    CommandQueue(
-        const Context& context,
-        const Device& device,
-        cl_command_queue_properties properties = 0,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateCommandQueue(
-            context(), device(), properties, &error);
-
-        detail::errHandler(error, __CREATE_COMMAND_QUEUE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    CommandQueue() { }
-
-    CommandQueue(const CommandQueue& commandQueue) : detail::Wrapper<cl_type>(commandQueue) { }
-
-    CommandQueue& operator = (const CommandQueue& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_command_queue_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetCommandQueueInfo, object_, name, param),
-                __GET_COMMAND_QUEUE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_command_queue_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_command_queue_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int enqueueReadBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_BUFFER_ERR);
-    }
-
-    cl_int enqueueWriteBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        const void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBuffer(
-        const Buffer& src,
-        const Buffer& dst,
-        ::size_t src_offset,
-        ::size_t dst_offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBuffer(
-                object_, src(), dst(), src_offset, dst_offset, size,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int enqueueReadBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_READ_BUFFER_RECT_ERR);
-    }
-
-
-    cl_int enqueueWriteBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_RECT_ERR);
-    }
-
-    cl_int enqueueCopyBufferRect(
-        const Buffer& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        ::size_t src_row_pitch,
-        ::size_t src_slice_pitch,
-        ::size_t dst_row_pitch,
-        ::size_t dst_slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferRect(
-                object_, 
-                src(), 
-                dst(), 
-                (const ::size_t *)src_origin, 
-                (const ::size_t *)dst_origin, 
-                (const ::size_t *)region,
-                src_row_pitch,
-                src_slice_pitch,
-                dst_row_pitch,
-                dst_slice_pitch,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_RECT_ERR);
-    }
-#endif
-
-    cl_int enqueueReadImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_IMAGE_ERR);
-    }
-
-    cl_int enqueueWriteImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_WRITE_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImage(
-        const Image& src,
-        const Image& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImage(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *)dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImageToBuffer(
-        const Image& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& region,
-        ::size_t dst_offset,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImageToBuffer(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *) region, dst_offset,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBufferToImage(
-        const Buffer& src,
-        const Image& dst,
-        ::size_t src_offset,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferToImage(
-                object_, src(), dst(), src_offset,
-                (const ::size_t *) dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR);
-    }
-
-    void* enqueueMapBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        ::size_t offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapBuffer(
-            object_, buffer(), blocking, flags, offset, size,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-        return result;
-    }
-
-    void* enqueueMapImage(
-        const Image& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t * row_pitch,
-        ::size_t * slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapImage(
-            object_, buffer(), blocking, flags,
-            (const ::size_t *) origin, (const ::size_t *) region,
-            row_pitch, slice_pitch,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_IMAGE_ERR);
-        if (err != NULL) {
-              *err = error;
-        }
-        return result;
-    }
-
-    cl_int enqueueUnmapMemObject(
-        const Memory& memory,
-        void* mapped_ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueUnmapMemObject(
-                object_, memory(), mapped_ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_UNMAP_MEM_OBJECT_ERR);
-    }
-
-    cl_int enqueueNDRangeKernel(
-        const Kernel& kernel,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueNDRangeKernel(
-                object_, kernel(), (cl_uint) global.dimensions(),
-                offset.dimensions() != 0 ? (const ::size_t*) offset : NULL,
-                (const ::size_t*) global,
-                local.dimensions() != 0 ? (const ::size_t*) local : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NDRANGE_KERNEL_ERR);
-    }
-
-    cl_int enqueueTask(
-        const Kernel& kernel,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueTask(
-                object_, kernel(),
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_TASK_ERR);
-    }
-
-    cl_int enqueueNativeKernel(
-        void (*userFptr)(void *),
-        std::pair<void*, ::size_t> args,
-        const VECTOR_CLASS<Memory>* mem_objects = NULL,
-        const VECTOR_CLASS<const void*>* mem_locs = NULL,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        cl_mem * mems = (mem_objects != NULL && mem_objects->size() > 0) 
-            ? (cl_mem*) alloca(mem_objects->size() * sizeof(cl_mem))
-            : NULL;
-
-        if (mems != NULL) {
-            for (unsigned int i = 0; i < mem_objects->size(); i++) {
-                mems[i] = ((*mem_objects)[i])();
-            }
-        }
-
-        return detail::errHandler(
-            ::clEnqueueNativeKernel(
-                object_, userFptr, args.first, args.second,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                mems,
-                (mem_locs != NULL) ? (const void **) &mem_locs->front() : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NATIVE_KERNEL);
-    }
-
-    cl_int enqueueMarker(Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueMarker(object_, (cl_event*) event),
-            __ENQUEUE_MARKER_ERR);
-    }
-
-    cl_int enqueueWaitForEvents(const VECTOR_CLASS<Event>& events) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWaitForEvents(
-                object_,
-                (cl_uint) events.size(),
-                (const cl_event*) &events.front()),
-            __ENQUEUE_WAIT_FOR_EVENTS_ERR);
-    }
-
-    cl_int enqueueAcquireGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueAcquireGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueReleaseGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_RELEASE_GL_ERR);
-     }
-
-#if defined (USE_DX_INTEROP)
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueAcquireD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects, cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueReleaseD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects,  cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-
-    cl_int enqueueAcquireD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         static PFN_clEnqueueAcquireD3D10ObjectsKHR pfn_clEnqueueAcquireD3D10ObjectsKHR = NULL;
-         __INIT_CL_EXT_FCN_PTR(clEnqueueAcquireD3D10ObjectsKHR);
-		
-         return detail::errHandler(
-             pfn_clEnqueueAcquireD3D10ObjectsKHR(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-    {
-        static PFN_clEnqueueReleaseD3D10ObjectsKHR pfn_clEnqueueReleaseD3D10ObjectsKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clEnqueueReleaseD3D10ObjectsKHR);
-
-        return detail::errHandler(
-            pfn_clEnqueueReleaseD3D10ObjectsKHR(
-                object_,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_RELEASE_GL_ERR);
-    }
-#endif
-
-    cl_int enqueueBarrier() const
-    {
-        return detail::errHandler(
-            ::clEnqueueBarrier(object_),
-            __ENQUEUE_BARRIER_ERR);
-    }
-
-    cl_int flush() const
-    {
-        return detail::errHandler(::clFlush(object_), __FLUSH_ERR);
-    }
-
-    cl_int finish() const
-    {
-        return detail::errHandler(::clFinish(object_), __FINISH_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::CommandQueue)
-
-/*! \class KernelFunctor
- * \brief Kernel functor interface
- *
- * \note Currently only functors of zero to ten arguments are supported. It
- * is straightforward to add more and a more general solution, similar to
- * Boost.Lambda could be followed if required in the future.
- */
-class KernelFunctor
-{
-private:
-    Kernel kernel_;
-    CommandQueue queue_;
-    NDRange offset_;
-    NDRange global_;
-    NDRange local_;
-
-    cl_int err_;
-public:
-    KernelFunctor() { }
-
-    KernelFunctor(
-        const Kernel& kernel,
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local) :
-            kernel_(kernel),
-            queue_(queue),
-            offset_(offset),
-            global_(global),
-            local_(local),
-            err_(CL_SUCCESS)
-    {}
-
-    KernelFunctor& operator=(const KernelFunctor& rhs);
-
-    KernelFunctor(const KernelFunctor& rhs);
-
-    cl_int getError() { return err_; }
-
-    inline Event operator()(const VECTOR_CLASS<Event>* events = NULL);
-
-    template<typename A1>
-    inline Event operator()(
-        const A1& a1, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5, class A6>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4,
-             class A5, class A6, class A7>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12, 
-        const A13& a13,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14, class A15>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14, 
-        const A15& a15,
-        const VECTOR_CLASS<Event>* events = NULL);
-};
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& offset,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,offset,global,local);
-}
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,NullRange,global,local);
-}
-
-inline KernelFunctor& KernelFunctor::operator=(const KernelFunctor& rhs)
-{
-    if (this == &rhs) {
-        return *this;
-    }
-    
-    kernel_ = rhs.kernel_;
-    queue_  = rhs.queue_;
-    offset_ = rhs.offset_;
-    global_ = rhs.global_;
-    local_  = rhs.local_;
-    
-    return *this;
-}
-
-inline KernelFunctor::KernelFunctor(const KernelFunctor& rhs) :
-    kernel_(rhs.kernel_),
-    queue_(rhs.queue_),
-    offset_(rhs.offset_),
-    global_(rhs.global_),
-    local_(rhs.local_)
-{
-}
-
-Event KernelFunctor::operator()(const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4,
-         typename A5, typename A6, typename A7>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9, typename A10>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12, 
-    const A13& a13,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14, class A15>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14, 
-    const A15& a15,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-    kernel_.setArg(14,a15);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-#undef __ERR_STR
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#undef __GET_DEVICE_INFO_ERR
-#undef __GET_PLATFORM_INFO_ERR
-#undef __GET_DEVICE_IDS_ERR
-#undef __GET_CONTEXT_INFO_ERR
-#undef __GET_EVENT_INFO_ERR
-#undef __GET_EVENT_PROFILE_INFO_ERR
-#undef __GET_MEM_OBJECT_INFO_ERR
-#undef __GET_IMAGE_INFO_ERR
-#undef __GET_SAMPLER_INFO_ERR
-#undef __GET_KERNEL_INFO_ERR
-#undef __GET_KERNEL_WORK_GROUP_INFO_ERR
-#undef __GET_PROGRAM_INFO_ERR
-#undef __GET_PROGRAM_BUILD_INFO_ERR
-#undef __GET_COMMAND_QUEUE_INFO_ERR
-
-#undef __CREATE_CONTEXT_FROM_TYPE_ERR
-#undef __GET_SUPPORTED_IMAGE_FORMATS_ERR
-
-#undef __CREATE_BUFFER_ERR
-#undef __CREATE_SUBBUFFER_ERR
-#undef __CREATE_IMAGE2D_ERR
-#undef __CREATE_IMAGE3D_ERR
-#undef __CREATE_SAMPLER_ERR
-#undef __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR
-
-#undef __CREATE_USER_EVENT_ERR
-#undef __SET_USER_EVENT_STATUS_ERR
-#undef __SET_EVENT_CALLBACK_ERR
-
-#undef __WAIT_FOR_EVENTS_ERR
-
-#undef __CREATE_KERNEL_ERR
-#undef __SET_KERNEL_ARGS_ERR
-#undef __CREATE_PROGRAM_WITH_SOURCE_ERR
-#undef __CREATE_PROGRAM_WITH_BINARY_ERR
-#undef __BUILD_PROGRAM_ERR
-#undef __CREATE_KERNELS_IN_PROGRAM_ERR
-
-#undef __CREATE_COMMAND_QUEUE_ERR
-#undef __SET_COMMAND_QUEUE_PROPERTY_ERR
-#undef __ENQUEUE_READ_BUFFER_ERR
-#undef __ENQUEUE_WRITE_BUFFER_ERR
-#undef __ENQUEUE_READ_BUFFER_RECT_ERR
-#undef __ENQUEUE_WRITE_BUFFER_RECT_ERR
-#undef __ENQEUE_COPY_BUFFER_ERR
-#undef __ENQEUE_COPY_BUFFER_RECT_ERR
-#undef __ENQUEUE_READ_IMAGE_ERR
-#undef __ENQUEUE_WRITE_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR
-#undef __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR
-#undef __ENQUEUE_MAP_BUFFER_ERR
-#undef __ENQUEUE_MAP_IMAGE_ERR
-#undef __ENQUEUE_UNMAP_MEM_OBJECT_ERR
-#undef __ENQUEUE_NDRANGE_KERNEL_ERR
-#undef __ENQUEUE_TASK_ERR
-#undef __ENQUEUE_NATIVE_KERNEL
-
-#undef __UNLOAD_COMPILER_ERR
-#endif //__CL_USER_OVERRIDE_ERROR_STRINGS
-
-#undef __GET_INFO_HELPER_WITH_RETAIN
-
-// Extensions
-#undef __INIT_CL_EXT_FCN_PTR
-#undef __CREATE_SUB_DEVICES
-
-#if defined(USE_CL_DEVICE_FISSION)
-#undef __PARAM_NAME_DEVICE_FISSION
-#endif // USE_CL_DEVICE_FISSION
-
-} // namespace cl
-
-#endif // CL_HPP_
diff --git a/SpeedComparisons/GrayScott_OpenCL/gray_scott_opencl.cpp b/SpeedComparisons/GrayScott_OpenCL/gray_scott_opencl.cpp
deleted file mode 100644
index a7a70250c..000000000
--- a/SpeedComparisons/GrayScott_OpenCL/gray_scott_opencl.cpp
+++ /dev/null
@@ -1,271 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCL:
-#define __NO_STD_VECTOR // Use cl::vector instead of STL version
-#define __CL_ENABLE_EXCEPTIONS
-
-// cl.hpp is standard but doesn't come with most SDKs, so download it from here:
-// http://www.khronos.org/registry/cl/api/1.1/cl.hpp
-#ifdef __APPLE__
-# include "cl.hpp"
-#else
-# include <CL/cl.hpp>
-#endif
-
-using namespace cl;
-
-// STL:
-#include <fstream>
-#include <iostream>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-static int g_opt_device = 0;
-static int g_wrap = 1;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if ((i+1<argc) && (strcmp(argv[i],"-device")==0)) {
-            // select an output device
-            i++; g_opt_device = atoi(argv[i]);
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            std::cout << "Unrecognized argument: '" << argv[i] << "'\n";
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    float k, f;
-     k = 0.064f; f = 0.035f; // solitons with mitosis (spots that multiply)
-    // k = 0.06f; f = 0.035f; // stripes
-    // k = 0.065f; f = 0.056f; // long stripes
-    // k = 0.064f; f = 0.04f; // dots and stripes
-    // k = 0.0475f; f = 0.0118f; // spiral waves
-    float speed = 2.0f;
-    // ----------------
-    
-    float u[X][Y], v[X][Y];
-    const int MEM_SIZE = sizeof(float)*X*Y;
-
-    // put the initial conditions into each cell
-    init(u,v);
-
-    try { 
-        // Get available OpenCL platforms
-        vector<Platform> platforms;
-        Platform::get(&platforms);
- 
-        // Select the default platform and create a context using this platform and the GPU
-        cl_context_properties cps[3] = { 
-            CL_CONTEXT_PLATFORM, 
-            (cl_context_properties)(platforms[0])(), 
-            0 
-        };
-        Context context( CL_DEVICE_TYPE_GPU, cps);
- 
-        // Get a list of devices on this platform
-        vector<Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
-
-        // range-check the user's selection
-        int maxdev = devices.size() - 1;
-        g_opt_device = (g_opt_device > maxdev) ? maxdev :
-                                     ((g_opt_device < 0) ? 0 : g_opt_device);
-        std::cout << (maxdev+1) << " device(s) available; using device "
-                                                    << g_opt_device << ".\n";
-
-        Device &device = devices[g_opt_device];
-        std::cout << "Global memory: " << device.getInfo<CL_DEVICE_GLOBAL_MEM_SIZE>() << " bytes\n";
-        std::cout << "Local memory: " << device.getInfo<CL_DEVICE_LOCAL_MEM_SIZE>() << " bytes\n";
-        std::cout << "Local memory type: " << std::string((device.getInfo<CL_DEVICE_LOCAL_MEM_TYPE>()==CL_LOCAL)?"local":"global") << " \n";
-        std::cout << "CL_DEVICE_MAX_WORK_GROUP_SIZE: " << device.getInfo<CL_DEVICE_MAX_WORK_GROUP_SIZE>() << "\n";
-
-        // Create a command queue and use the selected device
-        if (maxdev < 0) {
-          std::cerr << "error -- need at least one OpenCL capable device.\n";
-          exit(-1);
-        }
-        CommandQueue queue = CommandQueue(context, device);
-        Event event;
- 
-        // Read source file
-        std::string kfn = CL_SOURCE_DIR; // (defined in CMakeLists.txt to be the source folder)
-        kfn += "/grayscott_kernel.cl";
-        std::ifstream sourceFile(kfn.c_str());
-        std::string sourceCode(
-            std::istreambuf_iterator<char>(sourceFile),
-            (std::istreambuf_iterator<char>()));
-        Program::Sources source(1, std::make_pair(sourceCode.c_str(), sourceCode.length()+1));
- 
-        // Make program of the source code in the context
-        Program program = Program(context, source);
- 
-        // Build program for these specific devices
-        // If wrap (toroidal) option is selected, we define a preprocessor flag
-        // that controls how the xm1, xp1, etc. are computed.
-        program.build(devices, g_wrap ? "-D WRAP" : NULL, NULL, NULL);
- 
-        // Make kernel
-        Kernel kernel(program, "grayscott_compute");
-
-        std::cout << "CL_KERNEL_WORK_GROUP_SIZE: " << kernel.getWorkGroupInfo<CL_KERNEL_WORK_GROUP_SIZE>(device) << "\n";
-
-        // Create memory buffers
-        Buffer bufferU = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferV = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferU2 = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferV2 = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-
-        // Copy lists A and B to the memory buffers
-        queue.enqueueWriteBuffer(bufferU, CL_TRUE, 0, MEM_SIZE, u);
-        queue.enqueueWriteBuffer(bufferV, CL_TRUE, 0, MEM_SIZE, v);
- 
-        NDRange global(X,Y);
-        NDRange local(1,512);
-
-        kernel.setArg(4, k);
-        kernel.setArg(5, f);
-        kernel.setArg(6, r_a);
-        kernel.setArg(7, r_b);
-        kernel.setArg(8, speed);
-
-        int iteration = 0;
-        float fps_avg = 0.0; // decaying average of fps
-        const int N_FRAMES_PER_DISPLAY = 2000;  // an even number, because of our double-buffering implementation
-        while(true) 
-        {
-            struct timeval tod_record;
-            double tod_before, tod_after, tod_elap;
-
-            gettimeofday(&tod_record, 0);
-            tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            // run a few iterations (without copying the data back)
-            for(int it=0;it<N_FRAMES_PER_DISPLAY/2;it++)
-            {
-                // (buffer-switching)
-
-                kernel.setArg(0, bufferU);
-                kernel.setArg(1, bufferV);
-                kernel.setArg(2, bufferU2);
-                kernel.setArg(3, bufferV2); // output in A2,B2
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-
-                kernel.setArg(0, bufferU2);
-                kernel.setArg(1, bufferV2);
-                kernel.setArg(2, bufferU);
-                kernel.setArg(3, bufferV); // output in A,B
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-            }
-
-            // no colour -- just read the A pattern back into a buffer
-            queue.enqueueReadBuffer(bufferU, CL_TRUE, 0, MEM_SIZE, u);
-
-            gettimeofday(&tod_record, 0);
-            tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            tod_elap = tod_after - tod_before;
-
-            char msg[1000];
-            float fps = 0.0;     // frames per second
-            if (tod_elap > 0)
-                fps = ((float)N_FRAMES_PER_DISPLAY) / tod_elap;
-            // We display an exponential moving average of the fps measurement
-            fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-	        double Mcgs = (fps_avg * ((double)X) * ((double)Y)) / 1.0e6;
-            sprintf(msg,"GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-            // display:
-            {
-                int quitnow = display(u,u,u,iteration,false,200.0f,1,10,msg);
-                if (quitnow)
-                    break;
-            }
-        }
-    } 
-    catch(Error error) 
-    {
-       std::cout << error.what() << "(" << error.err() << ")" << std::endl;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            // start with a uniform field with an approximate circle in the middle
-            //if(hypot(i%20-10/*-X/2*/,j%20-10/*-Y/2*/)<=frand(2,5)) {
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5))
-            {
-                a[i][j] = frand(0.0f,0.1f);
-                b[i][j] = frand(0.9f,1.0f);
-            }
-            else 
-            {
-                a[i][j] = frand(0.9f,1.0f);
-                b[i][j] = frand(0.0f,0.1f);
-            }
-            /*float v = frand(0.0f,1.0f);
-            a[i][j] = v;
-            b[i][j] = 1.0f-v;*/
-        }
-    }
-}
-
diff --git a/SpeedComparisons/GrayScott_OpenCL/grayscott_kernel.cl b/SpeedComparisons/GrayScott_OpenCL/grayscott_kernel.cl
deleted file mode 100644
index 6ce342a9d..000000000
--- a/SpeedComparisons/GrayScott_OpenCL/grayscott_kernel.cl
+++ /dev/null
@@ -1,68 +0,0 @@
-__kernel void grayscott_compute(
-    __global float *U,__global float *V,
-    __global float *U2, __global float *V2,
-    float k,float F,float D_u,float D_v,float delta_t)
-{
-    // Get the index of the current element.
-    const int x = get_global_id(0);
-    const int y = get_global_id(1);
-    const int X = get_global_size(0);
-    const int Y = get_global_size(1);
-    const int i = x*Y+y;
-
-    const float u = U[i];
-    const float v = V[i];
-    
-    // compute the Laplacians of u and v
-#ifdef WRAP
-    // speedy modulo operator for when X and Y are powers of 2 (we'll enforce this)
-    // http://forums.amd.com/devforum/messageview.cfm?catid=390&threadid=143648
-    const int xm1 = ((x-1+X) & (X-1));
-    const int xp1 = ((x+1) & (X-1));
-    const int ym1 = ((y-1+Y) & (Y-1));
-    const int yp1 = ((y+1) & (Y-1));
-    //const int xm1 = ((x-1+X)%X);
-    //const int xp1 = ((x+1)%X);
-    //const int ym1 = ((y-1+Y)%Y);
-    //const int yp1 = ((y+1)%Y);
-#else
-    const int xm1 = max(x-1,0);
-    const int xp1 = min(x+1,X-1);
-    const int ym1 = max(y-1,0);
-    const int yp1 = min(y+1,Y-1);
-#endif
-    const int iLeft = xm1*Y + y;
-    const int iRight = xp1*Y + y;
-    const int iUp = x*Y + ym1;
-    const int iDown = x*Y + yp1;
-
-    // Standard 5-point stencil
-    const float nabla_u = U[iLeft] + U[iRight] + U[iUp] + U[iDown] - 4*u;
-    const float nabla_v = V[iLeft] + V[iRight] + V[iUp] + V[iDown] - 4*v;
-
-   // 9-point stencil of Arad et al. 1997
-   //    Arad, A Yakhot, G Ben-Dor. A Highly Accurate Numerical Solution
-   //    of a Biharmonic Equation. Numer. Meth. PDE, 13, pp. 375-397, 1997.
-   //    PDF at www.bgu.ac.il/~yakhot/homepage/publications/nmpde_4_97.pdf
-   //    (see page 379)
-   // gives more correct results (no vertical/horizontal bias) but slows down
-   // the kernal a lot mainly because of the extra accesses to get neighboring
-   // U[] and V[] values.
-   //
-   // const int iUpLeft = xm1*Y + ym1;
-   // const int iUpRight = xp1*Y + ym1;
-   // const int iDownLeft = xm1*Y + yp1;
-   // const int iDownRight = xp1*Y + yp1;
-   // const float nabla_u = (U[iLeft]+U[iRight]+U[iUp]+U[iDown])*2.0f/3.0f
-   // +(U[iUpLeft]+U[iUpRight]+U[iDownLeft]+U[iDownRight])/6.0f - 10.0f*u/3.0f;
-   // const float nabla_v = (V[iLeft]+V[iRight]+V[iUp]+V[iDown])*2.0f/3.0f
-   // +(V[iUpLeft]+V[iUpRight]+V[iDownLeft]+V[iDownRight])/6.0f - 10.0f*v/3.0f;
-
-    // compute the new rate of change
-    const float delta_u = D_u * nabla_u - u*v*v + F*(1.0f-u);
-    const float delta_v = D_v * nabla_v + u*v*v - (F+k)*v;
-
-    // apply the change (to the new buffer)
-    U2[i] = u + delta_t * delta_u;
-    V2[i] = v + delta_t * delta_v;
-}
diff --git a/SpeedComparisons/GrayScott_OpenCL_2x2/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenCL_2x2/CMakeLists.txt
deleted file mode 100644
index 96a8439f9..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_2x2/CMakeLists.txt
+++ /dev/null
@@ -1,27 +0,0 @@
-project(GrayScott_OpenCL_2x2)
-
-set(CMAKE_MODULE_PATH ${GrayScott_OpenCL_2x2_SOURCE_DIR})
-# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
-
-find_package(OpenCV REQUIRED)
-include_directories( ${OPENCV_INCLUDE_DIR})
-link_libraries( ${OpenCV_LIBS} )
-
-# only build the OpenCL version if OpenCL was found
-find_package ( OpenCL )
-if(OPENCL_FOUND)
-  include_directories ( ${OPENCL_INCLUDE_DIRS} )
-  link_libraries ( ${OPENCL_LIBRARIES} )
-
-  # tell the code where the .cl file will live
-  add_definitions(-DCL_SOURCE_DIR="${GrayScott_OpenCL_2x2_SOURCE_DIR}")
-
-  INCLUDE_DIRECTORIES( "../Display" )
-
-  add_executable(GrayScott_OpenCL_2x2
-    gray_scott_opencl_2x2.cpp 
-    grayscott_kernel_2x2.cl 
-    ../Display/defs.h
-  )
-endif()
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_2x2/FindOpenCL.cmake b/SpeedComparisons/GrayScott_OpenCL_2x2/FindOpenCL.cmake
deleted file mode 100644
index fde90efae..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_2x2/FindOpenCL.cmake
+++ /dev/null
@@ -1,79 +0,0 @@
-# - Try to find OpenCL
-# This module tries to find an OpenCL implementation on your system. It supports
-# AMD / ATI, Apple and NVIDIA implementations, but shoudl work, too.
-#
-# Once done this will define
-#  OPENCL_FOUND        - system has OpenCL
-#  OPENCL_INCLUDE_DIRS  - the OpenCL include directory
-#  OPENCL_LIBRARIES    - link these to use OpenCL
-#
-# WIN32 should work, but is untested
-
-FIND_PACKAGE( PackageHandleStandardArgs )
-
-SET (OPENCL_VERSION_STRING "0.1.0")
-SET (OPENCL_VERSION_MAJOR 0)
-SET (OPENCL_VERSION_MINOR 1)
-SET (OPENCL_VERSION_PATCH 0)
-
-IF (APPLE)
-
-  FIND_LIBRARY(OPENCL_LIBRARIES OpenCL DOC "OpenCL lib for OSX")
-  FIND_PATH(OPENCL_INCLUDE_DIRS OpenCL/cl.h DOC "Include for OpenCL on OSX")
-  FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS OpenCL/cl.hpp DOC "Include for OpenCL CPP bindings on OSX")
-
-ELSE (APPLE)
-
-	IF (WIN32)
-	
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h)
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp)
-	
-	    # The AMD SDK currently installs both x86 and x86_64 libraries
-	    # This is only a hack to find out architecture
-	    IF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86_64")
-	    ELSE (${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64")
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86")
-	    ENDIF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    FIND_LIBRARY(OPENCL_LIBRARIES OpenCL.lib ${OPENCL_LIB_DIR})
-	    
-	    GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-	    
-	    # On Win32 search relative to the library
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS "${_OPENCL_INC_CAND}")
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS "${_OPENCL_INC_CAND}")
-	
-	ELSE (WIN32)
-
-            # Unix style platforms
-            FIND_LIBRARY(OPENCL_LIBRARIES OpenCL
-              ENV LD_LIBRARY_PATH
-            )
-
-            GET_FILENAME_COMPONENT(OPENCL_LIB_DIR ${OPENCL_LIBRARIES} PATH)
-            GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-
-            # The AMD SDK currently does not place its headers
-            # in /usr/include, therefore also search relative
-            # to the library
-            FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-            FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-
-	ENDIF (WIN32)
-
-ENDIF (APPLE)
-
-FIND_PACKAGE_HANDLE_STANDARD_ARGS( OpenCL DEFAULT_MSG OPENCL_LIBRARIES OPENCL_INCLUDE_DIRS )
-
-IF( _OPENCL_CPP_INCLUDE_DIRS )
-	SET( OPENCL_HAS_CPP_BINDINGS TRUE )
-	LIST( APPEND OPENCL_INCLUDE_DIRS ${_OPENCL_CPP_INCLUDE_DIRS} )
-	# This is often the same, so clean up
-	LIST( REMOVE_DUPLICATES OPENCL_INCLUDE_DIRS )
-ENDIF( _OPENCL_CPP_INCLUDE_DIRS )
-
-MARK_AS_ADVANCED(
-  OPENCL_INCLUDE_DIRS
-)
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_2x2/cl.hpp b/SpeedComparisons/GrayScott_OpenCL_2x2/cl.hpp
deleted file mode 100644
index 99b86a665..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_2x2/cl.hpp
+++ /dev/null
@@ -1,4011 +0,0 @@
-/*******************************************************************************
- * Copyright (c) 2008-2010 The Khronos Group Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and/or associated documentation files (the
- * "Materials"), to deal in the Materials without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sublicense, and/or sell copies of the Materials, and to
- * permit persons to whom the Materials are furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice shall be included
- * in all copies or substantial portions of the Materials.
- *
- * THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
- * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
- * CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
- * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
- * MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
- ******************************************************************************/
-
-/*! \file
- *
- *   \brief C++ bindings for OpenCL 1.0 (rev 48) and OpenCL 1.1 (rev 33)    
- *   \author Benedict R. Gaster and Laurent Morichetti
- *   
- *   Additions and fixes from Brian Cole, March 3rd 2010.
- *   
- *   \version 1.1
- *   \date June 2010
- *
- *   Optional extension support
- *
- *         cl
- *         cl_ext_device_fission
- *				#define USE_CL_DEVICE_FISSION
- */
-
-/*! \mainpage
- * \section intro Introduction
- * For many large applications C++ is the language of choice and so it seems
- * reasonable to define C++ bindings for OpenCL.
- *
- *
- * The interface is contained with a single C++ header file \em cl.hpp and all
- * definitions are contained within the namespace \em cl. There is no additional
- * requirement to include \em cl.h and to use either the C++ or original C
- * bindings it is enough to simply include \em cl.hpp.
- *
- * The bindings themselves are lightweight and correspond closely to the
- * underlying C API. Using the C++ bindings introduces no additional execution
- * overhead.
- *
- * For detail documentation on the bindings see:
- *
- * The OpenCL C++ Wrapper API 1.1 (revision 04)
- *  http://www.khronos.org/registry/cl/specs/opencl-cplusplus-1.1.pdf
- *
- * \section example Example
- *
- * The following example shows a general use case for the C++
- * bindings, including support for the optional exception feature and
- * also the supplied vector and string classes, see following sections for
- * decriptions of these features.
- *
- * \code
- * #define __CL_ENABLE_EXCEPTIONS
- * 
- * #if defined(__APPLE__) || defined(__MACOSX)
- * #include <OpenCL/cl.hpp>
- * #else
- * #include <CL/cl.hpp>
- * #endif
- * #include <cstdio>
- * #include <cstdlib>
- * #include <iostream>
- * 
- *  const char * helloStr  = "__kernel void "
- *                           "hello(void) "
- *                           "{ "
- *                           "  "
- *                           "} ";
- * 
- *  int
- *  main(void)
- *  {
- *     cl_int err = CL_SUCCESS;
- *     try {
- *
- *       std::vector<cl::Platform> platforms;
- *       cl::Platform::get(&platforms);
- *       if (platforms.size() == 0) {
- *           std::cout << "Platform size 0\n";
- *           return -1;
- *       }
- *
- *       cl_context_properties properties[] = 
- *          { CL_CONTEXT_PLATFORM, (cl_context_properties)(platforms[0])(), 0};
- *       cl::Context context(CL_DEVICE_TYPE_CPU, properties); 
- * 
- *       std::vector<cl::Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
- * 
- *       cl::Program::Sources source(1,
- *           std::make_pair(helloStr,strlen(helloStr)));
- *       cl::Program program_ = cl::Program(context, source);
- *       program_.build(devices);
- * 
- *       cl::Kernel kernel(program_, "hello", &err);
- * 
- *       cl::Event event;
- *       cl::CommandQueue queue(context, devices[0], 0, &err);
- *       queue.enqueueNDRangeKernel(
- *           kernel, 
- *           cl::NullRange, 
- *           cl::NDRange(4,4),
- *           cl::NullRange,
- *           NULL,
- *           &event); 
- * 
- *       event.wait();
- *     }
- *     catch (cl::Error err) {
- *        std::cerr 
- *           << "ERROR: "
- *           << err.what()
- *           << "("
- *           << err.err()
- *           << ")"
- *           << std::endl;
- *     }
- * 
- *    return EXIT_SUCCESS;
- *  }
- * 
- * \endcode
- *
- */
-#ifndef CL_HPP_
-#define CL_HPP_
-
-#ifdef _WIN32
-#include <windows.h>
-#include <malloc.h>
-#if defined(USE_DX_INTEROP)
-#include <CL/cl_d3d10.h>
-#endif
-#endif // _WIN32
-
-// 
-#if defined(USE_CL_DEVICE_FISSION)
-#include <CL/cl_ext.h>
-#endif
-
-#if defined(__APPLE__) || defined(__MACOSX)
-#include <OpenGL/OpenGL.h>
-#include <OpenCL/opencl.h>
-#else
-#include <GL/gl.h>
-#include <CL/opencl.h>
-#endif // !__APPLE__
-
-#if !defined(CL_CALLBACK)
-#define CL_CALLBACK
-#endif //CL_CALLBACK
-
-#include <utility>
-
-#if !defined(__NO_STD_VECTOR)
-#include <vector>
-#endif
-
-#if !defined(__NO_STD_STRING)
-#include <string>
-#endif 
-
-#if defined(linux) || defined(__APPLE__) || defined(__MACOSX)
-# include <alloca.h>
-#endif // linux
-
-#include <cstring>
-
-/*! \namespace cl
- *
- * \brief The OpenCL C++ bindings are defined within this namespace.
- *
- */
-namespace cl {
-
-#define __INIT_CL_EXT_FCN_PTR(name) \
-    if(!pfn_##name) { \
-        pfn_##name = (PFN_##name) \
-            clGetExtensionFunctionAddress(#name); \
-        if(!pfn_##name) { \
-        } \
-    }
-
-class Program;
-class Device;
-class Context;
-class CommandQueue;
-class Memory;
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-#include <exception>
-/*! \class Error
- * \brief Exception class
- */
-class Error : public std::exception
-{
-private:
-    cl_int err_;
-    const char * errStr_;
-public:
-    /*! Create a new CL error exception for a given error code
-     *  and corresponding message.
-     */
-    Error(cl_int err, const char * errStr = NULL) : err_(err), errStr_(errStr)
-    {}
-
-    ~Error() throw() {}
-
-    /*! \brief Get error string associated with exception
-     *
-     * \return A memory pointer to the error message string.
-     */
-    virtual const char * what() const throw ()
-    {
-        if (errStr_ == NULL) {
-            return "empty";
-        }
-        else {
-            return errStr_;
-        }
-    }
-
-    /*! \brief Get error code associated with exception
-     *
-     *  \return The error code.
-     */
-    const cl_int err(void) const { return err_; }
-};
-
-#define __ERR_STR(x) #x
-#else
-#define __ERR_STR(x) NULL
-#endif // __CL_ENABLE_EXCEPTIONS
-
-//! \cond DOXYGEN_DETAIL
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#define __GET_DEVICE_INFO_ERR               __ERR_STR(clgetDeviceInfo)
-#define __GET_PLATFORM_INFO_ERR             __ERR_STR(clGetPlatformInfo)
-#define __GET_DEVICE_IDS_ERR                __ERR_STR(clGetDeviceIDs)
-#define __GET_PLATFORM_IDS_ERR              __ERR_STR(clGetPlatformIDs)
-#define __GET_CONTEXT_INFO_ERR              __ERR_STR(clGetContextInfo)
-#define __GET_EVENT_INFO_ERR                __ERR_STR(clGetEventInfo)
-#define __GET_EVENT_PROFILE_INFO_ERR        __ERR_STR(clGetEventProfileInfo)
-#define __GET_MEM_OBJECT_INFO_ERR           __ERR_STR(clGetMemObjectInfo)
-#define __GET_IMAGE_INFO_ERR                __ERR_STR(clGetImageInfo)
-#define __GET_SAMPLER_INFO_ERR              __ERR_STR(clGetSamplerInfo)
-#define __GET_KERNEL_INFO_ERR               __ERR_STR(clGetKernelInfo)
-#define __GET_KERNEL_WORK_GROUP_INFO_ERR    __ERR_STR(clGetKernelWorkGroupInfo)
-#define __GET_PROGRAM_INFO_ERR              __ERR_STR(clGetProgramInfo)
-#define __GET_PROGRAM_BUILD_INFO_ERR        __ERR_STR(clGetProgramBuildInfo)
-#define __GET_COMMAND_QUEUE_INFO_ERR        __ERR_STR(clGetCommandQueueInfo)
-
-#define __CREATE_CONTEXT_FROM_TYPE_ERR      __ERR_STR(clCreateContextFromType)
-#define __GET_SUPPORTED_IMAGE_FORMATS_ERR   __ERR_STR(clGetSupportedImageFormats)
-
-#define __CREATE_BUFFER_ERR                 __ERR_STR(clCreateBuffer)
-#define __CREATE_SUBBUFFER_ERR              __ERR_STR(clCreateSubBuffer)
-#define __CREATE_GL_BUFFER_ERR              __ERR_STR(clCreateFromGLBuffer)
-#define __GET_GL_OBJECT_INFO_ERR            __ERR_STR(clGetGLObjectInfo)
-#define __CREATE_IMAGE2D_ERR                __ERR_STR(clCreateImage2D)
-#define __CREATE_IMAGE3D_ERR                __ERR_STR(clCreateImage3D)
-#define __CREATE_SAMPLER_ERR                __ERR_STR(clCreateSampler)
-#define __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR __ERR_STR(clSetMemObjectDestructorCallback)
-
-#define __CREATE_USER_EVENT_ERR             __ERR_STR(clCreateUserEvent)
-#define __SET_USER_EVENT_STATUS_ERR         __ERR_STR(clSetUserEventStatus)
-#define __SET_EVENT_CALLBACK_ERR            __ERR_STR(clSetEventCallback)
-#define __WAIT_FOR_EVENTS_ERR               __ERR_STR(clWaitForEvents)
-
-#define __CREATE_KERNEL_ERR                 __ERR_STR(clCreateKernel)
-#define __SET_KERNEL_ARGS_ERR               __ERR_STR(clSetKernelArg)
-#define __CREATE_PROGRAM_WITH_SOURCE_ERR    __ERR_STR(clCreateProgramWithSource)
-#define __CREATE_PROGRAM_WITH_BINARY_ERR    __ERR_STR(clCreateProgramWithBinary)
-#define __BUILD_PROGRAM_ERR                 __ERR_STR(clBuildProgram)
-#define __CREATE_KERNELS_IN_PROGRAM_ERR     __ERR_STR(clCreateKernelsInProgram)
-
-#define __CREATE_COMMAND_QUEUE_ERR          __ERR_STR(clCreateCommandQueue)
-#define __SET_COMMAND_QUEUE_PROPERTY_ERR    __ERR_STR(clSetCommandQueueProperty)
-#define __ENQUEUE_READ_BUFFER_ERR           __ERR_STR(clEnqueueReadBuffer)
-#define __ENQUEUE_READ_BUFFER_RECT_ERR      __ERR_STR(clEnqueueReadBufferRect)
-#define __ENQUEUE_WRITE_BUFFER_ERR          __ERR_STR(clEnqueueWriteBuffer)
-#define __ENQUEUE_WRITE_BUFFER_RECT_ERR     __ERR_STR(clEnqueueWriteBufferRect)
-#define __ENQEUE_COPY_BUFFER_ERR            __ERR_STR(clEnqueueCopyBuffer)
-#define __ENQEUE_COPY_BUFFER_RECT_ERR       __ERR_STR(clEnqueueCopyBufferRect)
-#define __ENQUEUE_READ_IMAGE_ERR            __ERR_STR(clEnqueueReadImage)
-#define __ENQUEUE_WRITE_IMAGE_ERR           __ERR_STR(clEnqueueWriteImage)
-#define __ENQUEUE_COPY_IMAGE_ERR            __ERR_STR(clEnqueueCopyImage)
-#define __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR  __ERR_STR(clEnqueueCopyImageToBuffer)
-#define __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR  __ERR_STR(clEnqueueCopyBufferToImage)
-#define __ENQUEUE_MAP_BUFFER_ERR            __ERR_STR(clEnqueueMapBuffer)
-#define __ENQUEUE_MAP_IMAGE_ERR             __ERR_STR(clEnqueueMapImage)
-#define __ENQUEUE_UNMAP_MEM_OBJECT_ERR      __ERR_STR(clEnqueueUnMapMemObject)
-#define __ENQUEUE_NDRANGE_KERNEL_ERR        __ERR_STR(clEnqueueNDRangeKernel)
-#define __ENQUEUE_TASK_ERR                  __ERR_STR(clEnqueueTask)
-#define __ENQUEUE_NATIVE_KERNEL             __ERR_STR(clEnqueueNativeKernel)
-#define __ENQUEUE_MARKER_ERR                __ERR_STR(clEnqueueMarker)
-#define __ENQUEUE_WAIT_FOR_EVENTS_ERR       __ERR_STR(clEnqueueWaitForEvents)
-#define __ENQUEUE_BARRIER_ERR               __ERR_STR(clEnqueueBarrier)
-
-#define __ENQUEUE_ACQUIRE_GL_ERR            __ERR_STR(clEnqueueAcquireGLObjects)
-#define __ENQUEUE_RELEASE_GL_ERR            __ERR_STR(clEnqueueReleaseGLObjects)
-
-#define __UNLOAD_COMPILER_ERR               __ERR_STR(clUnloadCompiler)
-
-#define __FLUSH_ERR                         __ERR_STR(clFlush)
-#define __FINISH_ERR                        __ERR_STR(clFinish)
-
-#define __CREATE_SUB_DEVICES                __ERR_STR(clCreateSubDevicesEXT)
-#endif // __CL_USER_OVERRIDE_ERROR_STRINGS
-//! \endcond
-
-/*! \class string
- * \brief Simple string class, that provides a limited subset of std::string
- * functionality but avoids many of the issues that come with that class.
- */
-class string
-{
-private:
-    ::size_t size_;
-    char * str_;
-public:
-    string(void) : size_(0), str_(NULL)
-    {
-    }
-
-    string(char * str, ::size_t size) :
-        size_(size),
-        str_(NULL)
-    {
-        str_ = new char[size_+1];
-        if (str_ != NULL) {
-            memcpy(str_, str, size_  * sizeof(char));
-            str_[size_] = '\0';
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string(char * str) :
-        str_(NULL)
-    {
-        size_= ::strlen(str);
-        str_ = new char[size_ + 1];
-        if (str_ != NULL) {
-            memcpy(str_, str, (size_ + 1) * sizeof(char));
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string& operator=(const string& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        if (rhs.size_ == 0 || rhs.str_ == NULL) {
-            size_ = 0;
-            str_  = NULL;
-        } 
-        else {
-            size_ = rhs.size_;
-            str_ = new char[size_ + 1];
-            if (str_ != NULL) {
-                memcpy(str_, rhs.str_, (size_ + 1) * sizeof(char));
-            }
-            else {
-                size_ = 0;
-            }
-        }
-
-        return *this;
-    }
-
-    string(const string& rhs)
-    {
-        *this = rhs;
-    }
-
-    ~string()
-    {
-        if (str_ != NULL) {
-            delete[] str_;
-        }
-    }
-
-    ::size_t size(void) const   { return size_; }
-    ::size_t length(void) const { return size(); }
-
-    const char * c_str(void) const { return (str_) ? str_ : "";}
-};
-
-#if !defined(__USE_DEV_STRING) && !defined(__NO_STD_STRING)
-#include <string>
-typedef std::string STRING_CLASS;
-#elif !defined(__USE_DEV_STRING) 
-typedef cl::string STRING_CLASS;
-#endif
-
-#if !defined(__USE_DEV_VECTOR) && !defined(__NO_STD_VECTOR)
-#include <vector>
-#define VECTOR_CLASS std::vector
-#elif !defined(__USE_DEV_VECTOR) 
-#define VECTOR_CLASS cl::vector 
-#endif
-
-#if !defined(__MAX_DEFAULT_VECTOR_SIZE)
-#define __MAX_DEFAULT_VECTOR_SIZE 10
-#endif
-
-/*! \class vector
- * \brief Fixed sized vector implementation that mirroring 
- * std::vector functionality.
- */
-template <typename T, unsigned int N = __MAX_DEFAULT_VECTOR_SIZE>
-class vector
-{
-private:
-    T data_[N];
-    unsigned int size_;
-    bool empty_;
-public:
-    vector() : 
-        size_(-1),
-        empty_(true)
-    {}
-
-    ~vector() {}
-
-    unsigned int size(void) const
-    {
-        return size_ + 1;
-    }
-
-    void clear()
-    {
-        size_ = -1;
-        empty_ = true;
-    }
-
-    void push_back (const T& x)
-    { 
-        if (size() < N) {
-            size_++;  
-            data_[size_] = x;
-            empty_ = false;
-        }
-    }
-
-    void pop_back(void)
-    {
-        if (!empty_) {
-            data_[size_].~T();
-            size_--;
-            if (size_ == -1) {
-                empty_ = true;
-            }
-        }
-    }
-  
-    vector(const vector<T, N>& vec) : 
-        size_(vec.size_),
-        empty_(vec.empty_)
-    {
-        if (!empty_) {
-            memcpy(&data_[0], &vec.data_[0], size() * sizeof(T));
-        }
-    } 
-
-    vector(unsigned int size, const T& val = T()) :
-        size_(-1),
-        empty_(true)
-    {
-        for (unsigned int i = 0; i < size; i++) {
-            push_back(val);
-        }
-    }
-
-    vector<T, N>& operator=(const vector<T, N>& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        size_  = rhs.size_;
-        empty_ = rhs.empty_;
-
-        if (!empty_) {	
-            memcpy(&data_[0], &rhs.data_[0], size() * sizeof(T));
-        }
-    
-        return *this;
-    }
-
-    bool operator==(vector<T,N> &vec)
-    {
-        if (empty_ && vec.empty_) {
-            return true;
-        }
-
-        if (size() != vec.size()) {
-            return false;
-        }
-
-        return memcmp(&data_[0], &vec.data_[0], size() * sizeof(T)) == 0 ? true : false;
-    }
-  
-    operator T* ()             { return data_; }
-    operator const T* () const { return data_; }
-   
-    bool empty (void) const
-    {
-        return empty_;
-    }
-  
-    unsigned int max_size (void) const
-    {
-        return N;
-    }
-
-    unsigned int capacity () const
-    {
-        return sizeof(T) * N;
-    }
-
-    T& operator[](int index)
-    {
-        return data_[index];
-    }
-  
-    T operator[](int index) const
-    {
-        return data_[index];
-    }
-  
-    template<class I>
-    void assign(I start, I end)
-    {
-        clear();   
-        while(start < end) {
-            push_back(*start);
-            start++;
-        }
-    }
-
-    /*! \class iterator
-     * \brief Iterator class for vectors
-     */
-    class iterator
-    {
-    private:
-        vector<T,N> vec_;
-        int index_;
-        bool initialized_;
-    public:
-        iterator(void) : 
-            index_(-1),
-            initialized_(false)
-        {
-            index_ = -1;
-            initialized_ = false;
-        }
-
-        ~iterator(void) {}
-
-        static iterator begin(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = 0;
-            }
-
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-
-        static iterator end(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = vec.size();
-            }
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-    
-        bool operator==(iterator i)
-        {
-            return ((vec_ == i.vec_) && 
-                    (index_ == i.index_) && 
-                    (initialized_ == i.initialized_));
-        }
-
-        bool operator!=(iterator i)
-        {
-            return (!(*this==i));
-        }
-
-        void operator++()
-        {
-            index_++;
-        }
-
-        void operator++(int x)
-        {
-            index_ += x;
-        }
-
-        void operator--()
-        {
-            index_--;
-        }
-
-        void operator--(int x)
-        {
-            index_ -= x;
-        }
-
-        T operator *()
-        {
-            return vec_[index_];
-        }
-    };
-
-    iterator begin(void)
-    {
-        return iterator::begin(*this);
-    }
-
-    iterator end(void)
-    {
-        return iterator::end(*this);
-    }
-
-    T& front(void)
-    {
-        return data_[0];
-    }
-
-    T& back(void)
-    {
-        return data_[size_];
-    }
-
-    const T& front(void) const
-    {
-        return data_[0];
-    }
-
-    const T& back(void) const
-    {
-        return data_[size_];
-    }
-};  
-    
-/*!
- * \brief size_t class used to interface between C++ and
- * OpenCL C calls that require arrays of size_t values, who's
- * size is known statically.
- */
-template <int N>
-struct size_t : public cl::vector< ::size_t, N> { };
-
-namespace detail {
-
-// GetInfo help struct
-template <typename Functor, typename T>
-struct GetInfoHelper
-{
-    static cl_int
-    get(Functor f, cl_uint name, T* param)
-    {
-        return f(name, sizeof(T), param, NULL);
-    }
-};
-
-// Specialized GetInfoHelper for VECTOR_CLASS params
-template <typename Func, typename T>
-struct GetInfoHelper<Func, VECTOR_CLASS<T> >
-{
-    static cl_int get(Func f, cl_uint name, VECTOR_CLASS<T>* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        T* value = (T*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        param->assign(&value[0], &value[required/sizeof(T)]);
-        return CL_SUCCESS;
-    }
-};
-
-// Specialized for getInfo<CL_PROGRAM_BINARIES>
-template <typename Func>
-struct GetInfoHelper<Func, VECTOR_CLASS<char *> >
-{
-    static cl_int
-    get(Func f, cl_uint name, VECTOR_CLASS<char *>* param)
-    {
-      cl_uint err = f(name, param->size() * sizeof(char *), &(*param)[0], NULL);
-      if (err != CL_SUCCESS) {
-        return err;
-      }
-      
-      return CL_SUCCESS;
-    }
-};
-
-// Specialized GetInfoHelper for STRING_CLASS params
-template <typename Func>
-struct GetInfoHelper<Func, STRING_CLASS>
-{
-    static cl_int get(Func f, cl_uint name, STRING_CLASS* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        char* value = (char*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        *param = value;
-        return CL_SUCCESS;
-    }
-};
-
-#define __GET_INFO_HELPER_WITH_RETAIN(CPP_TYPE) \
-namespace detail { \
-template <typename Func> \
-struct GetInfoHelper<Func, CPP_TYPE> \
-{ \
-    static cl_int get(Func f, cl_uint name, CPP_TYPE* param) \
-    { \
-      cl_uint err = f(name, sizeof(CPP_TYPE), param, NULL); \
-      if (err != CL_SUCCESS) { \
-        return err; \
-      } \
-      \
-      return ReferenceHandler<CPP_TYPE::cl_type>::retain((*param)()); \
-    } \
-}; \
-} 
-
-
-#define __PARAM_NAME_INFO_1_0(F) \
-    F(cl_platform_info, CL_PLATFORM_PROFILE, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VERSION, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_NAME, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VENDOR, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_device_info, CL_DEVICE_TYPE, cl_device_type) \
-    F(cl_device_info, CL_DEVICE_VENDOR_ID, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_COMPUTE_UNITS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_CLOCK_FREQUENCY, cl_uint) \
-    F(cl_device_info, CL_DEVICE_ADDRESS_BITS, cl_bitfield) \
-    F(cl_device_info, CL_DEVICE_MAX_READ_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WRITE_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_MEM_ALLOC_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_DEPTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE_SUPPORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_PARAMETER_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_SAMPLERS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MEM_BASE_ADDR_ALIGN, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_SINGLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_TYPE, cl_device_mem_cache_type) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE, cl_uint)\
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_TYPE, cl_device_local_mem_type) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_ERROR_CORRECTION_SUPPORT, cl_bool) \
-    F(cl_device_info, CL_DEVICE_PROFILING_TIMER_RESOLUTION, ::size_t) \
-    F(cl_device_info, CL_DEVICE_ENDIAN_LITTLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_COMPILER_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_EXECUTION_CAPABILITIES, cl_device_exec_capabilities) \
-    F(cl_device_info, CL_DEVICE_QUEUE_PROPERTIES, cl_command_queue_properties) \
-    F(cl_device_info, CL_DEVICE_PLATFORM, cl_platform_id) \
-    F(cl_device_info, CL_DEVICE_NAME, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VENDOR, STRING_CLASS) \
-    F(cl_device_info, CL_DRIVER_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_PROFILE, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_context_info, CL_CONTEXT_REFERENCE_COUNT, cl_uint) \
-    F(cl_context_info, CL_CONTEXT_DEVICES, VECTOR_CLASS<Device>) \
-    F(cl_context_info, CL_CONTEXT_PROPERTIES, VECTOR_CLASS<cl_context_properties>) \
-    \
-    F(cl_event_info, CL_EVENT_COMMAND_QUEUE, cl::CommandQueue) \
-    F(cl_event_info, CL_EVENT_COMMAND_TYPE, cl_command_type) \
-    F(cl_event_info, CL_EVENT_REFERENCE_COUNT, cl_uint) \
-    F(cl_event_info, CL_EVENT_COMMAND_EXECUTION_STATUS, cl_uint) \
-    \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_QUEUED, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_SUBMIT, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_START, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_END, cl_ulong) \
-    \
-    F(cl_mem_info, CL_MEM_TYPE, cl_mem_object_type) \
-    F(cl_mem_info, CL_MEM_FLAGS, cl_mem_flags) \
-    F(cl_mem_info, CL_MEM_SIZE, ::size_t) \
-    F(cl_mem_info, CL_MEM_HOST_PTR, void*) \
-    F(cl_mem_info, CL_MEM_MAP_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_REFERENCE_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_CONTEXT, cl::Context) \
-    \
-    F(cl_image_info, CL_IMAGE_FORMAT, cl_image_format) \
-    F(cl_image_info, CL_IMAGE_ELEMENT_SIZE, ::size_t) \
-    F(cl_image_info, CL_IMAGE_ROW_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_SLICE_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_WIDTH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_HEIGHT, ::size_t) \
-    F(cl_image_info, CL_IMAGE_DEPTH, ::size_t) \
-    \
-    F(cl_sampler_info, CL_SAMPLER_REFERENCE_COUNT, cl_uint) \
-    F(cl_sampler_info, CL_SAMPLER_CONTEXT, cl::Context) \
-    F(cl_sampler_info, CL_SAMPLER_NORMALIZED_COORDS, cl_addressing_mode) \
-    F(cl_sampler_info, CL_SAMPLER_ADDRESSING_MODE, cl_filter_mode) \
-    F(cl_sampler_info, CL_SAMPLER_FILTER_MODE, cl_bool) \
-    \
-    F(cl_program_info, CL_PROGRAM_REFERENCE_COUNT, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_CONTEXT, cl::Context) \
-    F(cl_program_info, CL_PROGRAM_NUM_DEVICES, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_DEVICES, VECTOR_CLASS<cl_device_id>) \
-    F(cl_program_info, CL_PROGRAM_SOURCE, STRING_CLASS) \
-    F(cl_program_info, CL_PROGRAM_BINARY_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_program_info, CL_PROGRAM_BINARIES, VECTOR_CLASS<char *>) \
-    \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_STATUS, cl_build_status) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_OPTIONS, STRING_CLASS) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_LOG, STRING_CLASS) \
-    \
-    F(cl_kernel_info, CL_KERNEL_FUNCTION_NAME, STRING_CLASS) \
-    F(cl_kernel_info, CL_KERNEL_NUM_ARGS, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_REFERENCE_COUNT, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_CONTEXT, cl::Context) \
-    F(cl_kernel_info, CL_KERNEL_PROGRAM, cl::Program) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_COMPILE_WORK_GROUP_SIZE, cl::size_t<3>) \
-    F(cl_kernel_work_group_info, CL_KERNEL_LOCAL_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_command_queue_info, CL_QUEUE_CONTEXT, cl::Context) \
-    F(cl_command_queue_info, CL_QUEUE_DEVICE, cl::Device) \
-    F(cl_command_queue_info, CL_QUEUE_REFERENCE_COUNT, cl_uint) \
-    F(cl_command_queue_info, CL_QUEUE_PROPERTIES, cl_command_queue_properties)
-
-#if defined(CL_VERSION_1_1)
-#define __PARAM_NAME_INFO_1_1(F) \
-    F(cl_context_info, CL_CONTEXT_NUM_DEVICES, cl_uint)\
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_DOUBLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HALF_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HOST_UNIFIED_MEMORY, cl_bool) \
-    \
-    F(cl_mem_info, CL_MEM_ASSOCIATED_MEMOBJECT, cl::Memory) \
-    F(cl_mem_info, CL_MEM_OFFSET, ::size_t) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_PRIVATE_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_event_info, CL_EVENT_CONTEXT, cl::Context)
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-#define __PARAM_NAME_DEVICE_FISSION(F) \
-    F(cl_device_info, CL_DEVICE_PARENT_DEVICE_EXT, cl_device_id) \
-	F(cl_device_info, CL_DEVICE_PARTITION_TYPES_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_AFFINITY_DOMAINS_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_REFERENCE_COUNT_EXT , cl_uint) \
-	F(cl_device_info, CL_DEVICE_PARTITION_STYLE_EXT, VECTOR_CLASS<cl_device_partition_property_ext>)
-#endif // USE_CL_DEVICE_FISSION
-
-template <typename enum_type, cl_int Name>
-struct param_traits {};
-
-#define __DECLARE_PARAM_TRAITS(token, param_name, T) \
-struct token;                                        \
-template<>                                           \
-struct param_traits<detail:: token,param_name>       \
-{                                                    \
-    enum { value = param_name };                     \
-    typedef T param_type;                            \
-};
-
-__PARAM_NAME_INFO_1_0(__DECLARE_PARAM_TRAITS);
-#if defined(CL_VERSION_1_1)
-__PARAM_NAME_INFO_1_1(__DECLARE_PARAM_TRAITS);
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-__PARAM_NAME_DEVICE_FISSION(__DECLARE_PARAM_TRAITS);
-#endif // USE_CL_DEVICE_FISSION
-
-#undef __DECLARE_PARAM_TRAITS
-
-// Convenience functions
-
-template <typename Func, typename T>
-inline cl_int
-getInfo(Func f, cl_uint name, T* param)
-{
-    return GetInfoHelper<Func, T>::get(f, name, param);
-}
-
-template <typename Func, typename Arg0>
-struct GetInfoFunctor0
-{
-    Func f_; const Arg0& arg0_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename Arg1>
-struct GetInfoFunctor1
-{
-    Func f_; const Arg0& arg0_; const Arg1& arg1_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, arg1_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, cl_uint name, T* param)
-{
-    GetInfoFunctor0<Func, Arg0> f0 = { f, arg0 };
-    return GetInfoHelper<GetInfoFunctor0<Func, Arg0>, T>
-        ::get(f0, name, param);
-}
-
-template <typename Func, typename Arg0, typename Arg1, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, const Arg1& arg1, cl_uint name, T* param)
-{
-    GetInfoFunctor1<Func, Arg0, Arg1> f0 = { f, arg0, arg1 };
-    return GetInfoHelper<GetInfoFunctor1<Func, Arg0, Arg1>, T>
-        ::get(f0, name, param);
-}
-
-template<typename T>
-struct ReferenceHandler
-{ };
-
-template <>
-struct ReferenceHandler<cl_device_id>
-{
-    // cl_device_id does not have retain().
-    static cl_int retain(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-    // cl_device_id does not have release().
-    static cl_int release(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-};
-
-template <>
-struct ReferenceHandler<cl_platform_id>
-{
-    // cl_platform_id does not have retain().
-    static cl_int retain(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-    // cl_platform_id does not have release().
-    static cl_int release(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-};
-
-template <>
-struct ReferenceHandler<cl_context>
-{
-    static cl_int retain(cl_context context)
-    { return ::clRetainContext(context); }
-    static cl_int release(cl_context context)
-    { return ::clReleaseContext(context); }
-};
-
-template <>
-struct ReferenceHandler<cl_command_queue>
-{
-    static cl_int retain(cl_command_queue queue)
-    { return ::clRetainCommandQueue(queue); }
-    static cl_int release(cl_command_queue queue)
-    { return ::clReleaseCommandQueue(queue); }
-};
-
-template <>
-struct ReferenceHandler<cl_mem>
-{
-    static cl_int retain(cl_mem memory)
-    { return ::clRetainMemObject(memory); }
-    static cl_int release(cl_mem memory)
-    { return ::clReleaseMemObject(memory); }
-};
-
-template <>
-struct ReferenceHandler<cl_sampler>
-{
-    static cl_int retain(cl_sampler sampler)
-    { return ::clRetainSampler(sampler); }
-    static cl_int release(cl_sampler sampler)
-    { return ::clReleaseSampler(sampler); }
-};
-
-template <>
-struct ReferenceHandler<cl_program>
-{
-    static cl_int retain(cl_program program)
-    { return ::clRetainProgram(program); }
-    static cl_int release(cl_program program)
-    { return ::clReleaseProgram(program); }
-};
-
-template <>
-struct ReferenceHandler<cl_kernel>
-{
-    static cl_int retain(cl_kernel kernel)
-    { return ::clRetainKernel(kernel); }
-    static cl_int release(cl_kernel kernel)
-    { return ::clReleaseKernel(kernel); }
-};
-
-template <>
-struct ReferenceHandler<cl_event>
-{
-    static cl_int retain(cl_event event)
-    { return ::clRetainEvent(event); }
-    static cl_int release(cl_event event)
-    { return ::clReleaseEvent(event); }
-};
-
-template <typename T>
-class Wrapper
-{
-public:
-    typedef T cl_type;
-
-protected:
-    cl_type object_;
-
-public:
-    Wrapper() : object_(NULL) { }
-
-    ~Wrapper()
-    {
-        if (object_ != NULL) { release(); }
-    }
-
-    Wrapper(const Wrapper<cl_type>& rhs)
-    {
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-    }
-
-    Wrapper<cl_type>& operator = (const Wrapper<cl_type>& rhs)
-    {
-        if (object_ != NULL) { release(); }
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-        return *this;
-    }
-
-    cl_type operator ()() const { return object_; }
-
-    cl_type& operator ()() { return object_; }
-
-protected:
-
-    cl_int retain() const
-    {
-        return ReferenceHandler<cl_type>::retain(object_);
-    }
-
-    cl_int release() const
-    {
-        return ReferenceHandler<cl_type>::release(object_);
-    }
-};
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-static inline cl_int errHandler (
-    cl_int err,
-    const char * errStr = NULL) throw(Error)
-{
-    if (err != CL_SUCCESS) {
-        throw Error(err, errStr);
-    }
-    return err;
-}
-#else
-static inline cl_int errHandler (cl_int err, const char * errStr = NULL)
-{
-    return err;
-}
-#endif // __CL_ENABLE_EXCEPTIONS
-
-} // namespace detail
-//! \endcond
-
-/*! \stuct ImageFormat
- * \brief ImageFormat interface fro cl_image_format.
- */
-struct ImageFormat : public cl_image_format
-{
-    ImageFormat(){}
-
-    ImageFormat(cl_channel_order order, cl_channel_type type)
-    {
-        image_channel_order = order;
-        image_channel_data_type = type;
-    }
-
-    ImageFormat& operator = (const ImageFormat& rhs)
-    {
-        if (this != &rhs) {
-            this->image_channel_data_type = rhs.image_channel_data_type;
-            this->image_channel_order     = rhs.image_channel_order;
-        }
-        return *this;
-    }
-};
-
-/*! \class Device
- * \brief Device interface for cl_device_id.
- */
-class Device : public detail::Wrapper<cl_device_id>
-{
-public:
-    Device(cl_device_id device) { object_ = device; }
-
-    Device() : detail::Wrapper<cl_type>() { }
-
-    Device(const Device& device) : detail::Wrapper<cl_type>(device) { }
-
-    Device& operator = (const Device& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_device_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetDeviceInfo, object_, name, param),
-            __GET_DEVICE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_device_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_device_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(USE_CL_DEVICE_FISSION)
-	cl_int createSubDevices(
-		const cl_device_partition_property_ext * properties,
-		VECTOR_CLASS<Device>* devices)
-	{
-		typedef CL_API_ENTRY cl_int 
-			( CL_API_CALL * PFN_clCreateSubDevicesEXT)(
-				cl_device_id /*in_device*/,
-                const cl_device_partition_property_ext * /* properties */,
-                cl_uint /*num_entries*/,
-                cl_device_id * /*out_devices*/,
-                cl_uint * /*num_devices*/ ) CL_EXT_SUFFIX__VERSION_1_1;
-
-		static PFN_clCreateSubDevicesEXT pfn_clCreateSubDevicesEXT = NULL;
-		__INIT_CL_EXT_FCN_PTR(clCreateSubDevicesEXT);
-
-		cl_uint n = 0;
-        cl_int err = pfn_clCreateSubDevicesEXT(object_, properties, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clCreateSubDevicesEXT(object_, properties, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
- 	}
-#endif
-};
-
-/*! \class Platform
- *  \brief Platform interface.
- */
-class Platform : public detail::Wrapper<cl_platform_id>
-{
-public:
-    static const Platform null();
-
-    Platform(cl_platform_id platform) { object_ = platform; }
-
-    Platform() : detail::Wrapper<cl_type>()  { }
-
-    Platform(const Platform& platform) : detail::Wrapper<cl_type>(platform) { }
-
-    Platform& operator = (const Platform& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getInfo(cl_platform_info name, STRING_CLASS* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetPlatformInfo, object_, name, param),
-            __GET_PLATFORM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_platform_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_platform_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getDevices(
-        cl_device_type type,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetDeviceIDs(object_, type, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = ::clGetDeviceIDs(object_, type, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-
-#if defined(USE_DX_INTEROP)
-   /*! \brief Get the list of available D3D10 devices.
-     *
-     *  \param d3d_device_source.
-     *
-     *  \param d3d_object.
-     *
-     *  \param d3d_device_set.
-     *
-     *  \param devices returns a vector of OpenCL D3D10 devices found. The cl::Device
-     *  values returned in devices can be used to identify a specific OpenCL
-     *  device. If \a devices argument is NULL, this argument is ignored.
-     *
-     *  \return One of the following values:
-     *    - CL_SUCCESS if the function is executed successfully.
-     *
-     *  The application can query specific capabilities of the OpenCL device(s)
-     *  returned by cl::getDevices. This can be used by the application to
-     *  determine which device(s) to use.
-     *
-     * \note In the case that exceptions are enabled and a return value
-     * other than CL_SUCCESS is generated, then cl::Error exception is
-     * generated.
-     */
-    cl_int getDevices(
-        cl_d3d10_device_source_khr d3d_device_source,
-        void *                     d3d_object,
-        cl_d3d10_device_set_khr    d3d_device_set,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clGetDeviceIDsFromD3D10KHR)(
-            cl_platform_id platform, 
-            cl_d3d10_device_source_khr d3d_device_source, 
-            void * d3d_object,
-            cl_d3d10_device_set_khr d3d_device_set,
-            cl_uint num_entries,
-            cl_device_id * devices,
-            cl_uint* num_devices);
-
-        static PFN_clGetDeviceIDsFromD3D10KHR pfn_clGetDeviceIDsFromD3D10KHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clGetDeviceIDsFromD3D10KHR);
-
-        cl_uint n = 0;
-        cl_int err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set, 
-            0, 
-            NULL, 
-            &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set,
-            n, 
-            ids, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-#endif
-
-    static cl_int get(
-        VECTOR_CLASS<Platform>* platforms)
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetPlatformIDs(0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        cl_platform_id* ids = (cl_platform_id*) alloca(
-            n * sizeof(cl_platform_id));
-        err = ::clGetPlatformIDs(n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        platforms->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-};
-
-static inline cl_int
-UnloadCompiler()
-{
-    return ::clUnloadCompiler();
-}
-
-class Context : public detail::Wrapper<cl_context>
-{
-public:
-    Context(
-        const VECTOR_CLASS<Device>& devices,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContext(
-            properties, (cl_uint) devices.size(),
-            (cl_device_id*) &devices.front(),
-            notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context(
-        cl_device_type type,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContextFromType(
-            properties, type, notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context() : detail::Wrapper<cl_type>() { }
-
-    Context(const Context& context) : detail::Wrapper<cl_type>(context) { }
-
-    Context& operator = (const Context& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_context_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetContextInfo, object_, name, param),
-            __GET_CONTEXT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_context_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_context_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getSupportedImageFormats(
-        cl_mem_flags flags,
-        cl_mem_object_type type,
-        VECTOR_CLASS<ImageFormat>* formats) const
-    {
-        cl_uint numEntries;
-        cl_int err = ::clGetSupportedImageFormats(
-           object_, 
-           flags,
-           type, 
-           0, 
-           NULL, 
-           &numEntries);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        ImageFormat* value = (ImageFormat*)
-            alloca(numEntries * sizeof(ImageFormat));
-        err = ::clGetSupportedImageFormats(
-            object_, 
-            flags, 
-            type, 
-            numEntries,
-            (cl_image_format*) value, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        formats->assign(&value[0], &value[numEntries]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Context)
-
-/*! \class Event
- * \brief Event interface for cl_event.
- */
-class Event : public detail::Wrapper<cl_event>
-{
-public:
-    Event() : detail::Wrapper<cl_type>() { }
-
-    Event(const Event& event) : detail::Wrapper<cl_type>(event) { }
-
-    Event& operator = (const Event& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_event_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetEventInfo, object_, name, param),
-            __GET_EVENT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_event_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_event_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getProfilingInfo(cl_profiling_info name, T* param) const
-    {
-        return detail::errHandler(detail::getInfo(
-            &::clGetEventProfilingInfo, object_, name, param),
-            __GET_EVENT_PROFILE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_profiling_info, name>::param_type
-    getProfilingInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_profiling_info, name>::param_type param;
-        cl_int result = getProfilingInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int wait() const
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(1, &object_),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setCallback(
-        cl_int type,
-        void (CL_CALLBACK * pfn_notify)(cl_event, cl_int, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetEventCallback(
-                object_,
-                type,
-                pfn_notify,
-                user_data), 
-            __SET_EVENT_CALLBACK_ERR);
-    }
-#endif
-
-    static cl_int
-    waitForEvents(const VECTOR_CLASS<Event>& events)
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(
-                (cl_uint) events.size(), (cl_event*)&events.front()),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Event)
-
-#if defined(CL_VERSION_1_1)
-/*! \class UserEvent
- * \brief User event interface for cl_event.
- */
-class UserEvent : public Event
-{
-public:
-    UserEvent(
-        const Context& context,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateUserEvent(
-            context(),
-            &error);
-
-        detail::errHandler(error, __CREATE_USER_EVENT_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    UserEvent() : Event() { }
-
-    UserEvent(const UserEvent& event) : Event(event) { }
-
-    UserEvent& operator = (const UserEvent& rhs)
-    {
-        if (this != &rhs) {
-            Event::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int setStatus(cl_int status)
-    {
-        return detail::errHandler(
-            ::clSetUserEventStatus(object_,status), 
-            __SET_USER_EVENT_STATUS_ERR);
-    }
-};
-#endif
-
-inline static cl_int
-WaitForEvents(const VECTOR_CLASS<Event>& events)
-{
-    return detail::errHandler(
-        ::clWaitForEvents(
-            (cl_uint) events.size(), (cl_event*)&events.front()),
-        __WAIT_FOR_EVENTS_ERR);
-}
-
-/*! \class Memory
- * \brief Memory interface for cl_mem.
- */
-class Memory : public detail::Wrapper<cl_mem>
-{
-public:
-    Memory() : detail::Wrapper<cl_type>() { }
-
-    Memory(const Memory& memory) : detail::Wrapper<cl_type>(memory) { }
-
-    Memory& operator = (const Memory& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_mem_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetMemObjectInfo, object_, name, param),
-            __GET_MEM_OBJECT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_mem_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_mem_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setDestructorCallback(
-        void (CL_CALLBACK * pfn_notify)(cl_mem, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetMemObjectDestructorCallback(
-                object_,
-                pfn_notify,
-                user_data), 
-            __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR);
-    }
-#endif
-
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Memory)
-
-/*! \class Buffer
- * \brief Memory buffer interface.
- */
-class Buffer : public Memory
-{
-public:
-    Buffer(
-        const Context& context,
-        cl_mem_flags flags,
-        ::size_t size,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateBuffer(context(), flags, size, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Buffer() : Memory() { }
-
-    Buffer(const Buffer& buffer) : Memory(buffer) { }
-
-    Buffer& operator = (const Buffer& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-
-#if defined(CL_VERSION_1_1)
-    Buffer createSubBuffer(
-        cl_mem_flags flags,
-        cl_buffer_create_type buffer_create_type,
-        const void * buffer_create_info,
-        cl_int * err = NULL)
-    {
-        Buffer result;
-        cl_int error;
-        result.object_ = ::clCreateSubBuffer(
-            object_, 
-            flags, 
-            buffer_create_type, 
-            buffer_create_info, 
-            &error);
-
-        detail::errHandler(error, __CREATE_SUBBUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-
-        return result;
-	}		
-#endif
-};
-
-#if defined (USE_DX_INTEROP)
-class BufferD3D10 : public Buffer
-{
-public:
-    typedef CL_API_ENTRY cl_mem (CL_API_CALL *PFN_clCreateFromD3D10BufferKHR)(
-    cl_context context, cl_mem_flags flags, ID3D10Buffer*  buffer,
-    cl_int* errcode_ret);
-
-    BufferD3D10(
-        const Context& context,
-        cl_mem_flags flags,
-        ID3D10Buffer* bufobj,
-        cl_int * err = NULL)
-    {
-        static PFN_clCreateFromD3D10BufferKHR pfn_clCreateFromD3D10BufferKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clCreateFromD3D10BufferKHR);
-
-        cl_int error;
-        object_ = pfn_clCreateFromD3D10BufferKHR(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferD3D10() : Buffer() { }
-
-    BufferD3D10(const BufferD3D10& buffer) : Buffer(buffer) { }
-
-    BufferD3D10& operator = (const BufferD3D10& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-};
-#endif
-
-/*! \class BufferGL
- * \brief Memory buffer interface for GL interop.
- */
-class BufferGL : public Buffer
-{
-public:
-    BufferGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLBuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferGL() : Buffer() { }
-
-    BufferGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferGL& operator = (const BufferGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class BufferRenderGL
- * \brief Memory buffer interface for GL interop with renderbuffer.
- */
-class BufferRenderGL : public Buffer
-{
-public:
-    BufferRenderGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLRenderbuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferRenderGL() : Buffer() { }
-
-    BufferRenderGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferRenderGL& operator = (const BufferRenderGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class Image
- * \brief Base class  interface for all images.
- */
-class Image : public Memory
-{
-protected:
-    Image() : Memory() { }
-
-    Image(const Image& image) : Memory(image) { }
-
-    Image& operator = (const Image& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-public:
-    template <typename T>
-    cl_int getImageInfo(cl_image_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetImageInfo, object_, name, param),
-            __GET_IMAGE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_image_info, name>::param_type
-    getImageInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_image_info, name>::param_type param;
-        cl_int result = getImageInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-/*! \class Image2D
- * \brief Image interface for 2D images.
- */
-class Image2D : public Image
-{
-public:
-    Image2D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t row_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage2D(
-            context(), flags,&format, width, height, row_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE2D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2D() { }
-
-    Image2D(const Image2D& image2D) : Image(image2D) { }
-
-    Image2D& operator = (const Image2D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image2DGL : public Image2D
-{
-public:
-    Image2DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture2D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2DGL() : Image2D() { }
-
-    Image2DGL(const Image2DGL& image) : Image2D(image) { }
-
-    Image2DGL& operator = (const Image2DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image2D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image3D
- * \brief Image interface for 3D images.
- */
-class Image3D : public Image
-{
-public:
-    Image3D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t depth,
-        ::size_t row_pitch = 0,
-        ::size_t slice_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage3D(
-            context(), flags, &format, width, height, depth, row_pitch,
-            slice_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE3D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3D() { }
-
-    Image3D(const Image3D& image3D) : Image(image3D) { }
-
-    Image3D& operator = (const Image3D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image3DGL : public Image3D
-{
-public:
-    Image3DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture3D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3DGL() : Image3D() { }
-
-    Image3DGL(const Image3DGL& image) : Image3D(image) { }
-
-    Image3DGL& operator = (const Image3DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image3D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Sampler
- * \brief Sampler interface for cl_sampler.
- */
-class Sampler : public detail::Wrapper<cl_sampler>
-{
-public:
-    Sampler() { }
-
-    Sampler(
-        const Context& context,
-        cl_bool normalized_coords,
-        cl_addressing_mode addressing_mode,
-        cl_filter_mode filter_mode,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateSampler(
-            context(), 
-            normalized_coords,
-            addressing_mode,
-            filter_mode,
-            &error);
-
-        detail::errHandler(error, __CREATE_SAMPLER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Sampler(const Sampler& sampler) : detail::Wrapper<cl_type>(sampler) { }
-
-    Sampler& operator = (const Sampler& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_sampler_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetSamplerInfo, object_, name, param),
-            __GET_SAMPLER_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_sampler_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_sampler_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Sampler)
-
-class Program;
-class CommandQueue;
-class Kernel;
-
-/*! \class NDRange
- * \brief NDRange interface
- */
-class NDRange
-{
-private:
-    size_t<3> sizes_;
-    cl_uint dimensions_;
-
-public:
-    NDRange()
-        : dimensions_(0)
-    { }
-
-    NDRange(::size_t size0)
-        : dimensions_(1)
-    {
-        sizes_.push_back(size0);
-    }
-
-    NDRange(::size_t size0, ::size_t size1)
-        : dimensions_(2)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-    }
-
-    NDRange(::size_t size0, ::size_t size1, ::size_t size2)
-        : dimensions_(3)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-        sizes_.push_back(size2);
-    }
-
-    operator const ::size_t*() const { return (const ::size_t*) sizes_; }
-    ::size_t dimensions() const { return dimensions_; }
-};
-
-static const NDRange NullRange;
-
-/*!
- * \struct LocalSpaceArg
- * \brief Local address raper for use with Kernel::setArg
- */
-struct LocalSpaceArg
-{
-    ::size_t size_;
-};
-
-namespace detail {
-
-template <typename T>
-struct KernelArgumentHandler
-{
-    static ::size_t size(const T&) { return sizeof(T); }
-    static T* ptr(T& value) { return &value; }
-};
-
-template <>
-struct KernelArgumentHandler<LocalSpaceArg>
-{
-    static ::size_t size(const LocalSpaceArg& value) { return value.size_; }
-    static void* ptr(LocalSpaceArg&) { return NULL; }
-};
-
-} 
-//! \endcond
-
-inline LocalSpaceArg
-__local(::size_t size)
-{
-    LocalSpaceArg ret = { size };
-    return ret;
-}
-
-class KernelFunctor;
-
-/*! \class Kernel
- * \brief Kernel interface that implements cl_kernel
- */
-class Kernel : public detail::Wrapper<cl_kernel>
-{
-public:
-    inline Kernel(const Program& program, const char* name, cl_int* err = NULL);
-
-    Kernel() { }
-
-    Kernel(const Kernel& kernel) : detail::Wrapper<cl_type>(kernel) { }
-
-    Kernel& operator = (const Kernel& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_kernel_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetKernelInfo, object_, name, param),
-            __GET_KERNEL_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_kernel_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getWorkGroupInfo(
-        const Device& device, cl_kernel_work_group_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetKernelWorkGroupInfo, object_, device(), name, param),
-                __GET_KERNEL_WORK_GROUP_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_work_group_info, name>::param_type
-        getWorkGroupInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-        detail::cl_kernel_work_group_info, name>::param_type param;
-        cl_int result = getWorkGroupInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int setArg(cl_uint index, T value)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(
-                object_,
-                index,
-                detail::KernelArgumentHandler<T>::size(value),
-                detail::KernelArgumentHandler<T>::ptr(value)),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    cl_int setArg(cl_uint index, ::size_t size, void* argPtr)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(object_, index, size, argPtr),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local);
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& global,
-        const NDRange& local);
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Kernel)
-
-/*! \class Program
- * \brief Program interface that implements cl_program.
- */
-class Program : public detail::Wrapper<cl_program>
-{
-public:
-    typedef VECTOR_CLASS<std::pair<const void*, ::size_t> > Binaries;
-    typedef VECTOR_CLASS<std::pair<const char*, ::size_t> > Sources;
-
-    Program(
-        const Context& context,
-        const Sources& sources,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-
-        const ::size_t n = (::size_t)sources.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const char** strings = (const char**) alloca(n * sizeof(const char*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            strings[i] = sources[(int)i].first;
-            lengths[i] = sources[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithSource(
-            context(), (cl_uint)n, strings, lengths, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_SOURCE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program(
-        const Context& context,
-        const VECTOR_CLASS<Device>& devices,
-        const Binaries& binaries,
-        VECTOR_CLASS<cl_int>* binaryStatus = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        const ::size_t n = binaries.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const unsigned char** images = (const unsigned char**) alloca(n * sizeof(const void*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            images[i] = (const unsigned char*)binaries[(int)i].first;
-            lengths[i] = binaries[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithBinary(
-            context(), (cl_uint) devices.size(),
-            (cl_device_id*)&devices.front(),
-            lengths, images, binaryStatus != NULL
-               ? (cl_int*) &binaryStatus->front()
-               : NULL, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_BINARY_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program() { }
-
-    Program(const Program& program) : detail::Wrapper<cl_type>(program) { }
-
-    Program& operator = (const Program& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int build(
-        const VECTOR_CLASS<Device>& devices,
-        const char* options = NULL,
-        void (CL_CALLBACK * notifyFptr)(cl_program, void *) = NULL,
-        void* data = NULL) const
-    {
-        return detail::errHandler(
-            ::clBuildProgram(
-                object_,
-                (cl_uint)
-                devices.size(),
-                (cl_device_id*)&devices.front(),
-                options,
-                notifyFptr,
-                data),
-                __BUILD_PROGRAM_ERR);
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_program_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetProgramInfo, object_, name, param),
-            __GET_PROGRAM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getBuildInfo(
-        const Device& device, cl_program_build_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetProgramBuildInfo, object_, device(), name, param),
-                __GET_PROGRAM_BUILD_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_build_info, name>::param_type
-    getBuildInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_build_info, name>::param_type param;
-        cl_int result = getBuildInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int createKernels(VECTOR_CLASS<Kernel>* kernels)
-    {
-        cl_uint numKernels;
-        cl_int err = ::clCreateKernelsInProgram(object_, 0, NULL, &numKernels);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        Kernel* value = (Kernel*) alloca(numKernels * sizeof(Kernel));
-        err = ::clCreateKernelsInProgram(
-            object_, numKernels, (cl_kernel*) value, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        kernels->assign(&value[0], &value[numKernels]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Program)
-
-inline Kernel::Kernel(const Program& program, const char* name, cl_int* err)
-{
-    cl_int error;
-
-    object_ = ::clCreateKernel(program(), name, &error);
-    detail::errHandler(error, __CREATE_KERNEL_ERR);
-
-    if (err != NULL) {
-        *err = error;
-    }
-
-}
-
-/*! \class CommandQueue
- * \brief CommandQueue interface for cl_command_queue.
- */
-class CommandQueue : public detail::Wrapper<cl_command_queue>
-{
-public:
-    CommandQueue(
-        const Context& context,
-        const Device& device,
-        cl_command_queue_properties properties = 0,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateCommandQueue(
-            context(), device(), properties, &error);
-
-        detail::errHandler(error, __CREATE_COMMAND_QUEUE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    CommandQueue() { }
-
-    CommandQueue(const CommandQueue& commandQueue) : detail::Wrapper<cl_type>(commandQueue) { }
-
-    CommandQueue& operator = (const CommandQueue& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_command_queue_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetCommandQueueInfo, object_, name, param),
-                __GET_COMMAND_QUEUE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_command_queue_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_command_queue_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int enqueueReadBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_BUFFER_ERR);
-    }
-
-    cl_int enqueueWriteBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        const void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBuffer(
-        const Buffer& src,
-        const Buffer& dst,
-        ::size_t src_offset,
-        ::size_t dst_offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBuffer(
-                object_, src(), dst(), src_offset, dst_offset, size,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int enqueueReadBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_READ_BUFFER_RECT_ERR);
-    }
-
-
-    cl_int enqueueWriteBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_RECT_ERR);
-    }
-
-    cl_int enqueueCopyBufferRect(
-        const Buffer& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        ::size_t src_row_pitch,
-        ::size_t src_slice_pitch,
-        ::size_t dst_row_pitch,
-        ::size_t dst_slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferRect(
-                object_, 
-                src(), 
-                dst(), 
-                (const ::size_t *)src_origin, 
-                (const ::size_t *)dst_origin, 
-                (const ::size_t *)region,
-                src_row_pitch,
-                src_slice_pitch,
-                dst_row_pitch,
-                dst_slice_pitch,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_RECT_ERR);
-    }
-#endif
-
-    cl_int enqueueReadImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_IMAGE_ERR);
-    }
-
-    cl_int enqueueWriteImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_WRITE_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImage(
-        const Image& src,
-        const Image& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImage(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *)dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImageToBuffer(
-        const Image& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& region,
-        ::size_t dst_offset,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImageToBuffer(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *) region, dst_offset,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBufferToImage(
-        const Buffer& src,
-        const Image& dst,
-        ::size_t src_offset,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferToImage(
-                object_, src(), dst(), src_offset,
-                (const ::size_t *) dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR);
-    }
-
-    void* enqueueMapBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        ::size_t offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapBuffer(
-            object_, buffer(), blocking, flags, offset, size,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-        return result;
-    }
-
-    void* enqueueMapImage(
-        const Image& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t * row_pitch,
-        ::size_t * slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapImage(
-            object_, buffer(), blocking, flags,
-            (const ::size_t *) origin, (const ::size_t *) region,
-            row_pitch, slice_pitch,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_IMAGE_ERR);
-        if (err != NULL) {
-              *err = error;
-        }
-        return result;
-    }
-
-    cl_int enqueueUnmapMemObject(
-        const Memory& memory,
-        void* mapped_ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueUnmapMemObject(
-                object_, memory(), mapped_ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_UNMAP_MEM_OBJECT_ERR);
-    }
-
-    cl_int enqueueNDRangeKernel(
-        const Kernel& kernel,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueNDRangeKernel(
-                object_, kernel(), (cl_uint) global.dimensions(),
-                offset.dimensions() != 0 ? (const ::size_t*) offset : NULL,
-                (const ::size_t*) global,
-                local.dimensions() != 0 ? (const ::size_t*) local : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NDRANGE_KERNEL_ERR);
-    }
-
-    cl_int enqueueTask(
-        const Kernel& kernel,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueTask(
-                object_, kernel(),
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_TASK_ERR);
-    }
-
-    cl_int enqueueNativeKernel(
-        void (*userFptr)(void *),
-        std::pair<void*, ::size_t> args,
-        const VECTOR_CLASS<Memory>* mem_objects = NULL,
-        const VECTOR_CLASS<const void*>* mem_locs = NULL,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        cl_mem * mems = (mem_objects != NULL && mem_objects->size() > 0) 
-            ? (cl_mem*) alloca(mem_objects->size() * sizeof(cl_mem))
-            : NULL;
-
-        if (mems != NULL) {
-            for (unsigned int i = 0; i < mem_objects->size(); i++) {
-                mems[i] = ((*mem_objects)[i])();
-            }
-        }
-
-        return detail::errHandler(
-            ::clEnqueueNativeKernel(
-                object_, userFptr, args.first, args.second,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                mems,
-                (mem_locs != NULL) ? (const void **) &mem_locs->front() : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NATIVE_KERNEL);
-    }
-
-    cl_int enqueueMarker(Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueMarker(object_, (cl_event*) event),
-            __ENQUEUE_MARKER_ERR);
-    }
-
-    cl_int enqueueWaitForEvents(const VECTOR_CLASS<Event>& events) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWaitForEvents(
-                object_,
-                (cl_uint) events.size(),
-                (const cl_event*) &events.front()),
-            __ENQUEUE_WAIT_FOR_EVENTS_ERR);
-    }
-
-    cl_int enqueueAcquireGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueAcquireGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueReleaseGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_RELEASE_GL_ERR);
-     }
-
-#if defined (USE_DX_INTEROP)
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueAcquireD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects, cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueReleaseD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects,  cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-
-    cl_int enqueueAcquireD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         static PFN_clEnqueueAcquireD3D10ObjectsKHR pfn_clEnqueueAcquireD3D10ObjectsKHR = NULL;
-         __INIT_CL_EXT_FCN_PTR(clEnqueueAcquireD3D10ObjectsKHR);
-		
-         return detail::errHandler(
-             pfn_clEnqueueAcquireD3D10ObjectsKHR(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-    {
-        static PFN_clEnqueueReleaseD3D10ObjectsKHR pfn_clEnqueueReleaseD3D10ObjectsKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clEnqueueReleaseD3D10ObjectsKHR);
-
-        return detail::errHandler(
-            pfn_clEnqueueReleaseD3D10ObjectsKHR(
-                object_,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_RELEASE_GL_ERR);
-    }
-#endif
-
-    cl_int enqueueBarrier() const
-    {
-        return detail::errHandler(
-            ::clEnqueueBarrier(object_),
-            __ENQUEUE_BARRIER_ERR);
-    }
-
-    cl_int flush() const
-    {
-        return detail::errHandler(::clFlush(object_), __FLUSH_ERR);
-    }
-
-    cl_int finish() const
-    {
-        return detail::errHandler(::clFinish(object_), __FINISH_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::CommandQueue)
-
-/*! \class KernelFunctor
- * \brief Kernel functor interface
- *
- * \note Currently only functors of zero to ten arguments are supported. It
- * is straightforward to add more and a more general solution, similar to
- * Boost.Lambda could be followed if required in the future.
- */
-class KernelFunctor
-{
-private:
-    Kernel kernel_;
-    CommandQueue queue_;
-    NDRange offset_;
-    NDRange global_;
-    NDRange local_;
-
-    cl_int err_;
-public:
-    KernelFunctor() { }
-
-    KernelFunctor(
-        const Kernel& kernel,
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local) :
-            kernel_(kernel),
-            queue_(queue),
-            offset_(offset),
-            global_(global),
-            local_(local),
-            err_(CL_SUCCESS)
-    {}
-
-    KernelFunctor& operator=(const KernelFunctor& rhs);
-
-    KernelFunctor(const KernelFunctor& rhs);
-
-    cl_int getError() { return err_; }
-
-    inline Event operator()(const VECTOR_CLASS<Event>* events = NULL);
-
-    template<typename A1>
-    inline Event operator()(
-        const A1& a1, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5, class A6>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4,
-             class A5, class A6, class A7>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12, 
-        const A13& a13,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14, class A15>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14, 
-        const A15& a15,
-        const VECTOR_CLASS<Event>* events = NULL);
-};
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& offset,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,offset,global,local);
-}
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,NullRange,global,local);
-}
-
-inline KernelFunctor& KernelFunctor::operator=(const KernelFunctor& rhs)
-{
-    if (this == &rhs) {
-        return *this;
-    }
-    
-    kernel_ = rhs.kernel_;
-    queue_  = rhs.queue_;
-    offset_ = rhs.offset_;
-    global_ = rhs.global_;
-    local_  = rhs.local_;
-    
-    return *this;
-}
-
-inline KernelFunctor::KernelFunctor(const KernelFunctor& rhs) :
-    kernel_(rhs.kernel_),
-    queue_(rhs.queue_),
-    offset_(rhs.offset_),
-    global_(rhs.global_),
-    local_(rhs.local_)
-{
-}
-
-Event KernelFunctor::operator()(const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4,
-         typename A5, typename A6, typename A7>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9, typename A10>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12, 
-    const A13& a13,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14, class A15>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14, 
-    const A15& a15,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-    kernel_.setArg(14,a15);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-#undef __ERR_STR
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#undef __GET_DEVICE_INFO_ERR
-#undef __GET_PLATFORM_INFO_ERR
-#undef __GET_DEVICE_IDS_ERR
-#undef __GET_CONTEXT_INFO_ERR
-#undef __GET_EVENT_INFO_ERR
-#undef __GET_EVENT_PROFILE_INFO_ERR
-#undef __GET_MEM_OBJECT_INFO_ERR
-#undef __GET_IMAGE_INFO_ERR
-#undef __GET_SAMPLER_INFO_ERR
-#undef __GET_KERNEL_INFO_ERR
-#undef __GET_KERNEL_WORK_GROUP_INFO_ERR
-#undef __GET_PROGRAM_INFO_ERR
-#undef __GET_PROGRAM_BUILD_INFO_ERR
-#undef __GET_COMMAND_QUEUE_INFO_ERR
-
-#undef __CREATE_CONTEXT_FROM_TYPE_ERR
-#undef __GET_SUPPORTED_IMAGE_FORMATS_ERR
-
-#undef __CREATE_BUFFER_ERR
-#undef __CREATE_SUBBUFFER_ERR
-#undef __CREATE_IMAGE2D_ERR
-#undef __CREATE_IMAGE3D_ERR
-#undef __CREATE_SAMPLER_ERR
-#undef __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR
-
-#undef __CREATE_USER_EVENT_ERR
-#undef __SET_USER_EVENT_STATUS_ERR
-#undef __SET_EVENT_CALLBACK_ERR
-
-#undef __WAIT_FOR_EVENTS_ERR
-
-#undef __CREATE_KERNEL_ERR
-#undef __SET_KERNEL_ARGS_ERR
-#undef __CREATE_PROGRAM_WITH_SOURCE_ERR
-#undef __CREATE_PROGRAM_WITH_BINARY_ERR
-#undef __BUILD_PROGRAM_ERR
-#undef __CREATE_KERNELS_IN_PROGRAM_ERR
-
-#undef __CREATE_COMMAND_QUEUE_ERR
-#undef __SET_COMMAND_QUEUE_PROPERTY_ERR
-#undef __ENQUEUE_READ_BUFFER_ERR
-#undef __ENQUEUE_WRITE_BUFFER_ERR
-#undef __ENQUEUE_READ_BUFFER_RECT_ERR
-#undef __ENQUEUE_WRITE_BUFFER_RECT_ERR
-#undef __ENQEUE_COPY_BUFFER_ERR
-#undef __ENQEUE_COPY_BUFFER_RECT_ERR
-#undef __ENQUEUE_READ_IMAGE_ERR
-#undef __ENQUEUE_WRITE_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR
-#undef __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR
-#undef __ENQUEUE_MAP_BUFFER_ERR
-#undef __ENQUEUE_MAP_IMAGE_ERR
-#undef __ENQUEUE_UNMAP_MEM_OBJECT_ERR
-#undef __ENQUEUE_NDRANGE_KERNEL_ERR
-#undef __ENQUEUE_TASK_ERR
-#undef __ENQUEUE_NATIVE_KERNEL
-
-#undef __UNLOAD_COMPILER_ERR
-#endif //__CL_USER_OVERRIDE_ERROR_STRINGS
-
-#undef __GET_INFO_HELPER_WITH_RETAIN
-
-// Extensions
-#undef __INIT_CL_EXT_FCN_PTR
-#undef __CREATE_SUB_DEVICES
-
-#if defined(USE_CL_DEVICE_FISSION)
-#undef __PARAM_NAME_DEVICE_FISSION
-#endif // USE_CL_DEVICE_FISSION
-
-} // namespace cl
-
-#endif // CL_HPP_
diff --git a/SpeedComparisons/GrayScott_OpenCL_2x2/gray_scott_opencl_2x2.cpp b/SpeedComparisons/GrayScott_OpenCL_2x2/gray_scott_opencl_2x2.cpp
deleted file mode 100644
index ecb7f0e07..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_2x2/gray_scott_opencl_2x2.cpp
+++ /dev/null
@@ -1,402 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCL:
-#define __NO_STD_VECTOR // Use cl::vector instead of STL version
-#define __CL_ENABLE_EXCEPTIONS
-
-// cl.hpp is standard but doesn't come with most SDKs, so download it from here:
-// http://www.khronos.org/registry/cl/api/1.1/cl.hpp
-#ifdef __APPLE__
-# include "cl.hpp"
-#else
-# include <CL/cl.hpp>
-#endif
-
-using namespace cl;
-
-// STL:
-#include <fstream>
-#include <iostream>
-
-// local:
-#include "defs.h"
-
-void init(float a[X][Y],float b[X][Y]);
-bool display(float r[X][Y],float g[X][Y],float b[X][Y],
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
-
-// we pack the values in 2x2 blocks:   x y
-//                                     z w
-float& float_at(float* arr,int x,int y) 
-{
-    return arr[ ( (x/2)*(Y/2) + y/2 ) * 4 + (y%2)*2 + x%2 ]; 
-}
-
-static int g_opt_device = 0;
-static int g_wrap = 1;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if ((i+1<argc) && (strcmp(argv[i],"-device")==0)) {
-            // select an output device
-            i++; g_opt_device = atoi(argv[i]);
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            std::cout << "Unrecognized argument: '" << argv[i] << "'\n";
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    float k, f;
-    k = 0.064f; f = 0.035f; // solitons with mitosis (spots that multiply)
-    // k = 0.06f; f = 0.035f; // stripes
-    // k = 0.065f; f = 0.056f; // long stripes
-    // k = 0.064f; f = 0.04f; // dots and stripes
-    // k = 0.0475f; f = 0.0118f; // spiral waves
-    float speed = 2.0f;
-    // ----------------
-    
-    float a[X][Y], b[X][Y];
-    const int MEM_SIZE = sizeof(float)*X*Y;
-
-    // put the initial conditions into each cell
-    init(a,b);
-
-    try { 
-        // Get available OpenCL platforms
-        cl::vector<Platform> platforms;
-        Platform::get(&platforms);
- 
-        // Select the default platform and create a context using this platform and the GPU
-        cl_context_properties cps[3] = { 
-            CL_CONTEXT_PLATFORM, 
-            (cl_context_properties)(platforms[0])(), 
-            0 
-        };
-        Context context( CL_DEVICE_TYPE_GPU, cps);
- 
-        // Get a list of devices on this platform
-        cl::vector<Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
-
-        int maxdev = devices.size() - 1;
-        g_opt_device = (g_opt_device > maxdev) ? maxdev :
-                                     ((g_opt_device < 0) ? 0 : g_opt_device);
-        std::cout << (maxdev+1) << " device(s) available; using device "
-                                                    << g_opt_device << ".\n";
-        Device &device = devices[g_opt_device];
-
-        CommandQueue queue = CommandQueue(context, device);
-        Event event;
- 
-        // Read source file
-        // (CL_SOURCE_DIR is defined in CMakeLists.txt to be the folder
-        // containing the source files, including this file 'gray_scott_opencl_2x2.cpp'
-        // and the kernel source 'gs_wrap_kernel_2x2.cl'.)
-        std::string kfn = CL_SOURCE_DIR;
-        kfn += "/grayscott_kernel_2x2.cl";
-        std::ifstream sourceFile(kfn.c_str());
-        std::string sourceCode(
-            std::istreambuf_iterator<char>(sourceFile),
-            (std::istreambuf_iterator<char>()));
-        Program::Sources source(1, std::make_pair(sourceCode.c_str(), sourceCode.length()+1));
-
-
-        // enable this code to display kernel compilation error if you get clBuildProgram(-11)
-        #if 0
-           const ::size_t n = (::size_t)source.size();
-           ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-           const char** strings = (const char**) alloca(n * sizeof(const char*));
-           for (::size_t i = 0; i < n; ++i) {
-               strings[i] = source[(int)i].first;
-               lengths[i] = source[(int)i].second;
-           }
-           cl_int err;
-           cl_program myprog = clCreateProgramWithSource(context(), (cl_uint)n, strings, lengths, &err);
-           err = clBuildProgram(myprog, (cl_uint)devices.size(), (cl_device_id*)&devices.front(), NULL, NULL, NULL);
-           char proglog[1024];
-           clGetProgramBuildInfo(myprog, device(), CL_PROGRAM_BUILD_LOG, 1024, proglog, 0);
-           printf("err=%d log=%s\n", err, proglog);
-           return 0;
-        #endif
-         
- 
-        // Make program of the source code in the context
-        Program program = Program(context, source);
- 
-        // Build program for these specific devices
-        // If wrap (toroidal) option is selected, we define a preprocessor flag
-        // that controls how the xm1, xp1, etc. are computed.
-        program.build(devices, g_wrap ? "-D WRAP" : NULL, NULL, NULL);
- 
-        // Make kernel
-        Kernel kernel(program, "grayscott_compute");
-
-        // Create memory buffers
-        Buffer bufferU = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferV = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferU2 = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferV2 = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-
-        // Copy lists A and B to the memory buffers
-        queue.enqueueWriteBuffer(bufferU, CL_TRUE, 0, MEM_SIZE, a);
-        queue.enqueueWriteBuffer(bufferV, CL_TRUE, 0, MEM_SIZE, b);
- 
-        NDRange global(X/2,Y/2);
-        NDRange local(1,128);
-
-        kernel.setArg(4, k);
-        kernel.setArg(5, f);
-        kernel.setArg(6, r_a);
-        kernel.setArg(7, r_b);
-        kernel.setArg(8, speed);
-
-        int iteration = 0;
-        float fps_avg = 0.0; // decaying average of fps
-        const int N_FRAMES_PER_DISPLAY = 2000;  // an even number, because of our double-buffering implementation
-        while(true) 
-        {
-            struct timeval tod_record;
-            double tod_before, tod_after, tod_elap;
-
-            gettimeofday(&tod_record, 0);
-            tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            // run a few iterations (without copying the data back)
-            for(int it=0;it<N_FRAMES_PER_DISPLAY/2;it++)
-            {
-                // (buffer-switching)
-
-                kernel.setArg(0, bufferU);
-                kernel.setArg(1, bufferV);
-                kernel.setArg(2, bufferU2);
-                kernel.setArg(3, bufferV2); // output in A2,B2
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-
-                kernel.setArg(0, bufferU2);
-                kernel.setArg(1, bufferV2);
-                kernel.setArg(2, bufferU);
-                kernel.setArg(3, bufferV); // output in A,B
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-            }
-
-            queue.enqueueReadBuffer(bufferU, CL_TRUE, 0, MEM_SIZE, a);
-
-            gettimeofday(&tod_record, 0);
-            tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            tod_elap = tod_after - tod_before;
-
-            char msg[1000];
-            float fps = 0.0;     // frames per second
-            if (tod_elap > 0)
-                fps = ((float)N_FRAMES_PER_DISPLAY) / tod_elap;
-            // We display an exponential moving average of the fps measurement
-            fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-                double Mcgs = (fps_avg * ((double)X) * ((double)Y)) / 1.0e6;
-            sprintf(msg,"GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-            // display:
-            if(display(a,a,a,iteration,false,200.0f,2,10,msg))
-                break;
-        }
-    } 
-    catch(Error error) 
-    {
-       std::cout << error.what() << "(" << error.err() << ")" << std::endl;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            // start with a uniform field with an approximate circle in the middle
-            //if(hypot(i%20-10/*-X/2*/,j%20-10/*-Y/2*/)<=frand(2,5)) {
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5))
-            {
-                float_at((float*)a,i,j) = frand(0.0f,0.1f);
-                float_at((float*)b,i,j) = frand(0.9f,1.0f);
-            }
-            else 
-            {
-                float_at((float*)a,i,j) = frand(0.9f,1.0f);
-                float_at((float*)b,i,j) = frand(0.0f,0.1f);
-            }
-        }
-    }
-}
-
-bool display(float r[X][Y],float g[X][Y],float b[X][Y],
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-        
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-        
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-        if(write_video)
-        {
-            video = cvCreateVideoWriter(title,CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im2),1);
-        }
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<X;i++)
-        {
-            for(int j=0;j<Y;j++)
-            {
-                if(r) {
-                    val = float_at((float*)r,i,j);
-                    if(val<minR) minR=val; if(val>maxR) maxR=val;
-                }
-                if(g) {
-                    val = float_at((float*)g,i,j);
-                    if(val<minG) minG=val; if(val>maxG) maxG=val;
-                }
-                if(b) {
-                    val = float_at((float*)b,i,j);
-                    if(val<minB) minB=val; if(val>maxB) maxB=val;
-                }
-            }
-        }
-    }
-    #pragma omp parallel for
-    for(int i=0;i<X;i++)
-    {
-        for(int j=0;j<Y;j++)
-        {
-            if(r) {
-                float val = float_at((float*)r,i,j);//Y-j-1);
-                if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 2] = (uchar)val;
-            }
-            if(g) {
-                float val = float_at((float*)g,i,j);//Y-j-1);
-                if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 1] = (uchar)val;
-            }
-            if(b) {
-                float val = float_at((float*)b,i,j);//Y-j-1);
-                if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 0] = (uchar)val;
-            }
-        }
-    }
-
-    cvResize(im,im2);
-
-    char txt[100];
-    if(!write_video)
-    {
-        sprintf(txt,"%d",iteration);
-        cvPutText(im2,txt,cvPoint(20,20),&font,white);
-    }
-
-    cvPutText(im2,message,cvPoint(20,40),&font,white);
-
-    if(write_video)
-        cvWriteFrame(video,im2);
-
-    cvShowImage(title,im2);
-    
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        if(write_video)
-            cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
diff --git a/SpeedComparisons/GrayScott_OpenCL_2x2/grayscott_kernel_2x2.cl b/SpeedComparisons/GrayScott_OpenCL_2x2/grayscott_kernel_2x2.cl
deleted file mode 100644
index 7b6cadd2c..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_2x2/grayscott_kernel_2x2.cl
+++ /dev/null
@@ -1,59 +0,0 @@
-__kernel void grayscott_compute(
-    __global float4 *U,__global float4 *V,
-    __global float4 *U2, __global float4 *V2,
-    float k,float F,float D_u,float D_v,float delta_t)
-{
-    const int x = get_global_id(0);
-    const int y = get_global_id(1);
-    const int X = get_global_size(0);
-    const int Y = get_global_size(1);
-    const int i = x*Y+y;
-
-    const float4 u = U[i];
-    const float4 v = V[i];
-    
-    // compute the Laplacians of a and b
-#ifdef WRAP
-    const int xm1 = ((x-1+X)%X);
-    const int xp1 = ((x+1)%X);
-    const int ym1 = ((y-1+Y)%Y);
-    const int yp1 = ((y+1)%Y);
-#else
-    const int xm1 = max(x-1,0);
-    const int xp1 = min(x+1,X-1);
-    const int ym1 = max(y-1,0);
-    const int yp1 = min(y+1,Y-1);
-#endif
-    const int iLeft = xm1*Y + y;
-    const int iRight = xp1*Y + y;
-    const int iUp = x*Y + ym1;
-    const int iDown = x*Y + yp1;
-    
-    const float4 u_left = U[iLeft];
-    const float4 u_right = U[iRight];
-    const float4 u_up = U[iUp];
-    const float4 u_down = U[iDown];
-    const float4 v_left = V[iLeft];
-    const float4 v_right = V[iRight];
-    const float4 v_up = V[iUp];
-    const float4 v_down = V[iDown];
-
-    const float4 nabla_u = (float4)(
-        u_left.y + u_up.z + u.y + u.z,
-        u.x + u_up.w + u_right.x + u.w,
-        u_left.w + u.x + u.w + u_down.x,
-        u.z + u.y + u_right.z + u_down.y) - 4.0f*u;
-    const float4 nabla_v = (float4)(
-        v_left.y + v_up.z + v.y + v.z,
-        v.x + v_up.w + v_right.x + v.w,
-        v_left.w + v.x + v.w + v_down.x,
-        v.z + v.y + v_right.z + v_down.y) - 4.0f*v;
-
-    // compute the new rate of change
-    const float4 delta_u = D_u * nabla_u - u*v*v + F*(1.0f-u);
-    const float4 delta_v = D_v * nabla_v + u*v*v - (F+k)*v;
-
-    // apply the change (to the new buffer)
-    U2[i] = u + delta_t * delta_u;
-    V2[i] = v + delta_t * delta_v;
-}
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenCL_Image/CMakeLists.txt
deleted file mode 100644
index 1ed3d9a95..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image/CMakeLists.txt
+++ /dev/null
@@ -1,27 +0,0 @@
-project(GrayScott_OpenCL_Image)
-
-set(CMAKE_MODULE_PATH ${GrayScott_OpenCL_SOURCE_DIR})
-# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
-
-find_package(OpenCV REQUIRED)
-include_directories( ${OPENCV_INCLUDE_DIR})
-link_libraries( ${OpenCV_LIBS} )
-
-# only build the OpenCL version if OpenCL was found
-find_package ( OpenCL )
-if(OPENCL_FOUND)
-  include_directories ( ${OPENCL_INCLUDE_DIRS} )
-  link_libraries ( ${OPENCL_LIBRARIES} )
-
-  # tell the code where the .cl file will live
-  add_definitions(-DCL_SOURCE_DIR="${GrayScott_OpenCL_Image_SOURCE_DIR}")
-  
-  include_directories(../Display)
-
-  add_executable(GrayScott_OpenCL_Image
-    gray_scott_opencl_image.cpp 
-    grayscott_kernel_image.cl 
-    ../Display/defs.h
-  )
-endif()
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image/FindOpenCL.cmake b/SpeedComparisons/GrayScott_OpenCL_Image/FindOpenCL.cmake
deleted file mode 100644
index fde90efae..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image/FindOpenCL.cmake
+++ /dev/null
@@ -1,79 +0,0 @@
-# - Try to find OpenCL
-# This module tries to find an OpenCL implementation on your system. It supports
-# AMD / ATI, Apple and NVIDIA implementations, but shoudl work, too.
-#
-# Once done this will define
-#  OPENCL_FOUND        - system has OpenCL
-#  OPENCL_INCLUDE_DIRS  - the OpenCL include directory
-#  OPENCL_LIBRARIES    - link these to use OpenCL
-#
-# WIN32 should work, but is untested
-
-FIND_PACKAGE( PackageHandleStandardArgs )
-
-SET (OPENCL_VERSION_STRING "0.1.0")
-SET (OPENCL_VERSION_MAJOR 0)
-SET (OPENCL_VERSION_MINOR 1)
-SET (OPENCL_VERSION_PATCH 0)
-
-IF (APPLE)
-
-  FIND_LIBRARY(OPENCL_LIBRARIES OpenCL DOC "OpenCL lib for OSX")
-  FIND_PATH(OPENCL_INCLUDE_DIRS OpenCL/cl.h DOC "Include for OpenCL on OSX")
-  FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS OpenCL/cl.hpp DOC "Include for OpenCL CPP bindings on OSX")
-
-ELSE (APPLE)
-
-	IF (WIN32)
-	
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h)
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp)
-	
-	    # The AMD SDK currently installs both x86 and x86_64 libraries
-	    # This is only a hack to find out architecture
-	    IF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86_64")
-	    ELSE (${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64")
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86")
-	    ENDIF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    FIND_LIBRARY(OPENCL_LIBRARIES OpenCL.lib ${OPENCL_LIB_DIR})
-	    
-	    GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-	    
-	    # On Win32 search relative to the library
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS "${_OPENCL_INC_CAND}")
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS "${_OPENCL_INC_CAND}")
-	
-	ELSE (WIN32)
-
-            # Unix style platforms
-            FIND_LIBRARY(OPENCL_LIBRARIES OpenCL
-              ENV LD_LIBRARY_PATH
-            )
-
-            GET_FILENAME_COMPONENT(OPENCL_LIB_DIR ${OPENCL_LIBRARIES} PATH)
-            GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-
-            # The AMD SDK currently does not place its headers
-            # in /usr/include, therefore also search relative
-            # to the library
-            FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-            FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-
-	ENDIF (WIN32)
-
-ENDIF (APPLE)
-
-FIND_PACKAGE_HANDLE_STANDARD_ARGS( OpenCL DEFAULT_MSG OPENCL_LIBRARIES OPENCL_INCLUDE_DIRS )
-
-IF( _OPENCL_CPP_INCLUDE_DIRS )
-	SET( OPENCL_HAS_CPP_BINDINGS TRUE )
-	LIST( APPEND OPENCL_INCLUDE_DIRS ${_OPENCL_CPP_INCLUDE_DIRS} )
-	# This is often the same, so clean up
-	LIST( REMOVE_DUPLICATES OPENCL_INCLUDE_DIRS )
-ENDIF( _OPENCL_CPP_INCLUDE_DIRS )
-
-MARK_AS_ADVANCED(
-  OPENCL_INCLUDE_DIRS
-)
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image/cl.hpp b/SpeedComparisons/GrayScott_OpenCL_Image/cl.hpp
deleted file mode 100644
index 99b86a665..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image/cl.hpp
+++ /dev/null
@@ -1,4011 +0,0 @@
-/*******************************************************************************
- * Copyright (c) 2008-2010 The Khronos Group Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and/or associated documentation files (the
- * "Materials"), to deal in the Materials without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sublicense, and/or sell copies of the Materials, and to
- * permit persons to whom the Materials are furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice shall be included
- * in all copies or substantial portions of the Materials.
- *
- * THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
- * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
- * CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
- * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
- * MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
- ******************************************************************************/
-
-/*! \file
- *
- *   \brief C++ bindings for OpenCL 1.0 (rev 48) and OpenCL 1.1 (rev 33)    
- *   \author Benedict R. Gaster and Laurent Morichetti
- *   
- *   Additions and fixes from Brian Cole, March 3rd 2010.
- *   
- *   \version 1.1
- *   \date June 2010
- *
- *   Optional extension support
- *
- *         cl
- *         cl_ext_device_fission
- *				#define USE_CL_DEVICE_FISSION
- */
-
-/*! \mainpage
- * \section intro Introduction
- * For many large applications C++ is the language of choice and so it seems
- * reasonable to define C++ bindings for OpenCL.
- *
- *
- * The interface is contained with a single C++ header file \em cl.hpp and all
- * definitions are contained within the namespace \em cl. There is no additional
- * requirement to include \em cl.h and to use either the C++ or original C
- * bindings it is enough to simply include \em cl.hpp.
- *
- * The bindings themselves are lightweight and correspond closely to the
- * underlying C API. Using the C++ bindings introduces no additional execution
- * overhead.
- *
- * For detail documentation on the bindings see:
- *
- * The OpenCL C++ Wrapper API 1.1 (revision 04)
- *  http://www.khronos.org/registry/cl/specs/opencl-cplusplus-1.1.pdf
- *
- * \section example Example
- *
- * The following example shows a general use case for the C++
- * bindings, including support for the optional exception feature and
- * also the supplied vector and string classes, see following sections for
- * decriptions of these features.
- *
- * \code
- * #define __CL_ENABLE_EXCEPTIONS
- * 
- * #if defined(__APPLE__) || defined(__MACOSX)
- * #include <OpenCL/cl.hpp>
- * #else
- * #include <CL/cl.hpp>
- * #endif
- * #include <cstdio>
- * #include <cstdlib>
- * #include <iostream>
- * 
- *  const char * helloStr  = "__kernel void "
- *                           "hello(void) "
- *                           "{ "
- *                           "  "
- *                           "} ";
- * 
- *  int
- *  main(void)
- *  {
- *     cl_int err = CL_SUCCESS;
- *     try {
- *
- *       std::vector<cl::Platform> platforms;
- *       cl::Platform::get(&platforms);
- *       if (platforms.size() == 0) {
- *           std::cout << "Platform size 0\n";
- *           return -1;
- *       }
- *
- *       cl_context_properties properties[] = 
- *          { CL_CONTEXT_PLATFORM, (cl_context_properties)(platforms[0])(), 0};
- *       cl::Context context(CL_DEVICE_TYPE_CPU, properties); 
- * 
- *       std::vector<cl::Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
- * 
- *       cl::Program::Sources source(1,
- *           std::make_pair(helloStr,strlen(helloStr)));
- *       cl::Program program_ = cl::Program(context, source);
- *       program_.build(devices);
- * 
- *       cl::Kernel kernel(program_, "hello", &err);
- * 
- *       cl::Event event;
- *       cl::CommandQueue queue(context, devices[0], 0, &err);
- *       queue.enqueueNDRangeKernel(
- *           kernel, 
- *           cl::NullRange, 
- *           cl::NDRange(4,4),
- *           cl::NullRange,
- *           NULL,
- *           &event); 
- * 
- *       event.wait();
- *     }
- *     catch (cl::Error err) {
- *        std::cerr 
- *           << "ERROR: "
- *           << err.what()
- *           << "("
- *           << err.err()
- *           << ")"
- *           << std::endl;
- *     }
- * 
- *    return EXIT_SUCCESS;
- *  }
- * 
- * \endcode
- *
- */
-#ifndef CL_HPP_
-#define CL_HPP_
-
-#ifdef _WIN32
-#include <windows.h>
-#include <malloc.h>
-#if defined(USE_DX_INTEROP)
-#include <CL/cl_d3d10.h>
-#endif
-#endif // _WIN32
-
-// 
-#if defined(USE_CL_DEVICE_FISSION)
-#include <CL/cl_ext.h>
-#endif
-
-#if defined(__APPLE__) || defined(__MACOSX)
-#include <OpenGL/OpenGL.h>
-#include <OpenCL/opencl.h>
-#else
-#include <GL/gl.h>
-#include <CL/opencl.h>
-#endif // !__APPLE__
-
-#if !defined(CL_CALLBACK)
-#define CL_CALLBACK
-#endif //CL_CALLBACK
-
-#include <utility>
-
-#if !defined(__NO_STD_VECTOR)
-#include <vector>
-#endif
-
-#if !defined(__NO_STD_STRING)
-#include <string>
-#endif 
-
-#if defined(linux) || defined(__APPLE__) || defined(__MACOSX)
-# include <alloca.h>
-#endif // linux
-
-#include <cstring>
-
-/*! \namespace cl
- *
- * \brief The OpenCL C++ bindings are defined within this namespace.
- *
- */
-namespace cl {
-
-#define __INIT_CL_EXT_FCN_PTR(name) \
-    if(!pfn_##name) { \
-        pfn_##name = (PFN_##name) \
-            clGetExtensionFunctionAddress(#name); \
-        if(!pfn_##name) { \
-        } \
-    }
-
-class Program;
-class Device;
-class Context;
-class CommandQueue;
-class Memory;
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-#include <exception>
-/*! \class Error
- * \brief Exception class
- */
-class Error : public std::exception
-{
-private:
-    cl_int err_;
-    const char * errStr_;
-public:
-    /*! Create a new CL error exception for a given error code
-     *  and corresponding message.
-     */
-    Error(cl_int err, const char * errStr = NULL) : err_(err), errStr_(errStr)
-    {}
-
-    ~Error() throw() {}
-
-    /*! \brief Get error string associated with exception
-     *
-     * \return A memory pointer to the error message string.
-     */
-    virtual const char * what() const throw ()
-    {
-        if (errStr_ == NULL) {
-            return "empty";
-        }
-        else {
-            return errStr_;
-        }
-    }
-
-    /*! \brief Get error code associated with exception
-     *
-     *  \return The error code.
-     */
-    const cl_int err(void) const { return err_; }
-};
-
-#define __ERR_STR(x) #x
-#else
-#define __ERR_STR(x) NULL
-#endif // __CL_ENABLE_EXCEPTIONS
-
-//! \cond DOXYGEN_DETAIL
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#define __GET_DEVICE_INFO_ERR               __ERR_STR(clgetDeviceInfo)
-#define __GET_PLATFORM_INFO_ERR             __ERR_STR(clGetPlatformInfo)
-#define __GET_DEVICE_IDS_ERR                __ERR_STR(clGetDeviceIDs)
-#define __GET_PLATFORM_IDS_ERR              __ERR_STR(clGetPlatformIDs)
-#define __GET_CONTEXT_INFO_ERR              __ERR_STR(clGetContextInfo)
-#define __GET_EVENT_INFO_ERR                __ERR_STR(clGetEventInfo)
-#define __GET_EVENT_PROFILE_INFO_ERR        __ERR_STR(clGetEventProfileInfo)
-#define __GET_MEM_OBJECT_INFO_ERR           __ERR_STR(clGetMemObjectInfo)
-#define __GET_IMAGE_INFO_ERR                __ERR_STR(clGetImageInfo)
-#define __GET_SAMPLER_INFO_ERR              __ERR_STR(clGetSamplerInfo)
-#define __GET_KERNEL_INFO_ERR               __ERR_STR(clGetKernelInfo)
-#define __GET_KERNEL_WORK_GROUP_INFO_ERR    __ERR_STR(clGetKernelWorkGroupInfo)
-#define __GET_PROGRAM_INFO_ERR              __ERR_STR(clGetProgramInfo)
-#define __GET_PROGRAM_BUILD_INFO_ERR        __ERR_STR(clGetProgramBuildInfo)
-#define __GET_COMMAND_QUEUE_INFO_ERR        __ERR_STR(clGetCommandQueueInfo)
-
-#define __CREATE_CONTEXT_FROM_TYPE_ERR      __ERR_STR(clCreateContextFromType)
-#define __GET_SUPPORTED_IMAGE_FORMATS_ERR   __ERR_STR(clGetSupportedImageFormats)
-
-#define __CREATE_BUFFER_ERR                 __ERR_STR(clCreateBuffer)
-#define __CREATE_SUBBUFFER_ERR              __ERR_STR(clCreateSubBuffer)
-#define __CREATE_GL_BUFFER_ERR              __ERR_STR(clCreateFromGLBuffer)
-#define __GET_GL_OBJECT_INFO_ERR            __ERR_STR(clGetGLObjectInfo)
-#define __CREATE_IMAGE2D_ERR                __ERR_STR(clCreateImage2D)
-#define __CREATE_IMAGE3D_ERR                __ERR_STR(clCreateImage3D)
-#define __CREATE_SAMPLER_ERR                __ERR_STR(clCreateSampler)
-#define __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR __ERR_STR(clSetMemObjectDestructorCallback)
-
-#define __CREATE_USER_EVENT_ERR             __ERR_STR(clCreateUserEvent)
-#define __SET_USER_EVENT_STATUS_ERR         __ERR_STR(clSetUserEventStatus)
-#define __SET_EVENT_CALLBACK_ERR            __ERR_STR(clSetEventCallback)
-#define __WAIT_FOR_EVENTS_ERR               __ERR_STR(clWaitForEvents)
-
-#define __CREATE_KERNEL_ERR                 __ERR_STR(clCreateKernel)
-#define __SET_KERNEL_ARGS_ERR               __ERR_STR(clSetKernelArg)
-#define __CREATE_PROGRAM_WITH_SOURCE_ERR    __ERR_STR(clCreateProgramWithSource)
-#define __CREATE_PROGRAM_WITH_BINARY_ERR    __ERR_STR(clCreateProgramWithBinary)
-#define __BUILD_PROGRAM_ERR                 __ERR_STR(clBuildProgram)
-#define __CREATE_KERNELS_IN_PROGRAM_ERR     __ERR_STR(clCreateKernelsInProgram)
-
-#define __CREATE_COMMAND_QUEUE_ERR          __ERR_STR(clCreateCommandQueue)
-#define __SET_COMMAND_QUEUE_PROPERTY_ERR    __ERR_STR(clSetCommandQueueProperty)
-#define __ENQUEUE_READ_BUFFER_ERR           __ERR_STR(clEnqueueReadBuffer)
-#define __ENQUEUE_READ_BUFFER_RECT_ERR      __ERR_STR(clEnqueueReadBufferRect)
-#define __ENQUEUE_WRITE_BUFFER_ERR          __ERR_STR(clEnqueueWriteBuffer)
-#define __ENQUEUE_WRITE_BUFFER_RECT_ERR     __ERR_STR(clEnqueueWriteBufferRect)
-#define __ENQEUE_COPY_BUFFER_ERR            __ERR_STR(clEnqueueCopyBuffer)
-#define __ENQEUE_COPY_BUFFER_RECT_ERR       __ERR_STR(clEnqueueCopyBufferRect)
-#define __ENQUEUE_READ_IMAGE_ERR            __ERR_STR(clEnqueueReadImage)
-#define __ENQUEUE_WRITE_IMAGE_ERR           __ERR_STR(clEnqueueWriteImage)
-#define __ENQUEUE_COPY_IMAGE_ERR            __ERR_STR(clEnqueueCopyImage)
-#define __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR  __ERR_STR(clEnqueueCopyImageToBuffer)
-#define __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR  __ERR_STR(clEnqueueCopyBufferToImage)
-#define __ENQUEUE_MAP_BUFFER_ERR            __ERR_STR(clEnqueueMapBuffer)
-#define __ENQUEUE_MAP_IMAGE_ERR             __ERR_STR(clEnqueueMapImage)
-#define __ENQUEUE_UNMAP_MEM_OBJECT_ERR      __ERR_STR(clEnqueueUnMapMemObject)
-#define __ENQUEUE_NDRANGE_KERNEL_ERR        __ERR_STR(clEnqueueNDRangeKernel)
-#define __ENQUEUE_TASK_ERR                  __ERR_STR(clEnqueueTask)
-#define __ENQUEUE_NATIVE_KERNEL             __ERR_STR(clEnqueueNativeKernel)
-#define __ENQUEUE_MARKER_ERR                __ERR_STR(clEnqueueMarker)
-#define __ENQUEUE_WAIT_FOR_EVENTS_ERR       __ERR_STR(clEnqueueWaitForEvents)
-#define __ENQUEUE_BARRIER_ERR               __ERR_STR(clEnqueueBarrier)
-
-#define __ENQUEUE_ACQUIRE_GL_ERR            __ERR_STR(clEnqueueAcquireGLObjects)
-#define __ENQUEUE_RELEASE_GL_ERR            __ERR_STR(clEnqueueReleaseGLObjects)
-
-#define __UNLOAD_COMPILER_ERR               __ERR_STR(clUnloadCompiler)
-
-#define __FLUSH_ERR                         __ERR_STR(clFlush)
-#define __FINISH_ERR                        __ERR_STR(clFinish)
-
-#define __CREATE_SUB_DEVICES                __ERR_STR(clCreateSubDevicesEXT)
-#endif // __CL_USER_OVERRIDE_ERROR_STRINGS
-//! \endcond
-
-/*! \class string
- * \brief Simple string class, that provides a limited subset of std::string
- * functionality but avoids many of the issues that come with that class.
- */
-class string
-{
-private:
-    ::size_t size_;
-    char * str_;
-public:
-    string(void) : size_(0), str_(NULL)
-    {
-    }
-
-    string(char * str, ::size_t size) :
-        size_(size),
-        str_(NULL)
-    {
-        str_ = new char[size_+1];
-        if (str_ != NULL) {
-            memcpy(str_, str, size_  * sizeof(char));
-            str_[size_] = '\0';
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string(char * str) :
-        str_(NULL)
-    {
-        size_= ::strlen(str);
-        str_ = new char[size_ + 1];
-        if (str_ != NULL) {
-            memcpy(str_, str, (size_ + 1) * sizeof(char));
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string& operator=(const string& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        if (rhs.size_ == 0 || rhs.str_ == NULL) {
-            size_ = 0;
-            str_  = NULL;
-        } 
-        else {
-            size_ = rhs.size_;
-            str_ = new char[size_ + 1];
-            if (str_ != NULL) {
-                memcpy(str_, rhs.str_, (size_ + 1) * sizeof(char));
-            }
-            else {
-                size_ = 0;
-            }
-        }
-
-        return *this;
-    }
-
-    string(const string& rhs)
-    {
-        *this = rhs;
-    }
-
-    ~string()
-    {
-        if (str_ != NULL) {
-            delete[] str_;
-        }
-    }
-
-    ::size_t size(void) const   { return size_; }
-    ::size_t length(void) const { return size(); }
-
-    const char * c_str(void) const { return (str_) ? str_ : "";}
-};
-
-#if !defined(__USE_DEV_STRING) && !defined(__NO_STD_STRING)
-#include <string>
-typedef std::string STRING_CLASS;
-#elif !defined(__USE_DEV_STRING) 
-typedef cl::string STRING_CLASS;
-#endif
-
-#if !defined(__USE_DEV_VECTOR) && !defined(__NO_STD_VECTOR)
-#include <vector>
-#define VECTOR_CLASS std::vector
-#elif !defined(__USE_DEV_VECTOR) 
-#define VECTOR_CLASS cl::vector 
-#endif
-
-#if !defined(__MAX_DEFAULT_VECTOR_SIZE)
-#define __MAX_DEFAULT_VECTOR_SIZE 10
-#endif
-
-/*! \class vector
- * \brief Fixed sized vector implementation that mirroring 
- * std::vector functionality.
- */
-template <typename T, unsigned int N = __MAX_DEFAULT_VECTOR_SIZE>
-class vector
-{
-private:
-    T data_[N];
-    unsigned int size_;
-    bool empty_;
-public:
-    vector() : 
-        size_(-1),
-        empty_(true)
-    {}
-
-    ~vector() {}
-
-    unsigned int size(void) const
-    {
-        return size_ + 1;
-    }
-
-    void clear()
-    {
-        size_ = -1;
-        empty_ = true;
-    }
-
-    void push_back (const T& x)
-    { 
-        if (size() < N) {
-            size_++;  
-            data_[size_] = x;
-            empty_ = false;
-        }
-    }
-
-    void pop_back(void)
-    {
-        if (!empty_) {
-            data_[size_].~T();
-            size_--;
-            if (size_ == -1) {
-                empty_ = true;
-            }
-        }
-    }
-  
-    vector(const vector<T, N>& vec) : 
-        size_(vec.size_),
-        empty_(vec.empty_)
-    {
-        if (!empty_) {
-            memcpy(&data_[0], &vec.data_[0], size() * sizeof(T));
-        }
-    } 
-
-    vector(unsigned int size, const T& val = T()) :
-        size_(-1),
-        empty_(true)
-    {
-        for (unsigned int i = 0; i < size; i++) {
-            push_back(val);
-        }
-    }
-
-    vector<T, N>& operator=(const vector<T, N>& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        size_  = rhs.size_;
-        empty_ = rhs.empty_;
-
-        if (!empty_) {	
-            memcpy(&data_[0], &rhs.data_[0], size() * sizeof(T));
-        }
-    
-        return *this;
-    }
-
-    bool operator==(vector<T,N> &vec)
-    {
-        if (empty_ && vec.empty_) {
-            return true;
-        }
-
-        if (size() != vec.size()) {
-            return false;
-        }
-
-        return memcmp(&data_[0], &vec.data_[0], size() * sizeof(T)) == 0 ? true : false;
-    }
-  
-    operator T* ()             { return data_; }
-    operator const T* () const { return data_; }
-   
-    bool empty (void) const
-    {
-        return empty_;
-    }
-  
-    unsigned int max_size (void) const
-    {
-        return N;
-    }
-
-    unsigned int capacity () const
-    {
-        return sizeof(T) * N;
-    }
-
-    T& operator[](int index)
-    {
-        return data_[index];
-    }
-  
-    T operator[](int index) const
-    {
-        return data_[index];
-    }
-  
-    template<class I>
-    void assign(I start, I end)
-    {
-        clear();   
-        while(start < end) {
-            push_back(*start);
-            start++;
-        }
-    }
-
-    /*! \class iterator
-     * \brief Iterator class for vectors
-     */
-    class iterator
-    {
-    private:
-        vector<T,N> vec_;
-        int index_;
-        bool initialized_;
-    public:
-        iterator(void) : 
-            index_(-1),
-            initialized_(false)
-        {
-            index_ = -1;
-            initialized_ = false;
-        }
-
-        ~iterator(void) {}
-
-        static iterator begin(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = 0;
-            }
-
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-
-        static iterator end(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = vec.size();
-            }
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-    
-        bool operator==(iterator i)
-        {
-            return ((vec_ == i.vec_) && 
-                    (index_ == i.index_) && 
-                    (initialized_ == i.initialized_));
-        }
-
-        bool operator!=(iterator i)
-        {
-            return (!(*this==i));
-        }
-
-        void operator++()
-        {
-            index_++;
-        }
-
-        void operator++(int x)
-        {
-            index_ += x;
-        }
-
-        void operator--()
-        {
-            index_--;
-        }
-
-        void operator--(int x)
-        {
-            index_ -= x;
-        }
-
-        T operator *()
-        {
-            return vec_[index_];
-        }
-    };
-
-    iterator begin(void)
-    {
-        return iterator::begin(*this);
-    }
-
-    iterator end(void)
-    {
-        return iterator::end(*this);
-    }
-
-    T& front(void)
-    {
-        return data_[0];
-    }
-
-    T& back(void)
-    {
-        return data_[size_];
-    }
-
-    const T& front(void) const
-    {
-        return data_[0];
-    }
-
-    const T& back(void) const
-    {
-        return data_[size_];
-    }
-};  
-    
-/*!
- * \brief size_t class used to interface between C++ and
- * OpenCL C calls that require arrays of size_t values, who's
- * size is known statically.
- */
-template <int N>
-struct size_t : public cl::vector< ::size_t, N> { };
-
-namespace detail {
-
-// GetInfo help struct
-template <typename Functor, typename T>
-struct GetInfoHelper
-{
-    static cl_int
-    get(Functor f, cl_uint name, T* param)
-    {
-        return f(name, sizeof(T), param, NULL);
-    }
-};
-
-// Specialized GetInfoHelper for VECTOR_CLASS params
-template <typename Func, typename T>
-struct GetInfoHelper<Func, VECTOR_CLASS<T> >
-{
-    static cl_int get(Func f, cl_uint name, VECTOR_CLASS<T>* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        T* value = (T*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        param->assign(&value[0], &value[required/sizeof(T)]);
-        return CL_SUCCESS;
-    }
-};
-
-// Specialized for getInfo<CL_PROGRAM_BINARIES>
-template <typename Func>
-struct GetInfoHelper<Func, VECTOR_CLASS<char *> >
-{
-    static cl_int
-    get(Func f, cl_uint name, VECTOR_CLASS<char *>* param)
-    {
-      cl_uint err = f(name, param->size() * sizeof(char *), &(*param)[0], NULL);
-      if (err != CL_SUCCESS) {
-        return err;
-      }
-      
-      return CL_SUCCESS;
-    }
-};
-
-// Specialized GetInfoHelper for STRING_CLASS params
-template <typename Func>
-struct GetInfoHelper<Func, STRING_CLASS>
-{
-    static cl_int get(Func f, cl_uint name, STRING_CLASS* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        char* value = (char*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        *param = value;
-        return CL_SUCCESS;
-    }
-};
-
-#define __GET_INFO_HELPER_WITH_RETAIN(CPP_TYPE) \
-namespace detail { \
-template <typename Func> \
-struct GetInfoHelper<Func, CPP_TYPE> \
-{ \
-    static cl_int get(Func f, cl_uint name, CPP_TYPE* param) \
-    { \
-      cl_uint err = f(name, sizeof(CPP_TYPE), param, NULL); \
-      if (err != CL_SUCCESS) { \
-        return err; \
-      } \
-      \
-      return ReferenceHandler<CPP_TYPE::cl_type>::retain((*param)()); \
-    } \
-}; \
-} 
-
-
-#define __PARAM_NAME_INFO_1_0(F) \
-    F(cl_platform_info, CL_PLATFORM_PROFILE, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VERSION, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_NAME, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VENDOR, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_device_info, CL_DEVICE_TYPE, cl_device_type) \
-    F(cl_device_info, CL_DEVICE_VENDOR_ID, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_COMPUTE_UNITS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_CLOCK_FREQUENCY, cl_uint) \
-    F(cl_device_info, CL_DEVICE_ADDRESS_BITS, cl_bitfield) \
-    F(cl_device_info, CL_DEVICE_MAX_READ_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WRITE_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_MEM_ALLOC_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_DEPTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE_SUPPORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_PARAMETER_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_SAMPLERS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MEM_BASE_ADDR_ALIGN, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_SINGLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_TYPE, cl_device_mem_cache_type) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE, cl_uint)\
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_TYPE, cl_device_local_mem_type) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_ERROR_CORRECTION_SUPPORT, cl_bool) \
-    F(cl_device_info, CL_DEVICE_PROFILING_TIMER_RESOLUTION, ::size_t) \
-    F(cl_device_info, CL_DEVICE_ENDIAN_LITTLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_COMPILER_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_EXECUTION_CAPABILITIES, cl_device_exec_capabilities) \
-    F(cl_device_info, CL_DEVICE_QUEUE_PROPERTIES, cl_command_queue_properties) \
-    F(cl_device_info, CL_DEVICE_PLATFORM, cl_platform_id) \
-    F(cl_device_info, CL_DEVICE_NAME, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VENDOR, STRING_CLASS) \
-    F(cl_device_info, CL_DRIVER_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_PROFILE, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_context_info, CL_CONTEXT_REFERENCE_COUNT, cl_uint) \
-    F(cl_context_info, CL_CONTEXT_DEVICES, VECTOR_CLASS<Device>) \
-    F(cl_context_info, CL_CONTEXT_PROPERTIES, VECTOR_CLASS<cl_context_properties>) \
-    \
-    F(cl_event_info, CL_EVENT_COMMAND_QUEUE, cl::CommandQueue) \
-    F(cl_event_info, CL_EVENT_COMMAND_TYPE, cl_command_type) \
-    F(cl_event_info, CL_EVENT_REFERENCE_COUNT, cl_uint) \
-    F(cl_event_info, CL_EVENT_COMMAND_EXECUTION_STATUS, cl_uint) \
-    \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_QUEUED, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_SUBMIT, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_START, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_END, cl_ulong) \
-    \
-    F(cl_mem_info, CL_MEM_TYPE, cl_mem_object_type) \
-    F(cl_mem_info, CL_MEM_FLAGS, cl_mem_flags) \
-    F(cl_mem_info, CL_MEM_SIZE, ::size_t) \
-    F(cl_mem_info, CL_MEM_HOST_PTR, void*) \
-    F(cl_mem_info, CL_MEM_MAP_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_REFERENCE_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_CONTEXT, cl::Context) \
-    \
-    F(cl_image_info, CL_IMAGE_FORMAT, cl_image_format) \
-    F(cl_image_info, CL_IMAGE_ELEMENT_SIZE, ::size_t) \
-    F(cl_image_info, CL_IMAGE_ROW_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_SLICE_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_WIDTH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_HEIGHT, ::size_t) \
-    F(cl_image_info, CL_IMAGE_DEPTH, ::size_t) \
-    \
-    F(cl_sampler_info, CL_SAMPLER_REFERENCE_COUNT, cl_uint) \
-    F(cl_sampler_info, CL_SAMPLER_CONTEXT, cl::Context) \
-    F(cl_sampler_info, CL_SAMPLER_NORMALIZED_COORDS, cl_addressing_mode) \
-    F(cl_sampler_info, CL_SAMPLER_ADDRESSING_MODE, cl_filter_mode) \
-    F(cl_sampler_info, CL_SAMPLER_FILTER_MODE, cl_bool) \
-    \
-    F(cl_program_info, CL_PROGRAM_REFERENCE_COUNT, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_CONTEXT, cl::Context) \
-    F(cl_program_info, CL_PROGRAM_NUM_DEVICES, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_DEVICES, VECTOR_CLASS<cl_device_id>) \
-    F(cl_program_info, CL_PROGRAM_SOURCE, STRING_CLASS) \
-    F(cl_program_info, CL_PROGRAM_BINARY_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_program_info, CL_PROGRAM_BINARIES, VECTOR_CLASS<char *>) \
-    \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_STATUS, cl_build_status) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_OPTIONS, STRING_CLASS) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_LOG, STRING_CLASS) \
-    \
-    F(cl_kernel_info, CL_KERNEL_FUNCTION_NAME, STRING_CLASS) \
-    F(cl_kernel_info, CL_KERNEL_NUM_ARGS, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_REFERENCE_COUNT, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_CONTEXT, cl::Context) \
-    F(cl_kernel_info, CL_KERNEL_PROGRAM, cl::Program) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_COMPILE_WORK_GROUP_SIZE, cl::size_t<3>) \
-    F(cl_kernel_work_group_info, CL_KERNEL_LOCAL_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_command_queue_info, CL_QUEUE_CONTEXT, cl::Context) \
-    F(cl_command_queue_info, CL_QUEUE_DEVICE, cl::Device) \
-    F(cl_command_queue_info, CL_QUEUE_REFERENCE_COUNT, cl_uint) \
-    F(cl_command_queue_info, CL_QUEUE_PROPERTIES, cl_command_queue_properties)
-
-#if defined(CL_VERSION_1_1)
-#define __PARAM_NAME_INFO_1_1(F) \
-    F(cl_context_info, CL_CONTEXT_NUM_DEVICES, cl_uint)\
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_DOUBLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HALF_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HOST_UNIFIED_MEMORY, cl_bool) \
-    \
-    F(cl_mem_info, CL_MEM_ASSOCIATED_MEMOBJECT, cl::Memory) \
-    F(cl_mem_info, CL_MEM_OFFSET, ::size_t) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_PRIVATE_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_event_info, CL_EVENT_CONTEXT, cl::Context)
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-#define __PARAM_NAME_DEVICE_FISSION(F) \
-    F(cl_device_info, CL_DEVICE_PARENT_DEVICE_EXT, cl_device_id) \
-	F(cl_device_info, CL_DEVICE_PARTITION_TYPES_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_AFFINITY_DOMAINS_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_REFERENCE_COUNT_EXT , cl_uint) \
-	F(cl_device_info, CL_DEVICE_PARTITION_STYLE_EXT, VECTOR_CLASS<cl_device_partition_property_ext>)
-#endif // USE_CL_DEVICE_FISSION
-
-template <typename enum_type, cl_int Name>
-struct param_traits {};
-
-#define __DECLARE_PARAM_TRAITS(token, param_name, T) \
-struct token;                                        \
-template<>                                           \
-struct param_traits<detail:: token,param_name>       \
-{                                                    \
-    enum { value = param_name };                     \
-    typedef T param_type;                            \
-};
-
-__PARAM_NAME_INFO_1_0(__DECLARE_PARAM_TRAITS);
-#if defined(CL_VERSION_1_1)
-__PARAM_NAME_INFO_1_1(__DECLARE_PARAM_TRAITS);
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-__PARAM_NAME_DEVICE_FISSION(__DECLARE_PARAM_TRAITS);
-#endif // USE_CL_DEVICE_FISSION
-
-#undef __DECLARE_PARAM_TRAITS
-
-// Convenience functions
-
-template <typename Func, typename T>
-inline cl_int
-getInfo(Func f, cl_uint name, T* param)
-{
-    return GetInfoHelper<Func, T>::get(f, name, param);
-}
-
-template <typename Func, typename Arg0>
-struct GetInfoFunctor0
-{
-    Func f_; const Arg0& arg0_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename Arg1>
-struct GetInfoFunctor1
-{
-    Func f_; const Arg0& arg0_; const Arg1& arg1_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, arg1_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, cl_uint name, T* param)
-{
-    GetInfoFunctor0<Func, Arg0> f0 = { f, arg0 };
-    return GetInfoHelper<GetInfoFunctor0<Func, Arg0>, T>
-        ::get(f0, name, param);
-}
-
-template <typename Func, typename Arg0, typename Arg1, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, const Arg1& arg1, cl_uint name, T* param)
-{
-    GetInfoFunctor1<Func, Arg0, Arg1> f0 = { f, arg0, arg1 };
-    return GetInfoHelper<GetInfoFunctor1<Func, Arg0, Arg1>, T>
-        ::get(f0, name, param);
-}
-
-template<typename T>
-struct ReferenceHandler
-{ };
-
-template <>
-struct ReferenceHandler<cl_device_id>
-{
-    // cl_device_id does not have retain().
-    static cl_int retain(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-    // cl_device_id does not have release().
-    static cl_int release(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-};
-
-template <>
-struct ReferenceHandler<cl_platform_id>
-{
-    // cl_platform_id does not have retain().
-    static cl_int retain(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-    // cl_platform_id does not have release().
-    static cl_int release(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-};
-
-template <>
-struct ReferenceHandler<cl_context>
-{
-    static cl_int retain(cl_context context)
-    { return ::clRetainContext(context); }
-    static cl_int release(cl_context context)
-    { return ::clReleaseContext(context); }
-};
-
-template <>
-struct ReferenceHandler<cl_command_queue>
-{
-    static cl_int retain(cl_command_queue queue)
-    { return ::clRetainCommandQueue(queue); }
-    static cl_int release(cl_command_queue queue)
-    { return ::clReleaseCommandQueue(queue); }
-};
-
-template <>
-struct ReferenceHandler<cl_mem>
-{
-    static cl_int retain(cl_mem memory)
-    { return ::clRetainMemObject(memory); }
-    static cl_int release(cl_mem memory)
-    { return ::clReleaseMemObject(memory); }
-};
-
-template <>
-struct ReferenceHandler<cl_sampler>
-{
-    static cl_int retain(cl_sampler sampler)
-    { return ::clRetainSampler(sampler); }
-    static cl_int release(cl_sampler sampler)
-    { return ::clReleaseSampler(sampler); }
-};
-
-template <>
-struct ReferenceHandler<cl_program>
-{
-    static cl_int retain(cl_program program)
-    { return ::clRetainProgram(program); }
-    static cl_int release(cl_program program)
-    { return ::clReleaseProgram(program); }
-};
-
-template <>
-struct ReferenceHandler<cl_kernel>
-{
-    static cl_int retain(cl_kernel kernel)
-    { return ::clRetainKernel(kernel); }
-    static cl_int release(cl_kernel kernel)
-    { return ::clReleaseKernel(kernel); }
-};
-
-template <>
-struct ReferenceHandler<cl_event>
-{
-    static cl_int retain(cl_event event)
-    { return ::clRetainEvent(event); }
-    static cl_int release(cl_event event)
-    { return ::clReleaseEvent(event); }
-};
-
-template <typename T>
-class Wrapper
-{
-public:
-    typedef T cl_type;
-
-protected:
-    cl_type object_;
-
-public:
-    Wrapper() : object_(NULL) { }
-
-    ~Wrapper()
-    {
-        if (object_ != NULL) { release(); }
-    }
-
-    Wrapper(const Wrapper<cl_type>& rhs)
-    {
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-    }
-
-    Wrapper<cl_type>& operator = (const Wrapper<cl_type>& rhs)
-    {
-        if (object_ != NULL) { release(); }
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-        return *this;
-    }
-
-    cl_type operator ()() const { return object_; }
-
-    cl_type& operator ()() { return object_; }
-
-protected:
-
-    cl_int retain() const
-    {
-        return ReferenceHandler<cl_type>::retain(object_);
-    }
-
-    cl_int release() const
-    {
-        return ReferenceHandler<cl_type>::release(object_);
-    }
-};
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-static inline cl_int errHandler (
-    cl_int err,
-    const char * errStr = NULL) throw(Error)
-{
-    if (err != CL_SUCCESS) {
-        throw Error(err, errStr);
-    }
-    return err;
-}
-#else
-static inline cl_int errHandler (cl_int err, const char * errStr = NULL)
-{
-    return err;
-}
-#endif // __CL_ENABLE_EXCEPTIONS
-
-} // namespace detail
-//! \endcond
-
-/*! \stuct ImageFormat
- * \brief ImageFormat interface fro cl_image_format.
- */
-struct ImageFormat : public cl_image_format
-{
-    ImageFormat(){}
-
-    ImageFormat(cl_channel_order order, cl_channel_type type)
-    {
-        image_channel_order = order;
-        image_channel_data_type = type;
-    }
-
-    ImageFormat& operator = (const ImageFormat& rhs)
-    {
-        if (this != &rhs) {
-            this->image_channel_data_type = rhs.image_channel_data_type;
-            this->image_channel_order     = rhs.image_channel_order;
-        }
-        return *this;
-    }
-};
-
-/*! \class Device
- * \brief Device interface for cl_device_id.
- */
-class Device : public detail::Wrapper<cl_device_id>
-{
-public:
-    Device(cl_device_id device) { object_ = device; }
-
-    Device() : detail::Wrapper<cl_type>() { }
-
-    Device(const Device& device) : detail::Wrapper<cl_type>(device) { }
-
-    Device& operator = (const Device& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_device_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetDeviceInfo, object_, name, param),
-            __GET_DEVICE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_device_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_device_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(USE_CL_DEVICE_FISSION)
-	cl_int createSubDevices(
-		const cl_device_partition_property_ext * properties,
-		VECTOR_CLASS<Device>* devices)
-	{
-		typedef CL_API_ENTRY cl_int 
-			( CL_API_CALL * PFN_clCreateSubDevicesEXT)(
-				cl_device_id /*in_device*/,
-                const cl_device_partition_property_ext * /* properties */,
-                cl_uint /*num_entries*/,
-                cl_device_id * /*out_devices*/,
-                cl_uint * /*num_devices*/ ) CL_EXT_SUFFIX__VERSION_1_1;
-
-		static PFN_clCreateSubDevicesEXT pfn_clCreateSubDevicesEXT = NULL;
-		__INIT_CL_EXT_FCN_PTR(clCreateSubDevicesEXT);
-
-		cl_uint n = 0;
-        cl_int err = pfn_clCreateSubDevicesEXT(object_, properties, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clCreateSubDevicesEXT(object_, properties, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
- 	}
-#endif
-};
-
-/*! \class Platform
- *  \brief Platform interface.
- */
-class Platform : public detail::Wrapper<cl_platform_id>
-{
-public:
-    static const Platform null();
-
-    Platform(cl_platform_id platform) { object_ = platform; }
-
-    Platform() : detail::Wrapper<cl_type>()  { }
-
-    Platform(const Platform& platform) : detail::Wrapper<cl_type>(platform) { }
-
-    Platform& operator = (const Platform& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getInfo(cl_platform_info name, STRING_CLASS* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetPlatformInfo, object_, name, param),
-            __GET_PLATFORM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_platform_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_platform_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getDevices(
-        cl_device_type type,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetDeviceIDs(object_, type, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = ::clGetDeviceIDs(object_, type, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-
-#if defined(USE_DX_INTEROP)
-   /*! \brief Get the list of available D3D10 devices.
-     *
-     *  \param d3d_device_source.
-     *
-     *  \param d3d_object.
-     *
-     *  \param d3d_device_set.
-     *
-     *  \param devices returns a vector of OpenCL D3D10 devices found. The cl::Device
-     *  values returned in devices can be used to identify a specific OpenCL
-     *  device. If \a devices argument is NULL, this argument is ignored.
-     *
-     *  \return One of the following values:
-     *    - CL_SUCCESS if the function is executed successfully.
-     *
-     *  The application can query specific capabilities of the OpenCL device(s)
-     *  returned by cl::getDevices. This can be used by the application to
-     *  determine which device(s) to use.
-     *
-     * \note In the case that exceptions are enabled and a return value
-     * other than CL_SUCCESS is generated, then cl::Error exception is
-     * generated.
-     */
-    cl_int getDevices(
-        cl_d3d10_device_source_khr d3d_device_source,
-        void *                     d3d_object,
-        cl_d3d10_device_set_khr    d3d_device_set,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clGetDeviceIDsFromD3D10KHR)(
-            cl_platform_id platform, 
-            cl_d3d10_device_source_khr d3d_device_source, 
-            void * d3d_object,
-            cl_d3d10_device_set_khr d3d_device_set,
-            cl_uint num_entries,
-            cl_device_id * devices,
-            cl_uint* num_devices);
-
-        static PFN_clGetDeviceIDsFromD3D10KHR pfn_clGetDeviceIDsFromD3D10KHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clGetDeviceIDsFromD3D10KHR);
-
-        cl_uint n = 0;
-        cl_int err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set, 
-            0, 
-            NULL, 
-            &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set,
-            n, 
-            ids, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-#endif
-
-    static cl_int get(
-        VECTOR_CLASS<Platform>* platforms)
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetPlatformIDs(0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        cl_platform_id* ids = (cl_platform_id*) alloca(
-            n * sizeof(cl_platform_id));
-        err = ::clGetPlatformIDs(n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        platforms->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-};
-
-static inline cl_int
-UnloadCompiler()
-{
-    return ::clUnloadCompiler();
-}
-
-class Context : public detail::Wrapper<cl_context>
-{
-public:
-    Context(
-        const VECTOR_CLASS<Device>& devices,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContext(
-            properties, (cl_uint) devices.size(),
-            (cl_device_id*) &devices.front(),
-            notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context(
-        cl_device_type type,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContextFromType(
-            properties, type, notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context() : detail::Wrapper<cl_type>() { }
-
-    Context(const Context& context) : detail::Wrapper<cl_type>(context) { }
-
-    Context& operator = (const Context& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_context_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetContextInfo, object_, name, param),
-            __GET_CONTEXT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_context_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_context_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getSupportedImageFormats(
-        cl_mem_flags flags,
-        cl_mem_object_type type,
-        VECTOR_CLASS<ImageFormat>* formats) const
-    {
-        cl_uint numEntries;
-        cl_int err = ::clGetSupportedImageFormats(
-           object_, 
-           flags,
-           type, 
-           0, 
-           NULL, 
-           &numEntries);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        ImageFormat* value = (ImageFormat*)
-            alloca(numEntries * sizeof(ImageFormat));
-        err = ::clGetSupportedImageFormats(
-            object_, 
-            flags, 
-            type, 
-            numEntries,
-            (cl_image_format*) value, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        formats->assign(&value[0], &value[numEntries]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Context)
-
-/*! \class Event
- * \brief Event interface for cl_event.
- */
-class Event : public detail::Wrapper<cl_event>
-{
-public:
-    Event() : detail::Wrapper<cl_type>() { }
-
-    Event(const Event& event) : detail::Wrapper<cl_type>(event) { }
-
-    Event& operator = (const Event& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_event_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetEventInfo, object_, name, param),
-            __GET_EVENT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_event_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_event_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getProfilingInfo(cl_profiling_info name, T* param) const
-    {
-        return detail::errHandler(detail::getInfo(
-            &::clGetEventProfilingInfo, object_, name, param),
-            __GET_EVENT_PROFILE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_profiling_info, name>::param_type
-    getProfilingInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_profiling_info, name>::param_type param;
-        cl_int result = getProfilingInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int wait() const
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(1, &object_),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setCallback(
-        cl_int type,
-        void (CL_CALLBACK * pfn_notify)(cl_event, cl_int, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetEventCallback(
-                object_,
-                type,
-                pfn_notify,
-                user_data), 
-            __SET_EVENT_CALLBACK_ERR);
-    }
-#endif
-
-    static cl_int
-    waitForEvents(const VECTOR_CLASS<Event>& events)
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(
-                (cl_uint) events.size(), (cl_event*)&events.front()),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Event)
-
-#if defined(CL_VERSION_1_1)
-/*! \class UserEvent
- * \brief User event interface for cl_event.
- */
-class UserEvent : public Event
-{
-public:
-    UserEvent(
-        const Context& context,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateUserEvent(
-            context(),
-            &error);
-
-        detail::errHandler(error, __CREATE_USER_EVENT_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    UserEvent() : Event() { }
-
-    UserEvent(const UserEvent& event) : Event(event) { }
-
-    UserEvent& operator = (const UserEvent& rhs)
-    {
-        if (this != &rhs) {
-            Event::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int setStatus(cl_int status)
-    {
-        return detail::errHandler(
-            ::clSetUserEventStatus(object_,status), 
-            __SET_USER_EVENT_STATUS_ERR);
-    }
-};
-#endif
-
-inline static cl_int
-WaitForEvents(const VECTOR_CLASS<Event>& events)
-{
-    return detail::errHandler(
-        ::clWaitForEvents(
-            (cl_uint) events.size(), (cl_event*)&events.front()),
-        __WAIT_FOR_EVENTS_ERR);
-}
-
-/*! \class Memory
- * \brief Memory interface for cl_mem.
- */
-class Memory : public detail::Wrapper<cl_mem>
-{
-public:
-    Memory() : detail::Wrapper<cl_type>() { }
-
-    Memory(const Memory& memory) : detail::Wrapper<cl_type>(memory) { }
-
-    Memory& operator = (const Memory& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_mem_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetMemObjectInfo, object_, name, param),
-            __GET_MEM_OBJECT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_mem_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_mem_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setDestructorCallback(
-        void (CL_CALLBACK * pfn_notify)(cl_mem, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetMemObjectDestructorCallback(
-                object_,
-                pfn_notify,
-                user_data), 
-            __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR);
-    }
-#endif
-
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Memory)
-
-/*! \class Buffer
- * \brief Memory buffer interface.
- */
-class Buffer : public Memory
-{
-public:
-    Buffer(
-        const Context& context,
-        cl_mem_flags flags,
-        ::size_t size,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateBuffer(context(), flags, size, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Buffer() : Memory() { }
-
-    Buffer(const Buffer& buffer) : Memory(buffer) { }
-
-    Buffer& operator = (const Buffer& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-
-#if defined(CL_VERSION_1_1)
-    Buffer createSubBuffer(
-        cl_mem_flags flags,
-        cl_buffer_create_type buffer_create_type,
-        const void * buffer_create_info,
-        cl_int * err = NULL)
-    {
-        Buffer result;
-        cl_int error;
-        result.object_ = ::clCreateSubBuffer(
-            object_, 
-            flags, 
-            buffer_create_type, 
-            buffer_create_info, 
-            &error);
-
-        detail::errHandler(error, __CREATE_SUBBUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-
-        return result;
-	}		
-#endif
-};
-
-#if defined (USE_DX_INTEROP)
-class BufferD3D10 : public Buffer
-{
-public:
-    typedef CL_API_ENTRY cl_mem (CL_API_CALL *PFN_clCreateFromD3D10BufferKHR)(
-    cl_context context, cl_mem_flags flags, ID3D10Buffer*  buffer,
-    cl_int* errcode_ret);
-
-    BufferD3D10(
-        const Context& context,
-        cl_mem_flags flags,
-        ID3D10Buffer* bufobj,
-        cl_int * err = NULL)
-    {
-        static PFN_clCreateFromD3D10BufferKHR pfn_clCreateFromD3D10BufferKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clCreateFromD3D10BufferKHR);
-
-        cl_int error;
-        object_ = pfn_clCreateFromD3D10BufferKHR(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferD3D10() : Buffer() { }
-
-    BufferD3D10(const BufferD3D10& buffer) : Buffer(buffer) { }
-
-    BufferD3D10& operator = (const BufferD3D10& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-};
-#endif
-
-/*! \class BufferGL
- * \brief Memory buffer interface for GL interop.
- */
-class BufferGL : public Buffer
-{
-public:
-    BufferGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLBuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferGL() : Buffer() { }
-
-    BufferGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferGL& operator = (const BufferGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class BufferRenderGL
- * \brief Memory buffer interface for GL interop with renderbuffer.
- */
-class BufferRenderGL : public Buffer
-{
-public:
-    BufferRenderGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLRenderbuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferRenderGL() : Buffer() { }
-
-    BufferRenderGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferRenderGL& operator = (const BufferRenderGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class Image
- * \brief Base class  interface for all images.
- */
-class Image : public Memory
-{
-protected:
-    Image() : Memory() { }
-
-    Image(const Image& image) : Memory(image) { }
-
-    Image& operator = (const Image& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-public:
-    template <typename T>
-    cl_int getImageInfo(cl_image_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetImageInfo, object_, name, param),
-            __GET_IMAGE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_image_info, name>::param_type
-    getImageInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_image_info, name>::param_type param;
-        cl_int result = getImageInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-/*! \class Image2D
- * \brief Image interface for 2D images.
- */
-class Image2D : public Image
-{
-public:
-    Image2D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t row_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage2D(
-            context(), flags,&format, width, height, row_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE2D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2D() { }
-
-    Image2D(const Image2D& image2D) : Image(image2D) { }
-
-    Image2D& operator = (const Image2D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image2DGL : public Image2D
-{
-public:
-    Image2DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture2D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2DGL() : Image2D() { }
-
-    Image2DGL(const Image2DGL& image) : Image2D(image) { }
-
-    Image2DGL& operator = (const Image2DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image2D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image3D
- * \brief Image interface for 3D images.
- */
-class Image3D : public Image
-{
-public:
-    Image3D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t depth,
-        ::size_t row_pitch = 0,
-        ::size_t slice_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage3D(
-            context(), flags, &format, width, height, depth, row_pitch,
-            slice_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE3D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3D() { }
-
-    Image3D(const Image3D& image3D) : Image(image3D) { }
-
-    Image3D& operator = (const Image3D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image3DGL : public Image3D
-{
-public:
-    Image3DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture3D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3DGL() : Image3D() { }
-
-    Image3DGL(const Image3DGL& image) : Image3D(image) { }
-
-    Image3DGL& operator = (const Image3DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image3D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Sampler
- * \brief Sampler interface for cl_sampler.
- */
-class Sampler : public detail::Wrapper<cl_sampler>
-{
-public:
-    Sampler() { }
-
-    Sampler(
-        const Context& context,
-        cl_bool normalized_coords,
-        cl_addressing_mode addressing_mode,
-        cl_filter_mode filter_mode,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateSampler(
-            context(), 
-            normalized_coords,
-            addressing_mode,
-            filter_mode,
-            &error);
-
-        detail::errHandler(error, __CREATE_SAMPLER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Sampler(const Sampler& sampler) : detail::Wrapper<cl_type>(sampler) { }
-
-    Sampler& operator = (const Sampler& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_sampler_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetSamplerInfo, object_, name, param),
-            __GET_SAMPLER_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_sampler_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_sampler_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Sampler)
-
-class Program;
-class CommandQueue;
-class Kernel;
-
-/*! \class NDRange
- * \brief NDRange interface
- */
-class NDRange
-{
-private:
-    size_t<3> sizes_;
-    cl_uint dimensions_;
-
-public:
-    NDRange()
-        : dimensions_(0)
-    { }
-
-    NDRange(::size_t size0)
-        : dimensions_(1)
-    {
-        sizes_.push_back(size0);
-    }
-
-    NDRange(::size_t size0, ::size_t size1)
-        : dimensions_(2)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-    }
-
-    NDRange(::size_t size0, ::size_t size1, ::size_t size2)
-        : dimensions_(3)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-        sizes_.push_back(size2);
-    }
-
-    operator const ::size_t*() const { return (const ::size_t*) sizes_; }
-    ::size_t dimensions() const { return dimensions_; }
-};
-
-static const NDRange NullRange;
-
-/*!
- * \struct LocalSpaceArg
- * \brief Local address raper for use with Kernel::setArg
- */
-struct LocalSpaceArg
-{
-    ::size_t size_;
-};
-
-namespace detail {
-
-template <typename T>
-struct KernelArgumentHandler
-{
-    static ::size_t size(const T&) { return sizeof(T); }
-    static T* ptr(T& value) { return &value; }
-};
-
-template <>
-struct KernelArgumentHandler<LocalSpaceArg>
-{
-    static ::size_t size(const LocalSpaceArg& value) { return value.size_; }
-    static void* ptr(LocalSpaceArg&) { return NULL; }
-};
-
-} 
-//! \endcond
-
-inline LocalSpaceArg
-__local(::size_t size)
-{
-    LocalSpaceArg ret = { size };
-    return ret;
-}
-
-class KernelFunctor;
-
-/*! \class Kernel
- * \brief Kernel interface that implements cl_kernel
- */
-class Kernel : public detail::Wrapper<cl_kernel>
-{
-public:
-    inline Kernel(const Program& program, const char* name, cl_int* err = NULL);
-
-    Kernel() { }
-
-    Kernel(const Kernel& kernel) : detail::Wrapper<cl_type>(kernel) { }
-
-    Kernel& operator = (const Kernel& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_kernel_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetKernelInfo, object_, name, param),
-            __GET_KERNEL_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_kernel_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getWorkGroupInfo(
-        const Device& device, cl_kernel_work_group_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetKernelWorkGroupInfo, object_, device(), name, param),
-                __GET_KERNEL_WORK_GROUP_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_work_group_info, name>::param_type
-        getWorkGroupInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-        detail::cl_kernel_work_group_info, name>::param_type param;
-        cl_int result = getWorkGroupInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int setArg(cl_uint index, T value)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(
-                object_,
-                index,
-                detail::KernelArgumentHandler<T>::size(value),
-                detail::KernelArgumentHandler<T>::ptr(value)),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    cl_int setArg(cl_uint index, ::size_t size, void* argPtr)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(object_, index, size, argPtr),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local);
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& global,
-        const NDRange& local);
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Kernel)
-
-/*! \class Program
- * \brief Program interface that implements cl_program.
- */
-class Program : public detail::Wrapper<cl_program>
-{
-public:
-    typedef VECTOR_CLASS<std::pair<const void*, ::size_t> > Binaries;
-    typedef VECTOR_CLASS<std::pair<const char*, ::size_t> > Sources;
-
-    Program(
-        const Context& context,
-        const Sources& sources,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-
-        const ::size_t n = (::size_t)sources.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const char** strings = (const char**) alloca(n * sizeof(const char*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            strings[i] = sources[(int)i].first;
-            lengths[i] = sources[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithSource(
-            context(), (cl_uint)n, strings, lengths, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_SOURCE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program(
-        const Context& context,
-        const VECTOR_CLASS<Device>& devices,
-        const Binaries& binaries,
-        VECTOR_CLASS<cl_int>* binaryStatus = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        const ::size_t n = binaries.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const unsigned char** images = (const unsigned char**) alloca(n * sizeof(const void*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            images[i] = (const unsigned char*)binaries[(int)i].first;
-            lengths[i] = binaries[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithBinary(
-            context(), (cl_uint) devices.size(),
-            (cl_device_id*)&devices.front(),
-            lengths, images, binaryStatus != NULL
-               ? (cl_int*) &binaryStatus->front()
-               : NULL, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_BINARY_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program() { }
-
-    Program(const Program& program) : detail::Wrapper<cl_type>(program) { }
-
-    Program& operator = (const Program& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int build(
-        const VECTOR_CLASS<Device>& devices,
-        const char* options = NULL,
-        void (CL_CALLBACK * notifyFptr)(cl_program, void *) = NULL,
-        void* data = NULL) const
-    {
-        return detail::errHandler(
-            ::clBuildProgram(
-                object_,
-                (cl_uint)
-                devices.size(),
-                (cl_device_id*)&devices.front(),
-                options,
-                notifyFptr,
-                data),
-                __BUILD_PROGRAM_ERR);
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_program_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetProgramInfo, object_, name, param),
-            __GET_PROGRAM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getBuildInfo(
-        const Device& device, cl_program_build_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetProgramBuildInfo, object_, device(), name, param),
-                __GET_PROGRAM_BUILD_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_build_info, name>::param_type
-    getBuildInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_build_info, name>::param_type param;
-        cl_int result = getBuildInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int createKernels(VECTOR_CLASS<Kernel>* kernels)
-    {
-        cl_uint numKernels;
-        cl_int err = ::clCreateKernelsInProgram(object_, 0, NULL, &numKernels);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        Kernel* value = (Kernel*) alloca(numKernels * sizeof(Kernel));
-        err = ::clCreateKernelsInProgram(
-            object_, numKernels, (cl_kernel*) value, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        kernels->assign(&value[0], &value[numKernels]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Program)
-
-inline Kernel::Kernel(const Program& program, const char* name, cl_int* err)
-{
-    cl_int error;
-
-    object_ = ::clCreateKernel(program(), name, &error);
-    detail::errHandler(error, __CREATE_KERNEL_ERR);
-
-    if (err != NULL) {
-        *err = error;
-    }
-
-}
-
-/*! \class CommandQueue
- * \brief CommandQueue interface for cl_command_queue.
- */
-class CommandQueue : public detail::Wrapper<cl_command_queue>
-{
-public:
-    CommandQueue(
-        const Context& context,
-        const Device& device,
-        cl_command_queue_properties properties = 0,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateCommandQueue(
-            context(), device(), properties, &error);
-
-        detail::errHandler(error, __CREATE_COMMAND_QUEUE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    CommandQueue() { }
-
-    CommandQueue(const CommandQueue& commandQueue) : detail::Wrapper<cl_type>(commandQueue) { }
-
-    CommandQueue& operator = (const CommandQueue& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_command_queue_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetCommandQueueInfo, object_, name, param),
-                __GET_COMMAND_QUEUE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_command_queue_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_command_queue_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int enqueueReadBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_BUFFER_ERR);
-    }
-
-    cl_int enqueueWriteBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        const void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBuffer(
-        const Buffer& src,
-        const Buffer& dst,
-        ::size_t src_offset,
-        ::size_t dst_offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBuffer(
-                object_, src(), dst(), src_offset, dst_offset, size,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int enqueueReadBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_READ_BUFFER_RECT_ERR);
-    }
-
-
-    cl_int enqueueWriteBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_RECT_ERR);
-    }
-
-    cl_int enqueueCopyBufferRect(
-        const Buffer& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        ::size_t src_row_pitch,
-        ::size_t src_slice_pitch,
-        ::size_t dst_row_pitch,
-        ::size_t dst_slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferRect(
-                object_, 
-                src(), 
-                dst(), 
-                (const ::size_t *)src_origin, 
-                (const ::size_t *)dst_origin, 
-                (const ::size_t *)region,
-                src_row_pitch,
-                src_slice_pitch,
-                dst_row_pitch,
-                dst_slice_pitch,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_RECT_ERR);
-    }
-#endif
-
-    cl_int enqueueReadImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_IMAGE_ERR);
-    }
-
-    cl_int enqueueWriteImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_WRITE_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImage(
-        const Image& src,
-        const Image& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImage(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *)dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImageToBuffer(
-        const Image& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& region,
-        ::size_t dst_offset,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImageToBuffer(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *) region, dst_offset,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBufferToImage(
-        const Buffer& src,
-        const Image& dst,
-        ::size_t src_offset,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferToImage(
-                object_, src(), dst(), src_offset,
-                (const ::size_t *) dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR);
-    }
-
-    void* enqueueMapBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        ::size_t offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapBuffer(
-            object_, buffer(), blocking, flags, offset, size,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-        return result;
-    }
-
-    void* enqueueMapImage(
-        const Image& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t * row_pitch,
-        ::size_t * slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapImage(
-            object_, buffer(), blocking, flags,
-            (const ::size_t *) origin, (const ::size_t *) region,
-            row_pitch, slice_pitch,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_IMAGE_ERR);
-        if (err != NULL) {
-              *err = error;
-        }
-        return result;
-    }
-
-    cl_int enqueueUnmapMemObject(
-        const Memory& memory,
-        void* mapped_ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueUnmapMemObject(
-                object_, memory(), mapped_ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_UNMAP_MEM_OBJECT_ERR);
-    }
-
-    cl_int enqueueNDRangeKernel(
-        const Kernel& kernel,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueNDRangeKernel(
-                object_, kernel(), (cl_uint) global.dimensions(),
-                offset.dimensions() != 0 ? (const ::size_t*) offset : NULL,
-                (const ::size_t*) global,
-                local.dimensions() != 0 ? (const ::size_t*) local : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NDRANGE_KERNEL_ERR);
-    }
-
-    cl_int enqueueTask(
-        const Kernel& kernel,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueTask(
-                object_, kernel(),
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_TASK_ERR);
-    }
-
-    cl_int enqueueNativeKernel(
-        void (*userFptr)(void *),
-        std::pair<void*, ::size_t> args,
-        const VECTOR_CLASS<Memory>* mem_objects = NULL,
-        const VECTOR_CLASS<const void*>* mem_locs = NULL,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        cl_mem * mems = (mem_objects != NULL && mem_objects->size() > 0) 
-            ? (cl_mem*) alloca(mem_objects->size() * sizeof(cl_mem))
-            : NULL;
-
-        if (mems != NULL) {
-            for (unsigned int i = 0; i < mem_objects->size(); i++) {
-                mems[i] = ((*mem_objects)[i])();
-            }
-        }
-
-        return detail::errHandler(
-            ::clEnqueueNativeKernel(
-                object_, userFptr, args.first, args.second,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                mems,
-                (mem_locs != NULL) ? (const void **) &mem_locs->front() : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NATIVE_KERNEL);
-    }
-
-    cl_int enqueueMarker(Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueMarker(object_, (cl_event*) event),
-            __ENQUEUE_MARKER_ERR);
-    }
-
-    cl_int enqueueWaitForEvents(const VECTOR_CLASS<Event>& events) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWaitForEvents(
-                object_,
-                (cl_uint) events.size(),
-                (const cl_event*) &events.front()),
-            __ENQUEUE_WAIT_FOR_EVENTS_ERR);
-    }
-
-    cl_int enqueueAcquireGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueAcquireGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueReleaseGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_RELEASE_GL_ERR);
-     }
-
-#if defined (USE_DX_INTEROP)
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueAcquireD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects, cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueReleaseD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects,  cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-
-    cl_int enqueueAcquireD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         static PFN_clEnqueueAcquireD3D10ObjectsKHR pfn_clEnqueueAcquireD3D10ObjectsKHR = NULL;
-         __INIT_CL_EXT_FCN_PTR(clEnqueueAcquireD3D10ObjectsKHR);
-		
-         return detail::errHandler(
-             pfn_clEnqueueAcquireD3D10ObjectsKHR(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-    {
-        static PFN_clEnqueueReleaseD3D10ObjectsKHR pfn_clEnqueueReleaseD3D10ObjectsKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clEnqueueReleaseD3D10ObjectsKHR);
-
-        return detail::errHandler(
-            pfn_clEnqueueReleaseD3D10ObjectsKHR(
-                object_,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_RELEASE_GL_ERR);
-    }
-#endif
-
-    cl_int enqueueBarrier() const
-    {
-        return detail::errHandler(
-            ::clEnqueueBarrier(object_),
-            __ENQUEUE_BARRIER_ERR);
-    }
-
-    cl_int flush() const
-    {
-        return detail::errHandler(::clFlush(object_), __FLUSH_ERR);
-    }
-
-    cl_int finish() const
-    {
-        return detail::errHandler(::clFinish(object_), __FINISH_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::CommandQueue)
-
-/*! \class KernelFunctor
- * \brief Kernel functor interface
- *
- * \note Currently only functors of zero to ten arguments are supported. It
- * is straightforward to add more and a more general solution, similar to
- * Boost.Lambda could be followed if required in the future.
- */
-class KernelFunctor
-{
-private:
-    Kernel kernel_;
-    CommandQueue queue_;
-    NDRange offset_;
-    NDRange global_;
-    NDRange local_;
-
-    cl_int err_;
-public:
-    KernelFunctor() { }
-
-    KernelFunctor(
-        const Kernel& kernel,
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local) :
-            kernel_(kernel),
-            queue_(queue),
-            offset_(offset),
-            global_(global),
-            local_(local),
-            err_(CL_SUCCESS)
-    {}
-
-    KernelFunctor& operator=(const KernelFunctor& rhs);
-
-    KernelFunctor(const KernelFunctor& rhs);
-
-    cl_int getError() { return err_; }
-
-    inline Event operator()(const VECTOR_CLASS<Event>* events = NULL);
-
-    template<typename A1>
-    inline Event operator()(
-        const A1& a1, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5, class A6>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4,
-             class A5, class A6, class A7>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12, 
-        const A13& a13,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14, class A15>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14, 
-        const A15& a15,
-        const VECTOR_CLASS<Event>* events = NULL);
-};
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& offset,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,offset,global,local);
-}
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,NullRange,global,local);
-}
-
-inline KernelFunctor& KernelFunctor::operator=(const KernelFunctor& rhs)
-{
-    if (this == &rhs) {
-        return *this;
-    }
-    
-    kernel_ = rhs.kernel_;
-    queue_  = rhs.queue_;
-    offset_ = rhs.offset_;
-    global_ = rhs.global_;
-    local_  = rhs.local_;
-    
-    return *this;
-}
-
-inline KernelFunctor::KernelFunctor(const KernelFunctor& rhs) :
-    kernel_(rhs.kernel_),
-    queue_(rhs.queue_),
-    offset_(rhs.offset_),
-    global_(rhs.global_),
-    local_(rhs.local_)
-{
-}
-
-Event KernelFunctor::operator()(const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4,
-         typename A5, typename A6, typename A7>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9, typename A10>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12, 
-    const A13& a13,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14, class A15>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14, 
-    const A15& a15,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-    kernel_.setArg(14,a15);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-#undef __ERR_STR
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#undef __GET_DEVICE_INFO_ERR
-#undef __GET_PLATFORM_INFO_ERR
-#undef __GET_DEVICE_IDS_ERR
-#undef __GET_CONTEXT_INFO_ERR
-#undef __GET_EVENT_INFO_ERR
-#undef __GET_EVENT_PROFILE_INFO_ERR
-#undef __GET_MEM_OBJECT_INFO_ERR
-#undef __GET_IMAGE_INFO_ERR
-#undef __GET_SAMPLER_INFO_ERR
-#undef __GET_KERNEL_INFO_ERR
-#undef __GET_KERNEL_WORK_GROUP_INFO_ERR
-#undef __GET_PROGRAM_INFO_ERR
-#undef __GET_PROGRAM_BUILD_INFO_ERR
-#undef __GET_COMMAND_QUEUE_INFO_ERR
-
-#undef __CREATE_CONTEXT_FROM_TYPE_ERR
-#undef __GET_SUPPORTED_IMAGE_FORMATS_ERR
-
-#undef __CREATE_BUFFER_ERR
-#undef __CREATE_SUBBUFFER_ERR
-#undef __CREATE_IMAGE2D_ERR
-#undef __CREATE_IMAGE3D_ERR
-#undef __CREATE_SAMPLER_ERR
-#undef __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR
-
-#undef __CREATE_USER_EVENT_ERR
-#undef __SET_USER_EVENT_STATUS_ERR
-#undef __SET_EVENT_CALLBACK_ERR
-
-#undef __WAIT_FOR_EVENTS_ERR
-
-#undef __CREATE_KERNEL_ERR
-#undef __SET_KERNEL_ARGS_ERR
-#undef __CREATE_PROGRAM_WITH_SOURCE_ERR
-#undef __CREATE_PROGRAM_WITH_BINARY_ERR
-#undef __BUILD_PROGRAM_ERR
-#undef __CREATE_KERNELS_IN_PROGRAM_ERR
-
-#undef __CREATE_COMMAND_QUEUE_ERR
-#undef __SET_COMMAND_QUEUE_PROPERTY_ERR
-#undef __ENQUEUE_READ_BUFFER_ERR
-#undef __ENQUEUE_WRITE_BUFFER_ERR
-#undef __ENQUEUE_READ_BUFFER_RECT_ERR
-#undef __ENQUEUE_WRITE_BUFFER_RECT_ERR
-#undef __ENQEUE_COPY_BUFFER_ERR
-#undef __ENQEUE_COPY_BUFFER_RECT_ERR
-#undef __ENQUEUE_READ_IMAGE_ERR
-#undef __ENQUEUE_WRITE_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR
-#undef __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR
-#undef __ENQUEUE_MAP_BUFFER_ERR
-#undef __ENQUEUE_MAP_IMAGE_ERR
-#undef __ENQUEUE_UNMAP_MEM_OBJECT_ERR
-#undef __ENQUEUE_NDRANGE_KERNEL_ERR
-#undef __ENQUEUE_TASK_ERR
-#undef __ENQUEUE_NATIVE_KERNEL
-
-#undef __UNLOAD_COMPILER_ERR
-#endif //__CL_USER_OVERRIDE_ERROR_STRINGS
-
-#undef __GET_INFO_HELPER_WITH_RETAIN
-
-// Extensions
-#undef __INIT_CL_EXT_FCN_PTR
-#undef __CREATE_SUB_DEVICES
-
-#if defined(USE_CL_DEVICE_FISSION)
-#undef __PARAM_NAME_DEVICE_FISSION
-#endif // USE_CL_DEVICE_FISSION
-
-} // namespace cl
-
-#endif // CL_HPP_
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image/gray_scott_opencl_image.cpp b/SpeedComparisons/GrayScott_OpenCL_Image/gray_scott_opencl_image.cpp
deleted file mode 100644
index 40d305ad0..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image/gray_scott_opencl_image.cpp
+++ /dev/null
@@ -1,363 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCL:
-#define __NO_STD_VECTOR // Use cl::vector instead of STL version
-#define __CL_ENABLE_EXCEPTIONS
-
-// cl.hpp is standard but doesn't come with most SDKs, so download it from here:
-// http://www.khronos.org/registry/cl/api/1.1/cl.hpp
-#ifdef __APPLE__
-# include "cl.hpp"
-#else
-# include <CL/cl.hpp>
-#endif
-
-using namespace cl;
-
-// STL:
-#include <fstream>
-#include <iostream>
-
-// local:
-#include "defs.h"
-
-void init(float *chemicals);
-bool display(float *chemicals,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
-
-static int g_opt_device = 0;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if ((i+1<argc) && (strcmp(argv[i],"-device")==0)) {
-            // select an output device
-            i++; g_opt_device = atoi(argv[i]);
-        } else {
-            std::cout << "Unrecognized argument: '" << argv[i] << "'\n";
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    float k, f;
-    k = 0.064f; f = 0.035f; // solitons with mitosis (spots that multiply)
-    // k = 0.06f; f = 0.035f; // stripes
-    // k = 0.065f; f = 0.056f; // long stripes
-    //k = 0.064f; f = 0.04f; // dots and stripes
-    // k = 0.0475f; f = 0.0118f; // spiral waves
-    float speed = 2.0f;
-    // ----------------
-    
-    try { 
-        // Get available OpenCL platforms
-        cl::vector<Platform> platforms;
-        Platform::get(&platforms);
- 
-        // Select the default platform and create a context using this platform and the GPU
-        cl_context_properties cps[3] = { 
-            CL_CONTEXT_PLATFORM, 
-            (cl_context_properties)(platforms[0])(), 
-            0 
-        };
-        Context context( CL_DEVICE_TYPE_GPU, cps);
-
-        float *chem1 = new float[X*Y*4]; // we store up to 4 chemicals in an RGBA image
-        float *chem2 = new float[X*Y*4];
-        init(chem1);
-
-        // Get a list of devices on this platform
-        cl::vector<Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
-
-        // range-check the user's selection
-        int maxdev = devices.size() - 1;
-        g_opt_device = (g_opt_device > maxdev) ? maxdev :
-                                     ((g_opt_device < 0) ? 0 : g_opt_device);
-        std::cout << (maxdev+1) << " device(s) available; using device "
-                                                    << g_opt_device << ".\n";
-
-        Device &device = devices[g_opt_device];
-        cl::vector<Device> ourdevices = cl::vector<Device>(1, device);
-
-
-        bool is_ImageSupported = device.getInfo<CL_DEVICE_IMAGE_SUPPORT>();
-        if(!is_ImageSupported)
-        {
-            printf("Images not supported on this device.\n");
-            exit(-1);
-        }
-
-        // we make two images and swap between them
-        cl::Image2D chemicals1(context,CL_MEM_READ_WRITE,cl::ImageFormat(CL_RGBA,CL_FLOAT),X,Y);
-        cl::Image2D chemicals2(context,CL_MEM_READ_WRITE,cl::ImageFormat(CL_RGBA,CL_FLOAT),X,Y);
-
-        // Create a command queue and use the selected device
-        CommandQueue queue = CommandQueue(context, device);
-        Event event;
- 
-        // Copy to the memory buffers
-        cl::size_t<3> origin;
-        origin.push_back(0);
-        origin.push_back(0);
-        origin.push_back(0);
-        cl::size_t<3> region;
-        region.push_back(X);
-        region.push_back(Y);
-        region.push_back(1);
-        queue.enqueueWriteImage(chemicals1,true,origin,region,0,0,chem1);
-        queue.enqueueWriteImage(chemicals2,true,origin,region,0,0,chem2);
- 
-        // Read source file
-        std::string kfn = CL_SOURCE_DIR; // (defined in CMakeLists.txt to be the source folder)
-        kfn += "/grayscott_kernel_image.cl";
-        std::ifstream sourceFile(kfn.c_str());
-        std::string sourceCode(
-            std::istreambuf_iterator<char>(sourceFile),
-            (std::istreambuf_iterator<char>()));
-        Program::Sources source(1, std::make_pair(sourceCode.c_str(), sourceCode.length()+1));
- 
-        // enable this code to display kernel compilation error if you get clBuildProgram(-11)
-        #if 0
-           const ::size_t n = (::size_t)source.size();
-           ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-           const char** strings = (const char**) alloca(n * sizeof(const char*));
-           for (::size_t i = 0; i < n; ++i) {
-               strings[i] = source[(int)i].first;
-               lengths[i] = source[(int)i].second;
-           }
-           cl_int err;
-           cl_program myprog = clCreateProgramWithSource(context(), (cl_uint)n, strings, lengths, &err);
-           err = clBuildProgram(myprog, (cl_uint)ourdevices.size(), (cl_device_id*)&ourdevices.front(), NULL, NULL, NULL);
-           char proglog[1024];
-           clGetProgramBuildInfo(myprog, device(), CL_PROGRAM_BUILD_LOG, 1024, proglog, 0);
-           printf("err=%d log=%s\n", err, proglog);
-           return 0;
-        #endif
-
-        // Make program of the source code in the context
-        Program program = Program(context, source);
- 
-        // Build program for the specific device we are using
-        // IMPORTANT: If this program is running on a system that has multiple
-        // graphics cards, and if ANY of those cards does not support Images,
-        // then "program.build(devices);" will fail. Thus, we must build only
-        // on the device that is actually being used for the command queue.
-        program.build(ourdevices);
- 
-        // Make kernel
-        Kernel kernel(program, "grayscott_compute");
-
-        NDRange global(X,Y);
-        NDRange local(16,16);
-
-        kernel.setArg(2, f);
-        kernel.setArg(3, f+k);
-        kernel.setArg(4, r_a);
-        kernel.setArg(5, r_b);
-        kernel.setArg(6, speed);
-
-        int iteration = 0;
-        float fps_avg = 0.0; // decaying average of fps
-        const int N_FRAMES_PER_DISPLAY = 5000;  // an even number, because of our double-buffering implementation
-        while(true) 
-        {
-            struct timeval tod_record;
-            double tod_before, tod_after, tod_elap;
-
-            gettimeofday(&tod_record, 0);
-            tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            // run a few iterations (without copying the data back)
-            for(int it=0;it<N_FRAMES_PER_DISPLAY/2;it++)
-            {
-                // (buffer-switching)
-
-                kernel.setArg(0, chemicals1);
-                kernel.setArg(1, chemicals2);
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-
-                kernel.setArg(0, chemicals2);
-                kernel.setArg(1, chemicals1);
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-            }
-            iteration += N_FRAMES_PER_DISPLAY;
-
-            // read back into a buffer
-            queue.enqueueReadImage(chemicals1,true,origin,region,0,0,chem1);
-            //queue.enqueueReadImage(chemicals2,true,origin,region,0,0,chem2);
-
-            gettimeofday(&tod_record, 0);
-            tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            tod_elap = tod_after - tod_before;
-
-            char msg[1000];
-            float fps = 0.0f;     // frames per second
-            if (tod_elap > 0)
-                fps = N_FRAMES_PER_DISPLAY / (float)tod_elap;
-            // We display an exponential moving average of the fps measurement
-            fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0f) + fps) / 11.0f);
-            sprintf(msg,"GrayScott - %0.2f fps (%.2f Mcgs)", fps_avg,fps_avg*X*Y/1e6);
-
-            // display:
-            {
-                int quitnow = display(chem1,iteration,false,200.0f,2,10,msg);
-                if (quitnow)
-                    break;
-            }
-        }
-    } 
-    catch(Error error) 
-    {
-       std::cout << error.what() << "(" << error.err() << ")" << std::endl;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float* chemicals)
-{
-    srand((unsigned int)time(NULL));
-
-    // figure the values
-    for(int i = 0; i < X; i++) 
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            int index = (j*X+i)*4;
-            // start with a uniform field with an approximate circle in the middle
-            if(hypot(i-X/3,(j-Y/4)/1.5)<=frand(2,5))
-            {
-                chemicals[index+0] = frand(0.0f,0.1f);
-                chemicals[index+1] = frand(0.9f,1.0f);
-            }
-            else 
-            {
-                chemicals[index+0] = frand(0.9f,1.0f);
-                chemicals[index+1] = frand(0.0f,0.1f);
-            }
-
-        }
-    }
-}
-
-
-bool display(float *chemicals,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2,*im3;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    for(int col=0;col<X;col++)
-    {
-        for(int row=0;row<Y;row++)
-        {
-            float val = chemicals[(row*X+col)*4 + 0];
-            val *= manual_brighten;
-            if(val<0) val=0; if(val>255) val=255;
-            ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 2] = (uchar)val;
-            val = chemicals[(row*X+col)*4 + 0];
-            val *= manual_brighten;
-            if(val<0) val=0; if(val>255) val=255;
-            ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 1] = (uchar)val;
-            val = chemicals[(row*X+col)*4 + 0];
-            val *= manual_brighten;
-            if(val<0) val=0; if(val>255) val=255;
-            ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 0] = (uchar)val;
-        }
-    }
-
-    cvResize(im,im2);
-
-    {
-        char txt[100];
-        sprintf(txt,"%d",iteration);
-        cvPutText(im2,txt,cvPoint(20,20),&font,white);
-        cvPutText(im2,message,cvPoint(20,40),&font,white);
-    }
-
-    cvShowImage(title,im2);
-
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        return true;
-    }
-    return false;
-}
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image/grayscott_kernel_image.cl b/SpeedComparisons/GrayScott_OpenCL_Image/grayscott_kernel_image.cl
deleted file mode 100644
index 386704898..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image/grayscott_kernel_image.cl
+++ /dev/null
@@ -1,32 +0,0 @@
-__kernel void grayscott_compute (
-    read_only image2d_t input, 
-    write_only image2d_t output,
-    float f,float f_plus_k,
-    float r_a,float r_b,
-    float speed)
-{
-    const sampler_t smp = CLK_NORMALIZED_COORDS_FALSE | CLK_ADDRESS_REPEAT | CLK_FILTER_NEAREST;
-
-    const int x = get_global_id(0);
-    const int y = get_global_id(1);
-
-    const float4 pixel = read_imagef(input, smp, (int2)(x, y));
-    const float4 left = read_imagef(input, smp, (int2)(x-1, y));
-    const float4 right = read_imagef(input, smp, (int2)(x+1, y));
-    const float4 up = read_imagef(input, smp, (int2)(x, y-1));
-    const float4 down = read_imagef(input, smp, (int2)(x, y+1));
- 
-    const float4 laplacian = left + right + up + down - 4.0f*pixel;
-
-    {    
-        // for Gray-Scott we only use the first two components:
-        
-        const float da = r_a * laplacian.x - pixel.x*pixel.y*pixel.y + f*(1-pixel.x);
-        const float db = r_b * laplacian.y + pixel.x*pixel.y*pixel.y - f_plus_k*pixel.y;
-        
-        pixel.x += speed * da;
-        pixel.y += speed * db;
-    }
-
-    write_imagef(output, (int2)(x, y), pixel);
-}
\ No newline at end of file
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenCL_Image_2x2/CMakeLists.txt
deleted file mode 100644
index 8bf3ec0d8..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/CMakeLists.txt
+++ /dev/null
@@ -1,27 +0,0 @@
-project(GrayScott_OpenCL_Image_2x2)
-
-set(CMAKE_MODULE_PATH ${GrayScott_OpenCL_Image_2x2_SOURCE_DIR})
-# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
-
-find_package(OpenCV REQUIRED)
-include_directories( ${OPENCV_INCLUDE_DIR})
-link_libraries( ${OpenCV_LIBS} )
-
-# only build the OpenCL version if OpenCL was found
-find_package ( OpenCL )
-if(OPENCL_FOUND)
-  include_directories ( ${OPENCL_INCLUDE_DIRS} )
-  link_libraries ( ${OPENCL_LIBRARIES} )
-
-  # tell the code where the .cl file will live
-  add_definitions(-DCL_SOURCE_DIR="${GrayScott_OpenCL_Image_2x2_SOURCE_DIR}")
-  
-  include_directories(../Display)
-
-  add_executable(GrayScott_OpenCL_Image_2x2
-    gray_scott_opencl_image_2x2.cpp 
-    grayscott_kernel_image_2x2.cl 
-    ../Display/defs.h
-  )
-endif()
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/FindOpenCL.cmake b/SpeedComparisons/GrayScott_OpenCL_Image_2x2/FindOpenCL.cmake
deleted file mode 100644
index fde90efae..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/FindOpenCL.cmake
+++ /dev/null
@@ -1,79 +0,0 @@
-# - Try to find OpenCL
-# This module tries to find an OpenCL implementation on your system. It supports
-# AMD / ATI, Apple and NVIDIA implementations, but shoudl work, too.
-#
-# Once done this will define
-#  OPENCL_FOUND        - system has OpenCL
-#  OPENCL_INCLUDE_DIRS  - the OpenCL include directory
-#  OPENCL_LIBRARIES    - link these to use OpenCL
-#
-# WIN32 should work, but is untested
-
-FIND_PACKAGE( PackageHandleStandardArgs )
-
-SET (OPENCL_VERSION_STRING "0.1.0")
-SET (OPENCL_VERSION_MAJOR 0)
-SET (OPENCL_VERSION_MINOR 1)
-SET (OPENCL_VERSION_PATCH 0)
-
-IF (APPLE)
-
-  FIND_LIBRARY(OPENCL_LIBRARIES OpenCL DOC "OpenCL lib for OSX")
-  FIND_PATH(OPENCL_INCLUDE_DIRS OpenCL/cl.h DOC "Include for OpenCL on OSX")
-  FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS OpenCL/cl.hpp DOC "Include for OpenCL CPP bindings on OSX")
-
-ELSE (APPLE)
-
-	IF (WIN32)
-	
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h)
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp)
-	
-	    # The AMD SDK currently installs both x86 and x86_64 libraries
-	    # This is only a hack to find out architecture
-	    IF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86_64")
-	    ELSE (${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64")
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86")
-	    ENDIF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    FIND_LIBRARY(OPENCL_LIBRARIES OpenCL.lib ${OPENCL_LIB_DIR})
-	    
-	    GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-	    
-	    # On Win32 search relative to the library
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS "${_OPENCL_INC_CAND}")
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS "${_OPENCL_INC_CAND}")
-	
-	ELSE (WIN32)
-
-            # Unix style platforms
-            FIND_LIBRARY(OPENCL_LIBRARIES OpenCL
-              ENV LD_LIBRARY_PATH
-            )
-
-            GET_FILENAME_COMPONENT(OPENCL_LIB_DIR ${OPENCL_LIBRARIES} PATH)
-            GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-
-            # The AMD SDK currently does not place its headers
-            # in /usr/include, therefore also search relative
-            # to the library
-            FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-            FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-
-	ENDIF (WIN32)
-
-ENDIF (APPLE)
-
-FIND_PACKAGE_HANDLE_STANDARD_ARGS( OpenCL DEFAULT_MSG OPENCL_LIBRARIES OPENCL_INCLUDE_DIRS )
-
-IF( _OPENCL_CPP_INCLUDE_DIRS )
-	SET( OPENCL_HAS_CPP_BINDINGS TRUE )
-	LIST( APPEND OPENCL_INCLUDE_DIRS ${_OPENCL_CPP_INCLUDE_DIRS} )
-	# This is often the same, so clean up
-	LIST( REMOVE_DUPLICATES OPENCL_INCLUDE_DIRS )
-ENDIF( _OPENCL_CPP_INCLUDE_DIRS )
-
-MARK_AS_ADVANCED(
-  OPENCL_INCLUDE_DIRS
-)
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/cl.hpp b/SpeedComparisons/GrayScott_OpenCL_Image_2x2/cl.hpp
deleted file mode 100644
index 99b86a665..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/cl.hpp
+++ /dev/null
@@ -1,4011 +0,0 @@
-/*******************************************************************************
- * Copyright (c) 2008-2010 The Khronos Group Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and/or associated documentation files (the
- * "Materials"), to deal in the Materials without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sublicense, and/or sell copies of the Materials, and to
- * permit persons to whom the Materials are furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice shall be included
- * in all copies or substantial portions of the Materials.
- *
- * THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
- * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
- * CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
- * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
- * MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
- ******************************************************************************/
-
-/*! \file
- *
- *   \brief C++ bindings for OpenCL 1.0 (rev 48) and OpenCL 1.1 (rev 33)    
- *   \author Benedict R. Gaster and Laurent Morichetti
- *   
- *   Additions and fixes from Brian Cole, March 3rd 2010.
- *   
- *   \version 1.1
- *   \date June 2010
- *
- *   Optional extension support
- *
- *         cl
- *         cl_ext_device_fission
- *				#define USE_CL_DEVICE_FISSION
- */
-
-/*! \mainpage
- * \section intro Introduction
- * For many large applications C++ is the language of choice and so it seems
- * reasonable to define C++ bindings for OpenCL.
- *
- *
- * The interface is contained with a single C++ header file \em cl.hpp and all
- * definitions are contained within the namespace \em cl. There is no additional
- * requirement to include \em cl.h and to use either the C++ or original C
- * bindings it is enough to simply include \em cl.hpp.
- *
- * The bindings themselves are lightweight and correspond closely to the
- * underlying C API. Using the C++ bindings introduces no additional execution
- * overhead.
- *
- * For detail documentation on the bindings see:
- *
- * The OpenCL C++ Wrapper API 1.1 (revision 04)
- *  http://www.khronos.org/registry/cl/specs/opencl-cplusplus-1.1.pdf
- *
- * \section example Example
- *
- * The following example shows a general use case for the C++
- * bindings, including support for the optional exception feature and
- * also the supplied vector and string classes, see following sections for
- * decriptions of these features.
- *
- * \code
- * #define __CL_ENABLE_EXCEPTIONS
- * 
- * #if defined(__APPLE__) || defined(__MACOSX)
- * #include <OpenCL/cl.hpp>
- * #else
- * #include <CL/cl.hpp>
- * #endif
- * #include <cstdio>
- * #include <cstdlib>
- * #include <iostream>
- * 
- *  const char * helloStr  = "__kernel void "
- *                           "hello(void) "
- *                           "{ "
- *                           "  "
- *                           "} ";
- * 
- *  int
- *  main(void)
- *  {
- *     cl_int err = CL_SUCCESS;
- *     try {
- *
- *       std::vector<cl::Platform> platforms;
- *       cl::Platform::get(&platforms);
- *       if (platforms.size() == 0) {
- *           std::cout << "Platform size 0\n";
- *           return -1;
- *       }
- *
- *       cl_context_properties properties[] = 
- *          { CL_CONTEXT_PLATFORM, (cl_context_properties)(platforms[0])(), 0};
- *       cl::Context context(CL_DEVICE_TYPE_CPU, properties); 
- * 
- *       std::vector<cl::Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
- * 
- *       cl::Program::Sources source(1,
- *           std::make_pair(helloStr,strlen(helloStr)));
- *       cl::Program program_ = cl::Program(context, source);
- *       program_.build(devices);
- * 
- *       cl::Kernel kernel(program_, "hello", &err);
- * 
- *       cl::Event event;
- *       cl::CommandQueue queue(context, devices[0], 0, &err);
- *       queue.enqueueNDRangeKernel(
- *           kernel, 
- *           cl::NullRange, 
- *           cl::NDRange(4,4),
- *           cl::NullRange,
- *           NULL,
- *           &event); 
- * 
- *       event.wait();
- *     }
- *     catch (cl::Error err) {
- *        std::cerr 
- *           << "ERROR: "
- *           << err.what()
- *           << "("
- *           << err.err()
- *           << ")"
- *           << std::endl;
- *     }
- * 
- *    return EXIT_SUCCESS;
- *  }
- * 
- * \endcode
- *
- */
-#ifndef CL_HPP_
-#define CL_HPP_
-
-#ifdef _WIN32
-#include <windows.h>
-#include <malloc.h>
-#if defined(USE_DX_INTEROP)
-#include <CL/cl_d3d10.h>
-#endif
-#endif // _WIN32
-
-// 
-#if defined(USE_CL_DEVICE_FISSION)
-#include <CL/cl_ext.h>
-#endif
-
-#if defined(__APPLE__) || defined(__MACOSX)
-#include <OpenGL/OpenGL.h>
-#include <OpenCL/opencl.h>
-#else
-#include <GL/gl.h>
-#include <CL/opencl.h>
-#endif // !__APPLE__
-
-#if !defined(CL_CALLBACK)
-#define CL_CALLBACK
-#endif //CL_CALLBACK
-
-#include <utility>
-
-#if !defined(__NO_STD_VECTOR)
-#include <vector>
-#endif
-
-#if !defined(__NO_STD_STRING)
-#include <string>
-#endif 
-
-#if defined(linux) || defined(__APPLE__) || defined(__MACOSX)
-# include <alloca.h>
-#endif // linux
-
-#include <cstring>
-
-/*! \namespace cl
- *
- * \brief The OpenCL C++ bindings are defined within this namespace.
- *
- */
-namespace cl {
-
-#define __INIT_CL_EXT_FCN_PTR(name) \
-    if(!pfn_##name) { \
-        pfn_##name = (PFN_##name) \
-            clGetExtensionFunctionAddress(#name); \
-        if(!pfn_##name) { \
-        } \
-    }
-
-class Program;
-class Device;
-class Context;
-class CommandQueue;
-class Memory;
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-#include <exception>
-/*! \class Error
- * \brief Exception class
- */
-class Error : public std::exception
-{
-private:
-    cl_int err_;
-    const char * errStr_;
-public:
-    /*! Create a new CL error exception for a given error code
-     *  and corresponding message.
-     */
-    Error(cl_int err, const char * errStr = NULL) : err_(err), errStr_(errStr)
-    {}
-
-    ~Error() throw() {}
-
-    /*! \brief Get error string associated with exception
-     *
-     * \return A memory pointer to the error message string.
-     */
-    virtual const char * what() const throw ()
-    {
-        if (errStr_ == NULL) {
-            return "empty";
-        }
-        else {
-            return errStr_;
-        }
-    }
-
-    /*! \brief Get error code associated with exception
-     *
-     *  \return The error code.
-     */
-    const cl_int err(void) const { return err_; }
-};
-
-#define __ERR_STR(x) #x
-#else
-#define __ERR_STR(x) NULL
-#endif // __CL_ENABLE_EXCEPTIONS
-
-//! \cond DOXYGEN_DETAIL
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#define __GET_DEVICE_INFO_ERR               __ERR_STR(clgetDeviceInfo)
-#define __GET_PLATFORM_INFO_ERR             __ERR_STR(clGetPlatformInfo)
-#define __GET_DEVICE_IDS_ERR                __ERR_STR(clGetDeviceIDs)
-#define __GET_PLATFORM_IDS_ERR              __ERR_STR(clGetPlatformIDs)
-#define __GET_CONTEXT_INFO_ERR              __ERR_STR(clGetContextInfo)
-#define __GET_EVENT_INFO_ERR                __ERR_STR(clGetEventInfo)
-#define __GET_EVENT_PROFILE_INFO_ERR        __ERR_STR(clGetEventProfileInfo)
-#define __GET_MEM_OBJECT_INFO_ERR           __ERR_STR(clGetMemObjectInfo)
-#define __GET_IMAGE_INFO_ERR                __ERR_STR(clGetImageInfo)
-#define __GET_SAMPLER_INFO_ERR              __ERR_STR(clGetSamplerInfo)
-#define __GET_KERNEL_INFO_ERR               __ERR_STR(clGetKernelInfo)
-#define __GET_KERNEL_WORK_GROUP_INFO_ERR    __ERR_STR(clGetKernelWorkGroupInfo)
-#define __GET_PROGRAM_INFO_ERR              __ERR_STR(clGetProgramInfo)
-#define __GET_PROGRAM_BUILD_INFO_ERR        __ERR_STR(clGetProgramBuildInfo)
-#define __GET_COMMAND_QUEUE_INFO_ERR        __ERR_STR(clGetCommandQueueInfo)
-
-#define __CREATE_CONTEXT_FROM_TYPE_ERR      __ERR_STR(clCreateContextFromType)
-#define __GET_SUPPORTED_IMAGE_FORMATS_ERR   __ERR_STR(clGetSupportedImageFormats)
-
-#define __CREATE_BUFFER_ERR                 __ERR_STR(clCreateBuffer)
-#define __CREATE_SUBBUFFER_ERR              __ERR_STR(clCreateSubBuffer)
-#define __CREATE_GL_BUFFER_ERR              __ERR_STR(clCreateFromGLBuffer)
-#define __GET_GL_OBJECT_INFO_ERR            __ERR_STR(clGetGLObjectInfo)
-#define __CREATE_IMAGE2D_ERR                __ERR_STR(clCreateImage2D)
-#define __CREATE_IMAGE3D_ERR                __ERR_STR(clCreateImage3D)
-#define __CREATE_SAMPLER_ERR                __ERR_STR(clCreateSampler)
-#define __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR __ERR_STR(clSetMemObjectDestructorCallback)
-
-#define __CREATE_USER_EVENT_ERR             __ERR_STR(clCreateUserEvent)
-#define __SET_USER_EVENT_STATUS_ERR         __ERR_STR(clSetUserEventStatus)
-#define __SET_EVENT_CALLBACK_ERR            __ERR_STR(clSetEventCallback)
-#define __WAIT_FOR_EVENTS_ERR               __ERR_STR(clWaitForEvents)
-
-#define __CREATE_KERNEL_ERR                 __ERR_STR(clCreateKernel)
-#define __SET_KERNEL_ARGS_ERR               __ERR_STR(clSetKernelArg)
-#define __CREATE_PROGRAM_WITH_SOURCE_ERR    __ERR_STR(clCreateProgramWithSource)
-#define __CREATE_PROGRAM_WITH_BINARY_ERR    __ERR_STR(clCreateProgramWithBinary)
-#define __BUILD_PROGRAM_ERR                 __ERR_STR(clBuildProgram)
-#define __CREATE_KERNELS_IN_PROGRAM_ERR     __ERR_STR(clCreateKernelsInProgram)
-
-#define __CREATE_COMMAND_QUEUE_ERR          __ERR_STR(clCreateCommandQueue)
-#define __SET_COMMAND_QUEUE_PROPERTY_ERR    __ERR_STR(clSetCommandQueueProperty)
-#define __ENQUEUE_READ_BUFFER_ERR           __ERR_STR(clEnqueueReadBuffer)
-#define __ENQUEUE_READ_BUFFER_RECT_ERR      __ERR_STR(clEnqueueReadBufferRect)
-#define __ENQUEUE_WRITE_BUFFER_ERR          __ERR_STR(clEnqueueWriteBuffer)
-#define __ENQUEUE_WRITE_BUFFER_RECT_ERR     __ERR_STR(clEnqueueWriteBufferRect)
-#define __ENQEUE_COPY_BUFFER_ERR            __ERR_STR(clEnqueueCopyBuffer)
-#define __ENQEUE_COPY_BUFFER_RECT_ERR       __ERR_STR(clEnqueueCopyBufferRect)
-#define __ENQUEUE_READ_IMAGE_ERR            __ERR_STR(clEnqueueReadImage)
-#define __ENQUEUE_WRITE_IMAGE_ERR           __ERR_STR(clEnqueueWriteImage)
-#define __ENQUEUE_COPY_IMAGE_ERR            __ERR_STR(clEnqueueCopyImage)
-#define __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR  __ERR_STR(clEnqueueCopyImageToBuffer)
-#define __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR  __ERR_STR(clEnqueueCopyBufferToImage)
-#define __ENQUEUE_MAP_BUFFER_ERR            __ERR_STR(clEnqueueMapBuffer)
-#define __ENQUEUE_MAP_IMAGE_ERR             __ERR_STR(clEnqueueMapImage)
-#define __ENQUEUE_UNMAP_MEM_OBJECT_ERR      __ERR_STR(clEnqueueUnMapMemObject)
-#define __ENQUEUE_NDRANGE_KERNEL_ERR        __ERR_STR(clEnqueueNDRangeKernel)
-#define __ENQUEUE_TASK_ERR                  __ERR_STR(clEnqueueTask)
-#define __ENQUEUE_NATIVE_KERNEL             __ERR_STR(clEnqueueNativeKernel)
-#define __ENQUEUE_MARKER_ERR                __ERR_STR(clEnqueueMarker)
-#define __ENQUEUE_WAIT_FOR_EVENTS_ERR       __ERR_STR(clEnqueueWaitForEvents)
-#define __ENQUEUE_BARRIER_ERR               __ERR_STR(clEnqueueBarrier)
-
-#define __ENQUEUE_ACQUIRE_GL_ERR            __ERR_STR(clEnqueueAcquireGLObjects)
-#define __ENQUEUE_RELEASE_GL_ERR            __ERR_STR(clEnqueueReleaseGLObjects)
-
-#define __UNLOAD_COMPILER_ERR               __ERR_STR(clUnloadCompiler)
-
-#define __FLUSH_ERR                         __ERR_STR(clFlush)
-#define __FINISH_ERR                        __ERR_STR(clFinish)
-
-#define __CREATE_SUB_DEVICES                __ERR_STR(clCreateSubDevicesEXT)
-#endif // __CL_USER_OVERRIDE_ERROR_STRINGS
-//! \endcond
-
-/*! \class string
- * \brief Simple string class, that provides a limited subset of std::string
- * functionality but avoids many of the issues that come with that class.
- */
-class string
-{
-private:
-    ::size_t size_;
-    char * str_;
-public:
-    string(void) : size_(0), str_(NULL)
-    {
-    }
-
-    string(char * str, ::size_t size) :
-        size_(size),
-        str_(NULL)
-    {
-        str_ = new char[size_+1];
-        if (str_ != NULL) {
-            memcpy(str_, str, size_  * sizeof(char));
-            str_[size_] = '\0';
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string(char * str) :
-        str_(NULL)
-    {
-        size_= ::strlen(str);
-        str_ = new char[size_ + 1];
-        if (str_ != NULL) {
-            memcpy(str_, str, (size_ + 1) * sizeof(char));
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string& operator=(const string& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        if (rhs.size_ == 0 || rhs.str_ == NULL) {
-            size_ = 0;
-            str_  = NULL;
-        } 
-        else {
-            size_ = rhs.size_;
-            str_ = new char[size_ + 1];
-            if (str_ != NULL) {
-                memcpy(str_, rhs.str_, (size_ + 1) * sizeof(char));
-            }
-            else {
-                size_ = 0;
-            }
-        }
-
-        return *this;
-    }
-
-    string(const string& rhs)
-    {
-        *this = rhs;
-    }
-
-    ~string()
-    {
-        if (str_ != NULL) {
-            delete[] str_;
-        }
-    }
-
-    ::size_t size(void) const   { return size_; }
-    ::size_t length(void) const { return size(); }
-
-    const char * c_str(void) const { return (str_) ? str_ : "";}
-};
-
-#if !defined(__USE_DEV_STRING) && !defined(__NO_STD_STRING)
-#include <string>
-typedef std::string STRING_CLASS;
-#elif !defined(__USE_DEV_STRING) 
-typedef cl::string STRING_CLASS;
-#endif
-
-#if !defined(__USE_DEV_VECTOR) && !defined(__NO_STD_VECTOR)
-#include <vector>
-#define VECTOR_CLASS std::vector
-#elif !defined(__USE_DEV_VECTOR) 
-#define VECTOR_CLASS cl::vector 
-#endif
-
-#if !defined(__MAX_DEFAULT_VECTOR_SIZE)
-#define __MAX_DEFAULT_VECTOR_SIZE 10
-#endif
-
-/*! \class vector
- * \brief Fixed sized vector implementation that mirroring 
- * std::vector functionality.
- */
-template <typename T, unsigned int N = __MAX_DEFAULT_VECTOR_SIZE>
-class vector
-{
-private:
-    T data_[N];
-    unsigned int size_;
-    bool empty_;
-public:
-    vector() : 
-        size_(-1),
-        empty_(true)
-    {}
-
-    ~vector() {}
-
-    unsigned int size(void) const
-    {
-        return size_ + 1;
-    }
-
-    void clear()
-    {
-        size_ = -1;
-        empty_ = true;
-    }
-
-    void push_back (const T& x)
-    { 
-        if (size() < N) {
-            size_++;  
-            data_[size_] = x;
-            empty_ = false;
-        }
-    }
-
-    void pop_back(void)
-    {
-        if (!empty_) {
-            data_[size_].~T();
-            size_--;
-            if (size_ == -1) {
-                empty_ = true;
-            }
-        }
-    }
-  
-    vector(const vector<T, N>& vec) : 
-        size_(vec.size_),
-        empty_(vec.empty_)
-    {
-        if (!empty_) {
-            memcpy(&data_[0], &vec.data_[0], size() * sizeof(T));
-        }
-    } 
-
-    vector(unsigned int size, const T& val = T()) :
-        size_(-1),
-        empty_(true)
-    {
-        for (unsigned int i = 0; i < size; i++) {
-            push_back(val);
-        }
-    }
-
-    vector<T, N>& operator=(const vector<T, N>& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        size_  = rhs.size_;
-        empty_ = rhs.empty_;
-
-        if (!empty_) {	
-            memcpy(&data_[0], &rhs.data_[0], size() * sizeof(T));
-        }
-    
-        return *this;
-    }
-
-    bool operator==(vector<T,N> &vec)
-    {
-        if (empty_ && vec.empty_) {
-            return true;
-        }
-
-        if (size() != vec.size()) {
-            return false;
-        }
-
-        return memcmp(&data_[0], &vec.data_[0], size() * sizeof(T)) == 0 ? true : false;
-    }
-  
-    operator T* ()             { return data_; }
-    operator const T* () const { return data_; }
-   
-    bool empty (void) const
-    {
-        return empty_;
-    }
-  
-    unsigned int max_size (void) const
-    {
-        return N;
-    }
-
-    unsigned int capacity () const
-    {
-        return sizeof(T) * N;
-    }
-
-    T& operator[](int index)
-    {
-        return data_[index];
-    }
-  
-    T operator[](int index) const
-    {
-        return data_[index];
-    }
-  
-    template<class I>
-    void assign(I start, I end)
-    {
-        clear();   
-        while(start < end) {
-            push_back(*start);
-            start++;
-        }
-    }
-
-    /*! \class iterator
-     * \brief Iterator class for vectors
-     */
-    class iterator
-    {
-    private:
-        vector<T,N> vec_;
-        int index_;
-        bool initialized_;
-    public:
-        iterator(void) : 
-            index_(-1),
-            initialized_(false)
-        {
-            index_ = -1;
-            initialized_ = false;
-        }
-
-        ~iterator(void) {}
-
-        static iterator begin(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = 0;
-            }
-
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-
-        static iterator end(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = vec.size();
-            }
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-    
-        bool operator==(iterator i)
-        {
-            return ((vec_ == i.vec_) && 
-                    (index_ == i.index_) && 
-                    (initialized_ == i.initialized_));
-        }
-
-        bool operator!=(iterator i)
-        {
-            return (!(*this==i));
-        }
-
-        void operator++()
-        {
-            index_++;
-        }
-
-        void operator++(int x)
-        {
-            index_ += x;
-        }
-
-        void operator--()
-        {
-            index_--;
-        }
-
-        void operator--(int x)
-        {
-            index_ -= x;
-        }
-
-        T operator *()
-        {
-            return vec_[index_];
-        }
-    };
-
-    iterator begin(void)
-    {
-        return iterator::begin(*this);
-    }
-
-    iterator end(void)
-    {
-        return iterator::end(*this);
-    }
-
-    T& front(void)
-    {
-        return data_[0];
-    }
-
-    T& back(void)
-    {
-        return data_[size_];
-    }
-
-    const T& front(void) const
-    {
-        return data_[0];
-    }
-
-    const T& back(void) const
-    {
-        return data_[size_];
-    }
-};  
-    
-/*!
- * \brief size_t class used to interface between C++ and
- * OpenCL C calls that require arrays of size_t values, who's
- * size is known statically.
- */
-template <int N>
-struct size_t : public cl::vector< ::size_t, N> { };
-
-namespace detail {
-
-// GetInfo help struct
-template <typename Functor, typename T>
-struct GetInfoHelper
-{
-    static cl_int
-    get(Functor f, cl_uint name, T* param)
-    {
-        return f(name, sizeof(T), param, NULL);
-    }
-};
-
-// Specialized GetInfoHelper for VECTOR_CLASS params
-template <typename Func, typename T>
-struct GetInfoHelper<Func, VECTOR_CLASS<T> >
-{
-    static cl_int get(Func f, cl_uint name, VECTOR_CLASS<T>* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        T* value = (T*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        param->assign(&value[0], &value[required/sizeof(T)]);
-        return CL_SUCCESS;
-    }
-};
-
-// Specialized for getInfo<CL_PROGRAM_BINARIES>
-template <typename Func>
-struct GetInfoHelper<Func, VECTOR_CLASS<char *> >
-{
-    static cl_int
-    get(Func f, cl_uint name, VECTOR_CLASS<char *>* param)
-    {
-      cl_uint err = f(name, param->size() * sizeof(char *), &(*param)[0], NULL);
-      if (err != CL_SUCCESS) {
-        return err;
-      }
-      
-      return CL_SUCCESS;
-    }
-};
-
-// Specialized GetInfoHelper for STRING_CLASS params
-template <typename Func>
-struct GetInfoHelper<Func, STRING_CLASS>
-{
-    static cl_int get(Func f, cl_uint name, STRING_CLASS* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        char* value = (char*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        *param = value;
-        return CL_SUCCESS;
-    }
-};
-
-#define __GET_INFO_HELPER_WITH_RETAIN(CPP_TYPE) \
-namespace detail { \
-template <typename Func> \
-struct GetInfoHelper<Func, CPP_TYPE> \
-{ \
-    static cl_int get(Func f, cl_uint name, CPP_TYPE* param) \
-    { \
-      cl_uint err = f(name, sizeof(CPP_TYPE), param, NULL); \
-      if (err != CL_SUCCESS) { \
-        return err; \
-      } \
-      \
-      return ReferenceHandler<CPP_TYPE::cl_type>::retain((*param)()); \
-    } \
-}; \
-} 
-
-
-#define __PARAM_NAME_INFO_1_0(F) \
-    F(cl_platform_info, CL_PLATFORM_PROFILE, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VERSION, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_NAME, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VENDOR, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_device_info, CL_DEVICE_TYPE, cl_device_type) \
-    F(cl_device_info, CL_DEVICE_VENDOR_ID, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_COMPUTE_UNITS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_CLOCK_FREQUENCY, cl_uint) \
-    F(cl_device_info, CL_DEVICE_ADDRESS_BITS, cl_bitfield) \
-    F(cl_device_info, CL_DEVICE_MAX_READ_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WRITE_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_MEM_ALLOC_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_DEPTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE_SUPPORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_PARAMETER_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_SAMPLERS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MEM_BASE_ADDR_ALIGN, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_SINGLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_TYPE, cl_device_mem_cache_type) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE, cl_uint)\
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_TYPE, cl_device_local_mem_type) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_ERROR_CORRECTION_SUPPORT, cl_bool) \
-    F(cl_device_info, CL_DEVICE_PROFILING_TIMER_RESOLUTION, ::size_t) \
-    F(cl_device_info, CL_DEVICE_ENDIAN_LITTLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_COMPILER_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_EXECUTION_CAPABILITIES, cl_device_exec_capabilities) \
-    F(cl_device_info, CL_DEVICE_QUEUE_PROPERTIES, cl_command_queue_properties) \
-    F(cl_device_info, CL_DEVICE_PLATFORM, cl_platform_id) \
-    F(cl_device_info, CL_DEVICE_NAME, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VENDOR, STRING_CLASS) \
-    F(cl_device_info, CL_DRIVER_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_PROFILE, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_context_info, CL_CONTEXT_REFERENCE_COUNT, cl_uint) \
-    F(cl_context_info, CL_CONTEXT_DEVICES, VECTOR_CLASS<Device>) \
-    F(cl_context_info, CL_CONTEXT_PROPERTIES, VECTOR_CLASS<cl_context_properties>) \
-    \
-    F(cl_event_info, CL_EVENT_COMMAND_QUEUE, cl::CommandQueue) \
-    F(cl_event_info, CL_EVENT_COMMAND_TYPE, cl_command_type) \
-    F(cl_event_info, CL_EVENT_REFERENCE_COUNT, cl_uint) \
-    F(cl_event_info, CL_EVENT_COMMAND_EXECUTION_STATUS, cl_uint) \
-    \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_QUEUED, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_SUBMIT, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_START, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_END, cl_ulong) \
-    \
-    F(cl_mem_info, CL_MEM_TYPE, cl_mem_object_type) \
-    F(cl_mem_info, CL_MEM_FLAGS, cl_mem_flags) \
-    F(cl_mem_info, CL_MEM_SIZE, ::size_t) \
-    F(cl_mem_info, CL_MEM_HOST_PTR, void*) \
-    F(cl_mem_info, CL_MEM_MAP_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_REFERENCE_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_CONTEXT, cl::Context) \
-    \
-    F(cl_image_info, CL_IMAGE_FORMAT, cl_image_format) \
-    F(cl_image_info, CL_IMAGE_ELEMENT_SIZE, ::size_t) \
-    F(cl_image_info, CL_IMAGE_ROW_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_SLICE_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_WIDTH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_HEIGHT, ::size_t) \
-    F(cl_image_info, CL_IMAGE_DEPTH, ::size_t) \
-    \
-    F(cl_sampler_info, CL_SAMPLER_REFERENCE_COUNT, cl_uint) \
-    F(cl_sampler_info, CL_SAMPLER_CONTEXT, cl::Context) \
-    F(cl_sampler_info, CL_SAMPLER_NORMALIZED_COORDS, cl_addressing_mode) \
-    F(cl_sampler_info, CL_SAMPLER_ADDRESSING_MODE, cl_filter_mode) \
-    F(cl_sampler_info, CL_SAMPLER_FILTER_MODE, cl_bool) \
-    \
-    F(cl_program_info, CL_PROGRAM_REFERENCE_COUNT, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_CONTEXT, cl::Context) \
-    F(cl_program_info, CL_PROGRAM_NUM_DEVICES, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_DEVICES, VECTOR_CLASS<cl_device_id>) \
-    F(cl_program_info, CL_PROGRAM_SOURCE, STRING_CLASS) \
-    F(cl_program_info, CL_PROGRAM_BINARY_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_program_info, CL_PROGRAM_BINARIES, VECTOR_CLASS<char *>) \
-    \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_STATUS, cl_build_status) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_OPTIONS, STRING_CLASS) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_LOG, STRING_CLASS) \
-    \
-    F(cl_kernel_info, CL_KERNEL_FUNCTION_NAME, STRING_CLASS) \
-    F(cl_kernel_info, CL_KERNEL_NUM_ARGS, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_REFERENCE_COUNT, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_CONTEXT, cl::Context) \
-    F(cl_kernel_info, CL_KERNEL_PROGRAM, cl::Program) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_COMPILE_WORK_GROUP_SIZE, cl::size_t<3>) \
-    F(cl_kernel_work_group_info, CL_KERNEL_LOCAL_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_command_queue_info, CL_QUEUE_CONTEXT, cl::Context) \
-    F(cl_command_queue_info, CL_QUEUE_DEVICE, cl::Device) \
-    F(cl_command_queue_info, CL_QUEUE_REFERENCE_COUNT, cl_uint) \
-    F(cl_command_queue_info, CL_QUEUE_PROPERTIES, cl_command_queue_properties)
-
-#if defined(CL_VERSION_1_1)
-#define __PARAM_NAME_INFO_1_1(F) \
-    F(cl_context_info, CL_CONTEXT_NUM_DEVICES, cl_uint)\
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_DOUBLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HALF_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HOST_UNIFIED_MEMORY, cl_bool) \
-    \
-    F(cl_mem_info, CL_MEM_ASSOCIATED_MEMOBJECT, cl::Memory) \
-    F(cl_mem_info, CL_MEM_OFFSET, ::size_t) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_PRIVATE_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_event_info, CL_EVENT_CONTEXT, cl::Context)
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-#define __PARAM_NAME_DEVICE_FISSION(F) \
-    F(cl_device_info, CL_DEVICE_PARENT_DEVICE_EXT, cl_device_id) \
-	F(cl_device_info, CL_DEVICE_PARTITION_TYPES_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_AFFINITY_DOMAINS_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_REFERENCE_COUNT_EXT , cl_uint) \
-	F(cl_device_info, CL_DEVICE_PARTITION_STYLE_EXT, VECTOR_CLASS<cl_device_partition_property_ext>)
-#endif // USE_CL_DEVICE_FISSION
-
-template <typename enum_type, cl_int Name>
-struct param_traits {};
-
-#define __DECLARE_PARAM_TRAITS(token, param_name, T) \
-struct token;                                        \
-template<>                                           \
-struct param_traits<detail:: token,param_name>       \
-{                                                    \
-    enum { value = param_name };                     \
-    typedef T param_type;                            \
-};
-
-__PARAM_NAME_INFO_1_0(__DECLARE_PARAM_TRAITS);
-#if defined(CL_VERSION_1_1)
-__PARAM_NAME_INFO_1_1(__DECLARE_PARAM_TRAITS);
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-__PARAM_NAME_DEVICE_FISSION(__DECLARE_PARAM_TRAITS);
-#endif // USE_CL_DEVICE_FISSION
-
-#undef __DECLARE_PARAM_TRAITS
-
-// Convenience functions
-
-template <typename Func, typename T>
-inline cl_int
-getInfo(Func f, cl_uint name, T* param)
-{
-    return GetInfoHelper<Func, T>::get(f, name, param);
-}
-
-template <typename Func, typename Arg0>
-struct GetInfoFunctor0
-{
-    Func f_; const Arg0& arg0_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename Arg1>
-struct GetInfoFunctor1
-{
-    Func f_; const Arg0& arg0_; const Arg1& arg1_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, arg1_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, cl_uint name, T* param)
-{
-    GetInfoFunctor0<Func, Arg0> f0 = { f, arg0 };
-    return GetInfoHelper<GetInfoFunctor0<Func, Arg0>, T>
-        ::get(f0, name, param);
-}
-
-template <typename Func, typename Arg0, typename Arg1, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, const Arg1& arg1, cl_uint name, T* param)
-{
-    GetInfoFunctor1<Func, Arg0, Arg1> f0 = { f, arg0, arg1 };
-    return GetInfoHelper<GetInfoFunctor1<Func, Arg0, Arg1>, T>
-        ::get(f0, name, param);
-}
-
-template<typename T>
-struct ReferenceHandler
-{ };
-
-template <>
-struct ReferenceHandler<cl_device_id>
-{
-    // cl_device_id does not have retain().
-    static cl_int retain(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-    // cl_device_id does not have release().
-    static cl_int release(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-};
-
-template <>
-struct ReferenceHandler<cl_platform_id>
-{
-    // cl_platform_id does not have retain().
-    static cl_int retain(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-    // cl_platform_id does not have release().
-    static cl_int release(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-};
-
-template <>
-struct ReferenceHandler<cl_context>
-{
-    static cl_int retain(cl_context context)
-    { return ::clRetainContext(context); }
-    static cl_int release(cl_context context)
-    { return ::clReleaseContext(context); }
-};
-
-template <>
-struct ReferenceHandler<cl_command_queue>
-{
-    static cl_int retain(cl_command_queue queue)
-    { return ::clRetainCommandQueue(queue); }
-    static cl_int release(cl_command_queue queue)
-    { return ::clReleaseCommandQueue(queue); }
-};
-
-template <>
-struct ReferenceHandler<cl_mem>
-{
-    static cl_int retain(cl_mem memory)
-    { return ::clRetainMemObject(memory); }
-    static cl_int release(cl_mem memory)
-    { return ::clReleaseMemObject(memory); }
-};
-
-template <>
-struct ReferenceHandler<cl_sampler>
-{
-    static cl_int retain(cl_sampler sampler)
-    { return ::clRetainSampler(sampler); }
-    static cl_int release(cl_sampler sampler)
-    { return ::clReleaseSampler(sampler); }
-};
-
-template <>
-struct ReferenceHandler<cl_program>
-{
-    static cl_int retain(cl_program program)
-    { return ::clRetainProgram(program); }
-    static cl_int release(cl_program program)
-    { return ::clReleaseProgram(program); }
-};
-
-template <>
-struct ReferenceHandler<cl_kernel>
-{
-    static cl_int retain(cl_kernel kernel)
-    { return ::clRetainKernel(kernel); }
-    static cl_int release(cl_kernel kernel)
-    { return ::clReleaseKernel(kernel); }
-};
-
-template <>
-struct ReferenceHandler<cl_event>
-{
-    static cl_int retain(cl_event event)
-    { return ::clRetainEvent(event); }
-    static cl_int release(cl_event event)
-    { return ::clReleaseEvent(event); }
-};
-
-template <typename T>
-class Wrapper
-{
-public:
-    typedef T cl_type;
-
-protected:
-    cl_type object_;
-
-public:
-    Wrapper() : object_(NULL) { }
-
-    ~Wrapper()
-    {
-        if (object_ != NULL) { release(); }
-    }
-
-    Wrapper(const Wrapper<cl_type>& rhs)
-    {
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-    }
-
-    Wrapper<cl_type>& operator = (const Wrapper<cl_type>& rhs)
-    {
-        if (object_ != NULL) { release(); }
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-        return *this;
-    }
-
-    cl_type operator ()() const { return object_; }
-
-    cl_type& operator ()() { return object_; }
-
-protected:
-
-    cl_int retain() const
-    {
-        return ReferenceHandler<cl_type>::retain(object_);
-    }
-
-    cl_int release() const
-    {
-        return ReferenceHandler<cl_type>::release(object_);
-    }
-};
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-static inline cl_int errHandler (
-    cl_int err,
-    const char * errStr = NULL) throw(Error)
-{
-    if (err != CL_SUCCESS) {
-        throw Error(err, errStr);
-    }
-    return err;
-}
-#else
-static inline cl_int errHandler (cl_int err, const char * errStr = NULL)
-{
-    return err;
-}
-#endif // __CL_ENABLE_EXCEPTIONS
-
-} // namespace detail
-//! \endcond
-
-/*! \stuct ImageFormat
- * \brief ImageFormat interface fro cl_image_format.
- */
-struct ImageFormat : public cl_image_format
-{
-    ImageFormat(){}
-
-    ImageFormat(cl_channel_order order, cl_channel_type type)
-    {
-        image_channel_order = order;
-        image_channel_data_type = type;
-    }
-
-    ImageFormat& operator = (const ImageFormat& rhs)
-    {
-        if (this != &rhs) {
-            this->image_channel_data_type = rhs.image_channel_data_type;
-            this->image_channel_order     = rhs.image_channel_order;
-        }
-        return *this;
-    }
-};
-
-/*! \class Device
- * \brief Device interface for cl_device_id.
- */
-class Device : public detail::Wrapper<cl_device_id>
-{
-public:
-    Device(cl_device_id device) { object_ = device; }
-
-    Device() : detail::Wrapper<cl_type>() { }
-
-    Device(const Device& device) : detail::Wrapper<cl_type>(device) { }
-
-    Device& operator = (const Device& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_device_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetDeviceInfo, object_, name, param),
-            __GET_DEVICE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_device_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_device_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(USE_CL_DEVICE_FISSION)
-	cl_int createSubDevices(
-		const cl_device_partition_property_ext * properties,
-		VECTOR_CLASS<Device>* devices)
-	{
-		typedef CL_API_ENTRY cl_int 
-			( CL_API_CALL * PFN_clCreateSubDevicesEXT)(
-				cl_device_id /*in_device*/,
-                const cl_device_partition_property_ext * /* properties */,
-                cl_uint /*num_entries*/,
-                cl_device_id * /*out_devices*/,
-                cl_uint * /*num_devices*/ ) CL_EXT_SUFFIX__VERSION_1_1;
-
-		static PFN_clCreateSubDevicesEXT pfn_clCreateSubDevicesEXT = NULL;
-		__INIT_CL_EXT_FCN_PTR(clCreateSubDevicesEXT);
-
-		cl_uint n = 0;
-        cl_int err = pfn_clCreateSubDevicesEXT(object_, properties, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clCreateSubDevicesEXT(object_, properties, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
- 	}
-#endif
-};
-
-/*! \class Platform
- *  \brief Platform interface.
- */
-class Platform : public detail::Wrapper<cl_platform_id>
-{
-public:
-    static const Platform null();
-
-    Platform(cl_platform_id platform) { object_ = platform; }
-
-    Platform() : detail::Wrapper<cl_type>()  { }
-
-    Platform(const Platform& platform) : detail::Wrapper<cl_type>(platform) { }
-
-    Platform& operator = (const Platform& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getInfo(cl_platform_info name, STRING_CLASS* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetPlatformInfo, object_, name, param),
-            __GET_PLATFORM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_platform_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_platform_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getDevices(
-        cl_device_type type,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetDeviceIDs(object_, type, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = ::clGetDeviceIDs(object_, type, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-
-#if defined(USE_DX_INTEROP)
-   /*! \brief Get the list of available D3D10 devices.
-     *
-     *  \param d3d_device_source.
-     *
-     *  \param d3d_object.
-     *
-     *  \param d3d_device_set.
-     *
-     *  \param devices returns a vector of OpenCL D3D10 devices found. The cl::Device
-     *  values returned in devices can be used to identify a specific OpenCL
-     *  device. If \a devices argument is NULL, this argument is ignored.
-     *
-     *  \return One of the following values:
-     *    - CL_SUCCESS if the function is executed successfully.
-     *
-     *  The application can query specific capabilities of the OpenCL device(s)
-     *  returned by cl::getDevices. This can be used by the application to
-     *  determine which device(s) to use.
-     *
-     * \note In the case that exceptions are enabled and a return value
-     * other than CL_SUCCESS is generated, then cl::Error exception is
-     * generated.
-     */
-    cl_int getDevices(
-        cl_d3d10_device_source_khr d3d_device_source,
-        void *                     d3d_object,
-        cl_d3d10_device_set_khr    d3d_device_set,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clGetDeviceIDsFromD3D10KHR)(
-            cl_platform_id platform, 
-            cl_d3d10_device_source_khr d3d_device_source, 
-            void * d3d_object,
-            cl_d3d10_device_set_khr d3d_device_set,
-            cl_uint num_entries,
-            cl_device_id * devices,
-            cl_uint* num_devices);
-
-        static PFN_clGetDeviceIDsFromD3D10KHR pfn_clGetDeviceIDsFromD3D10KHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clGetDeviceIDsFromD3D10KHR);
-
-        cl_uint n = 0;
-        cl_int err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set, 
-            0, 
-            NULL, 
-            &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set,
-            n, 
-            ids, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-#endif
-
-    static cl_int get(
-        VECTOR_CLASS<Platform>* platforms)
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetPlatformIDs(0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        cl_platform_id* ids = (cl_platform_id*) alloca(
-            n * sizeof(cl_platform_id));
-        err = ::clGetPlatformIDs(n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        platforms->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-};
-
-static inline cl_int
-UnloadCompiler()
-{
-    return ::clUnloadCompiler();
-}
-
-class Context : public detail::Wrapper<cl_context>
-{
-public:
-    Context(
-        const VECTOR_CLASS<Device>& devices,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContext(
-            properties, (cl_uint) devices.size(),
-            (cl_device_id*) &devices.front(),
-            notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context(
-        cl_device_type type,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContextFromType(
-            properties, type, notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context() : detail::Wrapper<cl_type>() { }
-
-    Context(const Context& context) : detail::Wrapper<cl_type>(context) { }
-
-    Context& operator = (const Context& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_context_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetContextInfo, object_, name, param),
-            __GET_CONTEXT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_context_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_context_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getSupportedImageFormats(
-        cl_mem_flags flags,
-        cl_mem_object_type type,
-        VECTOR_CLASS<ImageFormat>* formats) const
-    {
-        cl_uint numEntries;
-        cl_int err = ::clGetSupportedImageFormats(
-           object_, 
-           flags,
-           type, 
-           0, 
-           NULL, 
-           &numEntries);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        ImageFormat* value = (ImageFormat*)
-            alloca(numEntries * sizeof(ImageFormat));
-        err = ::clGetSupportedImageFormats(
-            object_, 
-            flags, 
-            type, 
-            numEntries,
-            (cl_image_format*) value, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        formats->assign(&value[0], &value[numEntries]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Context)
-
-/*! \class Event
- * \brief Event interface for cl_event.
- */
-class Event : public detail::Wrapper<cl_event>
-{
-public:
-    Event() : detail::Wrapper<cl_type>() { }
-
-    Event(const Event& event) : detail::Wrapper<cl_type>(event) { }
-
-    Event& operator = (const Event& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_event_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetEventInfo, object_, name, param),
-            __GET_EVENT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_event_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_event_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getProfilingInfo(cl_profiling_info name, T* param) const
-    {
-        return detail::errHandler(detail::getInfo(
-            &::clGetEventProfilingInfo, object_, name, param),
-            __GET_EVENT_PROFILE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_profiling_info, name>::param_type
-    getProfilingInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_profiling_info, name>::param_type param;
-        cl_int result = getProfilingInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int wait() const
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(1, &object_),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setCallback(
-        cl_int type,
-        void (CL_CALLBACK * pfn_notify)(cl_event, cl_int, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetEventCallback(
-                object_,
-                type,
-                pfn_notify,
-                user_data), 
-            __SET_EVENT_CALLBACK_ERR);
-    }
-#endif
-
-    static cl_int
-    waitForEvents(const VECTOR_CLASS<Event>& events)
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(
-                (cl_uint) events.size(), (cl_event*)&events.front()),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Event)
-
-#if defined(CL_VERSION_1_1)
-/*! \class UserEvent
- * \brief User event interface for cl_event.
- */
-class UserEvent : public Event
-{
-public:
-    UserEvent(
-        const Context& context,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateUserEvent(
-            context(),
-            &error);
-
-        detail::errHandler(error, __CREATE_USER_EVENT_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    UserEvent() : Event() { }
-
-    UserEvent(const UserEvent& event) : Event(event) { }
-
-    UserEvent& operator = (const UserEvent& rhs)
-    {
-        if (this != &rhs) {
-            Event::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int setStatus(cl_int status)
-    {
-        return detail::errHandler(
-            ::clSetUserEventStatus(object_,status), 
-            __SET_USER_EVENT_STATUS_ERR);
-    }
-};
-#endif
-
-inline static cl_int
-WaitForEvents(const VECTOR_CLASS<Event>& events)
-{
-    return detail::errHandler(
-        ::clWaitForEvents(
-            (cl_uint) events.size(), (cl_event*)&events.front()),
-        __WAIT_FOR_EVENTS_ERR);
-}
-
-/*! \class Memory
- * \brief Memory interface for cl_mem.
- */
-class Memory : public detail::Wrapper<cl_mem>
-{
-public:
-    Memory() : detail::Wrapper<cl_type>() { }
-
-    Memory(const Memory& memory) : detail::Wrapper<cl_type>(memory) { }
-
-    Memory& operator = (const Memory& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_mem_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetMemObjectInfo, object_, name, param),
-            __GET_MEM_OBJECT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_mem_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_mem_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setDestructorCallback(
-        void (CL_CALLBACK * pfn_notify)(cl_mem, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetMemObjectDestructorCallback(
-                object_,
-                pfn_notify,
-                user_data), 
-            __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR);
-    }
-#endif
-
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Memory)
-
-/*! \class Buffer
- * \brief Memory buffer interface.
- */
-class Buffer : public Memory
-{
-public:
-    Buffer(
-        const Context& context,
-        cl_mem_flags flags,
-        ::size_t size,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateBuffer(context(), flags, size, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Buffer() : Memory() { }
-
-    Buffer(const Buffer& buffer) : Memory(buffer) { }
-
-    Buffer& operator = (const Buffer& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-
-#if defined(CL_VERSION_1_1)
-    Buffer createSubBuffer(
-        cl_mem_flags flags,
-        cl_buffer_create_type buffer_create_type,
-        const void * buffer_create_info,
-        cl_int * err = NULL)
-    {
-        Buffer result;
-        cl_int error;
-        result.object_ = ::clCreateSubBuffer(
-            object_, 
-            flags, 
-            buffer_create_type, 
-            buffer_create_info, 
-            &error);
-
-        detail::errHandler(error, __CREATE_SUBBUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-
-        return result;
-	}		
-#endif
-};
-
-#if defined (USE_DX_INTEROP)
-class BufferD3D10 : public Buffer
-{
-public:
-    typedef CL_API_ENTRY cl_mem (CL_API_CALL *PFN_clCreateFromD3D10BufferKHR)(
-    cl_context context, cl_mem_flags flags, ID3D10Buffer*  buffer,
-    cl_int* errcode_ret);
-
-    BufferD3D10(
-        const Context& context,
-        cl_mem_flags flags,
-        ID3D10Buffer* bufobj,
-        cl_int * err = NULL)
-    {
-        static PFN_clCreateFromD3D10BufferKHR pfn_clCreateFromD3D10BufferKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clCreateFromD3D10BufferKHR);
-
-        cl_int error;
-        object_ = pfn_clCreateFromD3D10BufferKHR(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferD3D10() : Buffer() { }
-
-    BufferD3D10(const BufferD3D10& buffer) : Buffer(buffer) { }
-
-    BufferD3D10& operator = (const BufferD3D10& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-};
-#endif
-
-/*! \class BufferGL
- * \brief Memory buffer interface for GL interop.
- */
-class BufferGL : public Buffer
-{
-public:
-    BufferGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLBuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferGL() : Buffer() { }
-
-    BufferGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferGL& operator = (const BufferGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class BufferRenderGL
- * \brief Memory buffer interface for GL interop with renderbuffer.
- */
-class BufferRenderGL : public Buffer
-{
-public:
-    BufferRenderGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLRenderbuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferRenderGL() : Buffer() { }
-
-    BufferRenderGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferRenderGL& operator = (const BufferRenderGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class Image
- * \brief Base class  interface for all images.
- */
-class Image : public Memory
-{
-protected:
-    Image() : Memory() { }
-
-    Image(const Image& image) : Memory(image) { }
-
-    Image& operator = (const Image& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-public:
-    template <typename T>
-    cl_int getImageInfo(cl_image_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetImageInfo, object_, name, param),
-            __GET_IMAGE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_image_info, name>::param_type
-    getImageInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_image_info, name>::param_type param;
-        cl_int result = getImageInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-/*! \class Image2D
- * \brief Image interface for 2D images.
- */
-class Image2D : public Image
-{
-public:
-    Image2D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t row_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage2D(
-            context(), flags,&format, width, height, row_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE2D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2D() { }
-
-    Image2D(const Image2D& image2D) : Image(image2D) { }
-
-    Image2D& operator = (const Image2D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image2DGL : public Image2D
-{
-public:
-    Image2DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture2D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2DGL() : Image2D() { }
-
-    Image2DGL(const Image2DGL& image) : Image2D(image) { }
-
-    Image2DGL& operator = (const Image2DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image2D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image3D
- * \brief Image interface for 3D images.
- */
-class Image3D : public Image
-{
-public:
-    Image3D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t depth,
-        ::size_t row_pitch = 0,
-        ::size_t slice_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage3D(
-            context(), flags, &format, width, height, depth, row_pitch,
-            slice_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE3D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3D() { }
-
-    Image3D(const Image3D& image3D) : Image(image3D) { }
-
-    Image3D& operator = (const Image3D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image3DGL : public Image3D
-{
-public:
-    Image3DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture3D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3DGL() : Image3D() { }
-
-    Image3DGL(const Image3DGL& image) : Image3D(image) { }
-
-    Image3DGL& operator = (const Image3DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image3D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Sampler
- * \brief Sampler interface for cl_sampler.
- */
-class Sampler : public detail::Wrapper<cl_sampler>
-{
-public:
-    Sampler() { }
-
-    Sampler(
-        const Context& context,
-        cl_bool normalized_coords,
-        cl_addressing_mode addressing_mode,
-        cl_filter_mode filter_mode,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateSampler(
-            context(), 
-            normalized_coords,
-            addressing_mode,
-            filter_mode,
-            &error);
-
-        detail::errHandler(error, __CREATE_SAMPLER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Sampler(const Sampler& sampler) : detail::Wrapper<cl_type>(sampler) { }
-
-    Sampler& operator = (const Sampler& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_sampler_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetSamplerInfo, object_, name, param),
-            __GET_SAMPLER_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_sampler_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_sampler_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Sampler)
-
-class Program;
-class CommandQueue;
-class Kernel;
-
-/*! \class NDRange
- * \brief NDRange interface
- */
-class NDRange
-{
-private:
-    size_t<3> sizes_;
-    cl_uint dimensions_;
-
-public:
-    NDRange()
-        : dimensions_(0)
-    { }
-
-    NDRange(::size_t size0)
-        : dimensions_(1)
-    {
-        sizes_.push_back(size0);
-    }
-
-    NDRange(::size_t size0, ::size_t size1)
-        : dimensions_(2)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-    }
-
-    NDRange(::size_t size0, ::size_t size1, ::size_t size2)
-        : dimensions_(3)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-        sizes_.push_back(size2);
-    }
-
-    operator const ::size_t*() const { return (const ::size_t*) sizes_; }
-    ::size_t dimensions() const { return dimensions_; }
-};
-
-static const NDRange NullRange;
-
-/*!
- * \struct LocalSpaceArg
- * \brief Local address raper for use with Kernel::setArg
- */
-struct LocalSpaceArg
-{
-    ::size_t size_;
-};
-
-namespace detail {
-
-template <typename T>
-struct KernelArgumentHandler
-{
-    static ::size_t size(const T&) { return sizeof(T); }
-    static T* ptr(T& value) { return &value; }
-};
-
-template <>
-struct KernelArgumentHandler<LocalSpaceArg>
-{
-    static ::size_t size(const LocalSpaceArg& value) { return value.size_; }
-    static void* ptr(LocalSpaceArg&) { return NULL; }
-};
-
-} 
-//! \endcond
-
-inline LocalSpaceArg
-__local(::size_t size)
-{
-    LocalSpaceArg ret = { size };
-    return ret;
-}
-
-class KernelFunctor;
-
-/*! \class Kernel
- * \brief Kernel interface that implements cl_kernel
- */
-class Kernel : public detail::Wrapper<cl_kernel>
-{
-public:
-    inline Kernel(const Program& program, const char* name, cl_int* err = NULL);
-
-    Kernel() { }
-
-    Kernel(const Kernel& kernel) : detail::Wrapper<cl_type>(kernel) { }
-
-    Kernel& operator = (const Kernel& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_kernel_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetKernelInfo, object_, name, param),
-            __GET_KERNEL_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_kernel_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getWorkGroupInfo(
-        const Device& device, cl_kernel_work_group_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetKernelWorkGroupInfo, object_, device(), name, param),
-                __GET_KERNEL_WORK_GROUP_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_work_group_info, name>::param_type
-        getWorkGroupInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-        detail::cl_kernel_work_group_info, name>::param_type param;
-        cl_int result = getWorkGroupInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int setArg(cl_uint index, T value)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(
-                object_,
-                index,
-                detail::KernelArgumentHandler<T>::size(value),
-                detail::KernelArgumentHandler<T>::ptr(value)),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    cl_int setArg(cl_uint index, ::size_t size, void* argPtr)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(object_, index, size, argPtr),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local);
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& global,
-        const NDRange& local);
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Kernel)
-
-/*! \class Program
- * \brief Program interface that implements cl_program.
- */
-class Program : public detail::Wrapper<cl_program>
-{
-public:
-    typedef VECTOR_CLASS<std::pair<const void*, ::size_t> > Binaries;
-    typedef VECTOR_CLASS<std::pair<const char*, ::size_t> > Sources;
-
-    Program(
-        const Context& context,
-        const Sources& sources,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-
-        const ::size_t n = (::size_t)sources.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const char** strings = (const char**) alloca(n * sizeof(const char*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            strings[i] = sources[(int)i].first;
-            lengths[i] = sources[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithSource(
-            context(), (cl_uint)n, strings, lengths, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_SOURCE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program(
-        const Context& context,
-        const VECTOR_CLASS<Device>& devices,
-        const Binaries& binaries,
-        VECTOR_CLASS<cl_int>* binaryStatus = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        const ::size_t n = binaries.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const unsigned char** images = (const unsigned char**) alloca(n * sizeof(const void*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            images[i] = (const unsigned char*)binaries[(int)i].first;
-            lengths[i] = binaries[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithBinary(
-            context(), (cl_uint) devices.size(),
-            (cl_device_id*)&devices.front(),
-            lengths, images, binaryStatus != NULL
-               ? (cl_int*) &binaryStatus->front()
-               : NULL, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_BINARY_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program() { }
-
-    Program(const Program& program) : detail::Wrapper<cl_type>(program) { }
-
-    Program& operator = (const Program& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int build(
-        const VECTOR_CLASS<Device>& devices,
-        const char* options = NULL,
-        void (CL_CALLBACK * notifyFptr)(cl_program, void *) = NULL,
-        void* data = NULL) const
-    {
-        return detail::errHandler(
-            ::clBuildProgram(
-                object_,
-                (cl_uint)
-                devices.size(),
-                (cl_device_id*)&devices.front(),
-                options,
-                notifyFptr,
-                data),
-                __BUILD_PROGRAM_ERR);
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_program_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetProgramInfo, object_, name, param),
-            __GET_PROGRAM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getBuildInfo(
-        const Device& device, cl_program_build_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetProgramBuildInfo, object_, device(), name, param),
-                __GET_PROGRAM_BUILD_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_build_info, name>::param_type
-    getBuildInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_build_info, name>::param_type param;
-        cl_int result = getBuildInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int createKernels(VECTOR_CLASS<Kernel>* kernels)
-    {
-        cl_uint numKernels;
-        cl_int err = ::clCreateKernelsInProgram(object_, 0, NULL, &numKernels);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        Kernel* value = (Kernel*) alloca(numKernels * sizeof(Kernel));
-        err = ::clCreateKernelsInProgram(
-            object_, numKernels, (cl_kernel*) value, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        kernels->assign(&value[0], &value[numKernels]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Program)
-
-inline Kernel::Kernel(const Program& program, const char* name, cl_int* err)
-{
-    cl_int error;
-
-    object_ = ::clCreateKernel(program(), name, &error);
-    detail::errHandler(error, __CREATE_KERNEL_ERR);
-
-    if (err != NULL) {
-        *err = error;
-    }
-
-}
-
-/*! \class CommandQueue
- * \brief CommandQueue interface for cl_command_queue.
- */
-class CommandQueue : public detail::Wrapper<cl_command_queue>
-{
-public:
-    CommandQueue(
-        const Context& context,
-        const Device& device,
-        cl_command_queue_properties properties = 0,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateCommandQueue(
-            context(), device(), properties, &error);
-
-        detail::errHandler(error, __CREATE_COMMAND_QUEUE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    CommandQueue() { }
-
-    CommandQueue(const CommandQueue& commandQueue) : detail::Wrapper<cl_type>(commandQueue) { }
-
-    CommandQueue& operator = (const CommandQueue& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_command_queue_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetCommandQueueInfo, object_, name, param),
-                __GET_COMMAND_QUEUE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_command_queue_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_command_queue_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int enqueueReadBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_BUFFER_ERR);
-    }
-
-    cl_int enqueueWriteBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        const void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBuffer(
-        const Buffer& src,
-        const Buffer& dst,
-        ::size_t src_offset,
-        ::size_t dst_offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBuffer(
-                object_, src(), dst(), src_offset, dst_offset, size,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int enqueueReadBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_READ_BUFFER_RECT_ERR);
-    }
-
-
-    cl_int enqueueWriteBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_RECT_ERR);
-    }
-
-    cl_int enqueueCopyBufferRect(
-        const Buffer& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        ::size_t src_row_pitch,
-        ::size_t src_slice_pitch,
-        ::size_t dst_row_pitch,
-        ::size_t dst_slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferRect(
-                object_, 
-                src(), 
-                dst(), 
-                (const ::size_t *)src_origin, 
-                (const ::size_t *)dst_origin, 
-                (const ::size_t *)region,
-                src_row_pitch,
-                src_slice_pitch,
-                dst_row_pitch,
-                dst_slice_pitch,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_RECT_ERR);
-    }
-#endif
-
-    cl_int enqueueReadImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_IMAGE_ERR);
-    }
-
-    cl_int enqueueWriteImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_WRITE_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImage(
-        const Image& src,
-        const Image& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImage(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *)dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImageToBuffer(
-        const Image& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& region,
-        ::size_t dst_offset,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImageToBuffer(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *) region, dst_offset,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBufferToImage(
-        const Buffer& src,
-        const Image& dst,
-        ::size_t src_offset,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferToImage(
-                object_, src(), dst(), src_offset,
-                (const ::size_t *) dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR);
-    }
-
-    void* enqueueMapBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        ::size_t offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapBuffer(
-            object_, buffer(), blocking, flags, offset, size,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-        return result;
-    }
-
-    void* enqueueMapImage(
-        const Image& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t * row_pitch,
-        ::size_t * slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapImage(
-            object_, buffer(), blocking, flags,
-            (const ::size_t *) origin, (const ::size_t *) region,
-            row_pitch, slice_pitch,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_IMAGE_ERR);
-        if (err != NULL) {
-              *err = error;
-        }
-        return result;
-    }
-
-    cl_int enqueueUnmapMemObject(
-        const Memory& memory,
-        void* mapped_ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueUnmapMemObject(
-                object_, memory(), mapped_ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_UNMAP_MEM_OBJECT_ERR);
-    }
-
-    cl_int enqueueNDRangeKernel(
-        const Kernel& kernel,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueNDRangeKernel(
-                object_, kernel(), (cl_uint) global.dimensions(),
-                offset.dimensions() != 0 ? (const ::size_t*) offset : NULL,
-                (const ::size_t*) global,
-                local.dimensions() != 0 ? (const ::size_t*) local : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NDRANGE_KERNEL_ERR);
-    }
-
-    cl_int enqueueTask(
-        const Kernel& kernel,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueTask(
-                object_, kernel(),
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_TASK_ERR);
-    }
-
-    cl_int enqueueNativeKernel(
-        void (*userFptr)(void *),
-        std::pair<void*, ::size_t> args,
-        const VECTOR_CLASS<Memory>* mem_objects = NULL,
-        const VECTOR_CLASS<const void*>* mem_locs = NULL,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        cl_mem * mems = (mem_objects != NULL && mem_objects->size() > 0) 
-            ? (cl_mem*) alloca(mem_objects->size() * sizeof(cl_mem))
-            : NULL;
-
-        if (mems != NULL) {
-            for (unsigned int i = 0; i < mem_objects->size(); i++) {
-                mems[i] = ((*mem_objects)[i])();
-            }
-        }
-
-        return detail::errHandler(
-            ::clEnqueueNativeKernel(
-                object_, userFptr, args.first, args.second,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                mems,
-                (mem_locs != NULL) ? (const void **) &mem_locs->front() : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NATIVE_KERNEL);
-    }
-
-    cl_int enqueueMarker(Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueMarker(object_, (cl_event*) event),
-            __ENQUEUE_MARKER_ERR);
-    }
-
-    cl_int enqueueWaitForEvents(const VECTOR_CLASS<Event>& events) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWaitForEvents(
-                object_,
-                (cl_uint) events.size(),
-                (const cl_event*) &events.front()),
-            __ENQUEUE_WAIT_FOR_EVENTS_ERR);
-    }
-
-    cl_int enqueueAcquireGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueAcquireGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueReleaseGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_RELEASE_GL_ERR);
-     }
-
-#if defined (USE_DX_INTEROP)
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueAcquireD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects, cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueReleaseD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects,  cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-
-    cl_int enqueueAcquireD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         static PFN_clEnqueueAcquireD3D10ObjectsKHR pfn_clEnqueueAcquireD3D10ObjectsKHR = NULL;
-         __INIT_CL_EXT_FCN_PTR(clEnqueueAcquireD3D10ObjectsKHR);
-		
-         return detail::errHandler(
-             pfn_clEnqueueAcquireD3D10ObjectsKHR(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-    {
-        static PFN_clEnqueueReleaseD3D10ObjectsKHR pfn_clEnqueueReleaseD3D10ObjectsKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clEnqueueReleaseD3D10ObjectsKHR);
-
-        return detail::errHandler(
-            pfn_clEnqueueReleaseD3D10ObjectsKHR(
-                object_,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_RELEASE_GL_ERR);
-    }
-#endif
-
-    cl_int enqueueBarrier() const
-    {
-        return detail::errHandler(
-            ::clEnqueueBarrier(object_),
-            __ENQUEUE_BARRIER_ERR);
-    }
-
-    cl_int flush() const
-    {
-        return detail::errHandler(::clFlush(object_), __FLUSH_ERR);
-    }
-
-    cl_int finish() const
-    {
-        return detail::errHandler(::clFinish(object_), __FINISH_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::CommandQueue)
-
-/*! \class KernelFunctor
- * \brief Kernel functor interface
- *
- * \note Currently only functors of zero to ten arguments are supported. It
- * is straightforward to add more and a more general solution, similar to
- * Boost.Lambda could be followed if required in the future.
- */
-class KernelFunctor
-{
-private:
-    Kernel kernel_;
-    CommandQueue queue_;
-    NDRange offset_;
-    NDRange global_;
-    NDRange local_;
-
-    cl_int err_;
-public:
-    KernelFunctor() { }
-
-    KernelFunctor(
-        const Kernel& kernel,
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local) :
-            kernel_(kernel),
-            queue_(queue),
-            offset_(offset),
-            global_(global),
-            local_(local),
-            err_(CL_SUCCESS)
-    {}
-
-    KernelFunctor& operator=(const KernelFunctor& rhs);
-
-    KernelFunctor(const KernelFunctor& rhs);
-
-    cl_int getError() { return err_; }
-
-    inline Event operator()(const VECTOR_CLASS<Event>* events = NULL);
-
-    template<typename A1>
-    inline Event operator()(
-        const A1& a1, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5, class A6>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4,
-             class A5, class A6, class A7>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12, 
-        const A13& a13,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14, class A15>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14, 
-        const A15& a15,
-        const VECTOR_CLASS<Event>* events = NULL);
-};
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& offset,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,offset,global,local);
-}
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,NullRange,global,local);
-}
-
-inline KernelFunctor& KernelFunctor::operator=(const KernelFunctor& rhs)
-{
-    if (this == &rhs) {
-        return *this;
-    }
-    
-    kernel_ = rhs.kernel_;
-    queue_  = rhs.queue_;
-    offset_ = rhs.offset_;
-    global_ = rhs.global_;
-    local_  = rhs.local_;
-    
-    return *this;
-}
-
-inline KernelFunctor::KernelFunctor(const KernelFunctor& rhs) :
-    kernel_(rhs.kernel_),
-    queue_(rhs.queue_),
-    offset_(rhs.offset_),
-    global_(rhs.global_),
-    local_(rhs.local_)
-{
-}
-
-Event KernelFunctor::operator()(const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4,
-         typename A5, typename A6, typename A7>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9, typename A10>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12, 
-    const A13& a13,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14, class A15>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14, 
-    const A15& a15,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-    kernel_.setArg(14,a15);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-#undef __ERR_STR
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#undef __GET_DEVICE_INFO_ERR
-#undef __GET_PLATFORM_INFO_ERR
-#undef __GET_DEVICE_IDS_ERR
-#undef __GET_CONTEXT_INFO_ERR
-#undef __GET_EVENT_INFO_ERR
-#undef __GET_EVENT_PROFILE_INFO_ERR
-#undef __GET_MEM_OBJECT_INFO_ERR
-#undef __GET_IMAGE_INFO_ERR
-#undef __GET_SAMPLER_INFO_ERR
-#undef __GET_KERNEL_INFO_ERR
-#undef __GET_KERNEL_WORK_GROUP_INFO_ERR
-#undef __GET_PROGRAM_INFO_ERR
-#undef __GET_PROGRAM_BUILD_INFO_ERR
-#undef __GET_COMMAND_QUEUE_INFO_ERR
-
-#undef __CREATE_CONTEXT_FROM_TYPE_ERR
-#undef __GET_SUPPORTED_IMAGE_FORMATS_ERR
-
-#undef __CREATE_BUFFER_ERR
-#undef __CREATE_SUBBUFFER_ERR
-#undef __CREATE_IMAGE2D_ERR
-#undef __CREATE_IMAGE3D_ERR
-#undef __CREATE_SAMPLER_ERR
-#undef __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR
-
-#undef __CREATE_USER_EVENT_ERR
-#undef __SET_USER_EVENT_STATUS_ERR
-#undef __SET_EVENT_CALLBACK_ERR
-
-#undef __WAIT_FOR_EVENTS_ERR
-
-#undef __CREATE_KERNEL_ERR
-#undef __SET_KERNEL_ARGS_ERR
-#undef __CREATE_PROGRAM_WITH_SOURCE_ERR
-#undef __CREATE_PROGRAM_WITH_BINARY_ERR
-#undef __BUILD_PROGRAM_ERR
-#undef __CREATE_KERNELS_IN_PROGRAM_ERR
-
-#undef __CREATE_COMMAND_QUEUE_ERR
-#undef __SET_COMMAND_QUEUE_PROPERTY_ERR
-#undef __ENQUEUE_READ_BUFFER_ERR
-#undef __ENQUEUE_WRITE_BUFFER_ERR
-#undef __ENQUEUE_READ_BUFFER_RECT_ERR
-#undef __ENQUEUE_WRITE_BUFFER_RECT_ERR
-#undef __ENQEUE_COPY_BUFFER_ERR
-#undef __ENQEUE_COPY_BUFFER_RECT_ERR
-#undef __ENQUEUE_READ_IMAGE_ERR
-#undef __ENQUEUE_WRITE_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR
-#undef __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR
-#undef __ENQUEUE_MAP_BUFFER_ERR
-#undef __ENQUEUE_MAP_IMAGE_ERR
-#undef __ENQUEUE_UNMAP_MEM_OBJECT_ERR
-#undef __ENQUEUE_NDRANGE_KERNEL_ERR
-#undef __ENQUEUE_TASK_ERR
-#undef __ENQUEUE_NATIVE_KERNEL
-
-#undef __UNLOAD_COMPILER_ERR
-#endif //__CL_USER_OVERRIDE_ERROR_STRINGS
-
-#undef __GET_INFO_HELPER_WITH_RETAIN
-
-// Extensions
-#undef __INIT_CL_EXT_FCN_PTR
-#undef __CREATE_SUB_DEVICES
-
-#if defined(USE_CL_DEVICE_FISSION)
-#undef __PARAM_NAME_DEVICE_FISSION
-#endif // USE_CL_DEVICE_FISSION
-
-} // namespace cl
-
-#endif // CL_HPP_
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/gray_scott_opencl_image_2x2.cpp b/SpeedComparisons/GrayScott_OpenCL_Image_2x2/gray_scott_opencl_image_2x2.cpp
deleted file mode 100644
index 5e609f40e..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/gray_scott_opencl_image_2x2.cpp
+++ /dev/null
@@ -1,365 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCL:
-#define __NO_STD_VECTOR // Use cl::vector instead of STL version
-#define __CL_ENABLE_EXCEPTIONS
-
-// cl.hpp is standard but doesn't come with most SDKs, so download it from here:
-// http://www.khronos.org/registry/cl/api/1.1/cl.hpp
-#ifdef __APPLE__
-# include "cl.hpp"
-#else
-# include <CL/cl.hpp>
-#endif
-
-using namespace cl;
-
-// STL:
-#include <fstream>
-#include <iostream>
-
-// local:
-#include "defs.h"
-
-void init(float *a,float *b);
-bool display(float *a,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
-
-// we pack the values in 2x2 blocks:   x y
-//                                     z w
-float& float_at(float* arr,int x,int y) 
-{
-    return arr[ ( (y/2)*(X/2) + x/2 ) * 4 + (y%2)*2 + x%2 ]; 
-}
-
-static int g_opt_device = 0;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if ((i+1<argc) && (strcmp(argv[i],"-device")==0)) {
-            // select an output device
-            i++; g_opt_device = atoi(argv[i]);
-        } else {
-            std::cout << "Unrecognized argument: '" << argv[i] << "'\n";
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    float k, f;
-    k = 0.064f; f = 0.035f; // solitons with mitosis (spots that multiply)
-    // k = 0.06f; f = 0.035f; // stripes
-    //k = 0.065f; f = 0.056f; // long stripes
-    //k = 0.064f; f = 0.04f; // dots and stripes
-    // k = 0.0475f; f = 0.0118f; // spiral waves
-    float speed = 2.0f;
-    // ----------------
-    
-    try { 
-        // Get available OpenCL platforms
-        cl::vector<Platform> platforms;
-        Platform::get(&platforms);
- 
-        // Select the default platform and create a context using this platform and the GPU
-        cl_context_properties cps[3] = { 
-            CL_CONTEXT_PLATFORM, 
-            (cl_context_properties)(platforms[0])(), 
-            0 
-        };
-        Context context( CL_DEVICE_TYPE_GPU, cps);
-
-        float *a = new float[X*Y];
-        float *b = new float[X*Y];
-        init(a,b);
-
-        // Get a list of devices on this platform
-        cl::vector<Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
-
-        // range-check the user's selection
-        int maxdev = devices.size() - 1;
-        g_opt_device = (g_opt_device > maxdev) ? maxdev :
-                                     ((g_opt_device < 0) ? 0 : g_opt_device);
-        std::cout << (maxdev+1) << " device(s) available; using device "
-                                                    << g_opt_device << ".\n";
-
-        Device &device = devices[g_opt_device];
-        cl::vector<Device> ourdevices = cl::vector<Device>(1, device);
-
-        bool is_ImageSupported = device.getInfo<CL_DEVICE_IMAGE_SUPPORT>();
-        if(!is_ImageSupported)
-        {
-            printf("Images not supported on this device.\n");
-            exit(-1);
-        }
-
-        // we make two images and swap between them
-        cl::Image2D a1(context,CL_MEM_READ_WRITE,cl::ImageFormat(CL_RGBA,CL_FLOAT),X/2,Y/2);
-        cl::Image2D b1(context,CL_MEM_READ_WRITE,cl::ImageFormat(CL_RGBA,CL_FLOAT),X/2,Y/2);
-        cl::Image2D a2(context,CL_MEM_READ_WRITE,cl::ImageFormat(CL_RGBA,CL_FLOAT),X/2,Y/2);
-        cl::Image2D b2(context,CL_MEM_READ_WRITE,cl::ImageFormat(CL_RGBA,CL_FLOAT),X/2,Y/2);
-
-        // Create a command queue and use the selected device
-        CommandQueue queue = CommandQueue(context, device);
-        Event event;
- 
-        // Copy to the memory buffers
-        cl::size_t<3> origin;
-        origin.push_back(0);
-        origin.push_back(0);
-        origin.push_back(0);
-        cl::size_t<3> region;
-        region.push_back(X/2);
-        region.push_back(Y/2);
-        region.push_back(1);
-        queue.enqueueWriteImage(a1,true,origin,region,0,0,a);
-        queue.enqueueWriteImage(b1,true,origin,region,0,0,b);
- 
-        // Read source file
-        std::string kfn = CL_SOURCE_DIR; // (defined in CMakeLists.txt to be the source folder)
-        kfn += "/grayscott_kernel_image_2x2.cl";
-        std::ifstream sourceFile(kfn.c_str());
-        std::string sourceCode(
-            std::istreambuf_iterator<char>(sourceFile),
-            (std::istreambuf_iterator<char>()));
-        Program::Sources source(1, std::make_pair(sourceCode.c_str(), sourceCode.length()+1));
- 
-        // enable this code to display kernel compilation error if you get clBuildProgram(-11)
-        #if 0
-           const ::size_t n = (::size_t)source.size();
-           ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-           const char** strings = (const char**) alloca(n * sizeof(const char*));
-           for (::size_t i = 0; i < n; ++i) {
-               strings[i] = source[(int)i].first;
-               lengths[i] = source[(int)i].second;
-           }
-           cl_int err;
-           cl_program myprog = clCreateProgramWithSource(context(), (cl_uint)n, strings, lengths, &err);
-           err = clBuildProgram(myprog, (cl_uint)ourdevices.size(), (cl_device_id*)&ourdevices.front(), NULL, NULL, NULL);
-           char proglog[1024];
-           clGetProgramBuildInfo(myprog, device(), CL_PROGRAM_BUILD_LOG, 1024, proglog, 0);
-           printf("err=%d log=%s\n", err, proglog);
-           return 0;
-        #endif
-
-        // Make program of the source code in the context
-        Program program = Program(context, source);
- 
-        // Build program for the specific device we are using
-        // IMPORTANT: If this program is running on a system that has multiple
-        // graphics cards, and if ANY of those cards does not support Images,
-        // then "program.build(devices);" will fail. Thus, we must build only
-        // on the device that is actually being used for the command queue.
-        program.build(ourdevices);
- 
-        // Make kernel
-        Kernel kernel(program, "grayscott_compute_2x2");
-
-        NDRange global(X/2,Y/2);
-        NDRange local(16,16);
-
-        kernel.setArg(4, f);
-        kernel.setArg(5, f+k);
-        kernel.setArg(6, r_a);
-        kernel.setArg(7, r_b);
-        kernel.setArg(8, speed);
-
-        int iteration = 0;
-        const int N_FRAMES_PER_DISPLAY = 5000;  // an even number, because of our double-buffering implementation
-        while(true) 
-        {
-            struct timeval tod_record;
-            double tod_before, tod_after, tod_elap;
-
-            gettimeofday(&tod_record, 0);
-            tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            // run a few iterations (without copying the data back)
-            for(int it=0;it<N_FRAMES_PER_DISPLAY/2;it++)
-            {
-                // (buffer-switching)
-
-                kernel.setArg(0, a1);
-                kernel.setArg(1, b1);
-                kernel.setArg(2, a2);
-                kernel.setArg(3, b2);
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-
-                kernel.setArg(0, a2);
-                kernel.setArg(1, b2);
-                kernel.setArg(2, a1);
-                kernel.setArg(3, b1);
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-            }
-            iteration += N_FRAMES_PER_DISPLAY;
-
-            // read back into a buffer
-            queue.enqueueReadImage(a1,true,origin,region,0,0,a);
-            //queue.enqueueReadImage(b1,true,origin,region,0,0,b);
-
-            gettimeofday(&tod_record, 0);
-            tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            tod_elap = tod_after - tod_before;
-
-            char msg[1000];
-            float fps = 0.0f;     // frames per second
-            if (tod_elap > 0)
-                fps = N_FRAMES_PER_DISPLAY / (float)tod_elap;
-            sprintf(msg,"GrayScott - %0.2f fps (%.2f Mcgs)",fps,fps*X*Y/1e6);
-
-            // display:
-            {
-                int quitnow = display(a,iteration,false,200.0f,2,10,msg);
-                if (quitnow)
-                    break;
-            }
-        }
-    } 
-    catch(Error error) 
-    {
-       std::cout << error.what() << "(" << error.err() << ")" << std::endl;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float* a,float* b)
-{
-    srand((unsigned int)time(NULL));
-
-    // figure the values
-    for(int i = 0; i < X; i++) 
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            // start with a uniform field with an approximate circle in the middle
-            if(hypot(i-X/3,(j-Y/4)/1.5)<=frand(2,5))
-            {
-                float_at(a,i,j) = frand(0.0f,0.1f);
-                float_at(b,i,j) = frand(0.9f,1.0f);
-            }
-            else 
-            {
-                float_at(a,i,j) = frand(0.9f,1.0f);
-                float_at(b,i,j) = frand(0.0f,0.1f);
-            }
-
-        }
-    }
-}
-
-
-bool display(float *a,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2,*im3;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    for(int col=0;col<X;col++)
-    {
-        for(int row=0;row<Y;row++)
-        {
-            float val = float_at(a,col,row);
-            val *= manual_brighten;
-            if(val<0) val=0; if(val>255) val=255;
-            ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 2] = (uchar)val;
-            ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 1] = (uchar)val;
-            ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 0] = (uchar)val;
-        }
-    }
-
-    cvResize(im,im2);
-
-    {
-        char txt[100];
-        sprintf(txt,"%d",iteration);
-        cvPutText(im2,txt,cvPoint(20,20),&font,white);
-        cvPutText(im2,message,cvPoint(20,40),&font,white);
-    }
-
-    cvShowImage(title,im2);
-
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        return true;
-    }
-    return false;
-}
diff --git a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/grayscott_kernel_image_2x2.cl b/SpeedComparisons/GrayScott_OpenCL_Image_2x2/grayscott_kernel_image_2x2.cl
deleted file mode 100644
index 938493c22..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Image_2x2/grayscott_kernel_image_2x2.cl
+++ /dev/null
@@ -1,42 +0,0 @@
-__kernel void grayscott_compute_2x2(
-    read_only image2d_t a, 
-    read_only image2d_t b,
-    write_only image2d_t a2, 
-    write_only image2d_t b2,
-    float f,float f_plus_k,
-    float r_a,float r_b,
-    float speed)
-{
-    const sampler_t smp = CLK_NORMALIZED_COORDS_FALSE | CLK_ADDRESS_REPEAT | CLK_FILTER_NEAREST;
-
-    const int x = get_global_id(0);
-    const int y = get_global_id(1);
-
-    const float4 a_pixel = read_imagef(a, smp, (int2)(x, y));
-    const float4 a_left = read_imagef(a, smp, (int2)(x-1, y));
-    const float4 a_right = read_imagef(a, smp, (int2)(x+1, y));
-    const float4 a_up = read_imagef(a, smp, (int2)(x, y-1));
-    const float4 a_down = read_imagef(a, smp, (int2)(x, y+1));
-    const float4 b_pixel = read_imagef(b, smp, (int2)(x, y));
-    const float4 b_left = read_imagef(b, smp, (int2)(x-1, y));
-    const float4 b_right = read_imagef(b, smp, (int2)(x+1, y));
-    const float4 b_up = read_imagef(b, smp, (int2)(x, y-1));
-    const float4 b_down = read_imagef(b, smp, (int2)(x, y+1));
- 
-    // we pack the values in 2x2 blocks:   x y
-    //                                     z w
-
-    write_imagef( a2, (int2)(x, y), a_pixel + speed * (r_a * ((float4)(
-        a_left.y + a_up.z + a_pixel.y + a_pixel.z,
-        a_pixel.x + a_up.w + a_right.x + a_pixel.w,
-        a_left.w + a_pixel.x + a_pixel.w + a_down.x,
-        a_pixel.z + a_pixel.y + a_right.z + a_down.y) - 4.0f*a_pixel) 
-        - a_pixel * b_pixel * b_pixel + f*(1.0f-a_pixel)) );
-    write_imagef( b2, (int2)(x, y), b_pixel + speed * (r_b * ((float4)(
-        b_left.y + b_up.z + b_pixel.y + b_pixel.z,
-        b_pixel.x + b_up.w + b_right.x + b_pixel.w,
-        b_left.w + b_pixel.x + b_pixel.w + b_down.x,
-        b_pixel.z + b_pixel.y + b_right.z + b_down.y) - 4.0f*b_pixel) 
-        + a_pixel * b_pixel * b_pixel - f_plus_k*b_pixel) );
-    // (it's faster than using local const floats to split up the computation, annoyingly)
-}
diff --git a/SpeedComparisons/GrayScott_OpenCL_Local/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenCL_Local/CMakeLists.txt
deleted file mode 100644
index 69b345140..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Local/CMakeLists.txt
+++ /dev/null
@@ -1,29 +0,0 @@
-project(GrayScott_OpenCL_Local)
-
-set(CMAKE_MODULE_PATH ${GrayScott_OpenCL_Local_SOURCE_DIR})
-# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
-
-find_package(OpenCV REQUIRED)
-include_directories( ${OPENCV_INCLUDE_DIR})
-link_libraries( ${OpenCV_LIBS} )
-
-# only build the OpenCL version if OpenCL was found
-find_package ( OpenCL )
-if(OPENCL_FOUND)
-  include_directories ( ${OPENCL_INCLUDE_DIRS} )
-  link_libraries ( ${OPENCL_LIBRARIES} )
-
-  # tell the code where the .cl file will live
-  add_definitions(-DCL_SOURCE_DIR="${GrayScott_OpenCL_Local_SOURCE_DIR}")
-
-  INCLUDE_DIRECTORIES( "../Display" )
-
-  add_executable(GrayScott_OpenCL_Local
-    gray_scott_opencl_local.cpp 
-    grayscott_kernel_local.cl 
-    ../Display/display.cpp
-    ../Display/display.h
-    ../Display/defs.h
-  )
-endif()
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_Local/FindOpenCL.cmake b/SpeedComparisons/GrayScott_OpenCL_Local/FindOpenCL.cmake
deleted file mode 100644
index fde90efae..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Local/FindOpenCL.cmake
+++ /dev/null
@@ -1,79 +0,0 @@
-# - Try to find OpenCL
-# This module tries to find an OpenCL implementation on your system. It supports
-# AMD / ATI, Apple and NVIDIA implementations, but shoudl work, too.
-#
-# Once done this will define
-#  OPENCL_FOUND        - system has OpenCL
-#  OPENCL_INCLUDE_DIRS  - the OpenCL include directory
-#  OPENCL_LIBRARIES    - link these to use OpenCL
-#
-# WIN32 should work, but is untested
-
-FIND_PACKAGE( PackageHandleStandardArgs )
-
-SET (OPENCL_VERSION_STRING "0.1.0")
-SET (OPENCL_VERSION_MAJOR 0)
-SET (OPENCL_VERSION_MINOR 1)
-SET (OPENCL_VERSION_PATCH 0)
-
-IF (APPLE)
-
-  FIND_LIBRARY(OPENCL_LIBRARIES OpenCL DOC "OpenCL lib for OSX")
-  FIND_PATH(OPENCL_INCLUDE_DIRS OpenCL/cl.h DOC "Include for OpenCL on OSX")
-  FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS OpenCL/cl.hpp DOC "Include for OpenCL CPP bindings on OSX")
-
-ELSE (APPLE)
-
-	IF (WIN32)
-	
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h)
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp)
-	
-	    # The AMD SDK currently installs both x86 and x86_64 libraries
-	    # This is only a hack to find out architecture
-	    IF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86_64")
-	    ELSE (${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64")
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86")
-	    ENDIF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    FIND_LIBRARY(OPENCL_LIBRARIES OpenCL.lib ${OPENCL_LIB_DIR})
-	    
-	    GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-	    
-	    # On Win32 search relative to the library
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS "${_OPENCL_INC_CAND}")
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS "${_OPENCL_INC_CAND}")
-	
-	ELSE (WIN32)
-
-            # Unix style platforms
-            FIND_LIBRARY(OPENCL_LIBRARIES OpenCL
-              ENV LD_LIBRARY_PATH
-            )
-
-            GET_FILENAME_COMPONENT(OPENCL_LIB_DIR ${OPENCL_LIBRARIES} PATH)
-            GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-
-            # The AMD SDK currently does not place its headers
-            # in /usr/include, therefore also search relative
-            # to the library
-            FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-            FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-
-	ENDIF (WIN32)
-
-ENDIF (APPLE)
-
-FIND_PACKAGE_HANDLE_STANDARD_ARGS( OpenCL DEFAULT_MSG OPENCL_LIBRARIES OPENCL_INCLUDE_DIRS )
-
-IF( _OPENCL_CPP_INCLUDE_DIRS )
-	SET( OPENCL_HAS_CPP_BINDINGS TRUE )
-	LIST( APPEND OPENCL_INCLUDE_DIRS ${_OPENCL_CPP_INCLUDE_DIRS} )
-	# This is often the same, so clean up
-	LIST( REMOVE_DUPLICATES OPENCL_INCLUDE_DIRS )
-ENDIF( _OPENCL_CPP_INCLUDE_DIRS )
-
-MARK_AS_ADVANCED(
-  OPENCL_INCLUDE_DIRS
-)
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_Local/cl.hpp b/SpeedComparisons/GrayScott_OpenCL_Local/cl.hpp
deleted file mode 100644
index 99b86a665..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Local/cl.hpp
+++ /dev/null
@@ -1,4011 +0,0 @@
-/*******************************************************************************
- * Copyright (c) 2008-2010 The Khronos Group Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and/or associated documentation files (the
- * "Materials"), to deal in the Materials without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sublicense, and/or sell copies of the Materials, and to
- * permit persons to whom the Materials are furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice shall be included
- * in all copies or substantial portions of the Materials.
- *
- * THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
- * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
- * CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
- * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
- * MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
- ******************************************************************************/
-
-/*! \file
- *
- *   \brief C++ bindings for OpenCL 1.0 (rev 48) and OpenCL 1.1 (rev 33)    
- *   \author Benedict R. Gaster and Laurent Morichetti
- *   
- *   Additions and fixes from Brian Cole, March 3rd 2010.
- *   
- *   \version 1.1
- *   \date June 2010
- *
- *   Optional extension support
- *
- *         cl
- *         cl_ext_device_fission
- *				#define USE_CL_DEVICE_FISSION
- */
-
-/*! \mainpage
- * \section intro Introduction
- * For many large applications C++ is the language of choice and so it seems
- * reasonable to define C++ bindings for OpenCL.
- *
- *
- * The interface is contained with a single C++ header file \em cl.hpp and all
- * definitions are contained within the namespace \em cl. There is no additional
- * requirement to include \em cl.h and to use either the C++ or original C
- * bindings it is enough to simply include \em cl.hpp.
- *
- * The bindings themselves are lightweight and correspond closely to the
- * underlying C API. Using the C++ bindings introduces no additional execution
- * overhead.
- *
- * For detail documentation on the bindings see:
- *
- * The OpenCL C++ Wrapper API 1.1 (revision 04)
- *  http://www.khronos.org/registry/cl/specs/opencl-cplusplus-1.1.pdf
- *
- * \section example Example
- *
- * The following example shows a general use case for the C++
- * bindings, including support for the optional exception feature and
- * also the supplied vector and string classes, see following sections for
- * decriptions of these features.
- *
- * \code
- * #define __CL_ENABLE_EXCEPTIONS
- * 
- * #if defined(__APPLE__) || defined(__MACOSX)
- * #include <OpenCL/cl.hpp>
- * #else
- * #include <CL/cl.hpp>
- * #endif
- * #include <cstdio>
- * #include <cstdlib>
- * #include <iostream>
- * 
- *  const char * helloStr  = "__kernel void "
- *                           "hello(void) "
- *                           "{ "
- *                           "  "
- *                           "} ";
- * 
- *  int
- *  main(void)
- *  {
- *     cl_int err = CL_SUCCESS;
- *     try {
- *
- *       std::vector<cl::Platform> platforms;
- *       cl::Platform::get(&platforms);
- *       if (platforms.size() == 0) {
- *           std::cout << "Platform size 0\n";
- *           return -1;
- *       }
- *
- *       cl_context_properties properties[] = 
- *          { CL_CONTEXT_PLATFORM, (cl_context_properties)(platforms[0])(), 0};
- *       cl::Context context(CL_DEVICE_TYPE_CPU, properties); 
- * 
- *       std::vector<cl::Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
- * 
- *       cl::Program::Sources source(1,
- *           std::make_pair(helloStr,strlen(helloStr)));
- *       cl::Program program_ = cl::Program(context, source);
- *       program_.build(devices);
- * 
- *       cl::Kernel kernel(program_, "hello", &err);
- * 
- *       cl::Event event;
- *       cl::CommandQueue queue(context, devices[0], 0, &err);
- *       queue.enqueueNDRangeKernel(
- *           kernel, 
- *           cl::NullRange, 
- *           cl::NDRange(4,4),
- *           cl::NullRange,
- *           NULL,
- *           &event); 
- * 
- *       event.wait();
- *     }
- *     catch (cl::Error err) {
- *        std::cerr 
- *           << "ERROR: "
- *           << err.what()
- *           << "("
- *           << err.err()
- *           << ")"
- *           << std::endl;
- *     }
- * 
- *    return EXIT_SUCCESS;
- *  }
- * 
- * \endcode
- *
- */
-#ifndef CL_HPP_
-#define CL_HPP_
-
-#ifdef _WIN32
-#include <windows.h>
-#include <malloc.h>
-#if defined(USE_DX_INTEROP)
-#include <CL/cl_d3d10.h>
-#endif
-#endif // _WIN32
-
-// 
-#if defined(USE_CL_DEVICE_FISSION)
-#include <CL/cl_ext.h>
-#endif
-
-#if defined(__APPLE__) || defined(__MACOSX)
-#include <OpenGL/OpenGL.h>
-#include <OpenCL/opencl.h>
-#else
-#include <GL/gl.h>
-#include <CL/opencl.h>
-#endif // !__APPLE__
-
-#if !defined(CL_CALLBACK)
-#define CL_CALLBACK
-#endif //CL_CALLBACK
-
-#include <utility>
-
-#if !defined(__NO_STD_VECTOR)
-#include <vector>
-#endif
-
-#if !defined(__NO_STD_STRING)
-#include <string>
-#endif 
-
-#if defined(linux) || defined(__APPLE__) || defined(__MACOSX)
-# include <alloca.h>
-#endif // linux
-
-#include <cstring>
-
-/*! \namespace cl
- *
- * \brief The OpenCL C++ bindings are defined within this namespace.
- *
- */
-namespace cl {
-
-#define __INIT_CL_EXT_FCN_PTR(name) \
-    if(!pfn_##name) { \
-        pfn_##name = (PFN_##name) \
-            clGetExtensionFunctionAddress(#name); \
-        if(!pfn_##name) { \
-        } \
-    }
-
-class Program;
-class Device;
-class Context;
-class CommandQueue;
-class Memory;
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-#include <exception>
-/*! \class Error
- * \brief Exception class
- */
-class Error : public std::exception
-{
-private:
-    cl_int err_;
-    const char * errStr_;
-public:
-    /*! Create a new CL error exception for a given error code
-     *  and corresponding message.
-     */
-    Error(cl_int err, const char * errStr = NULL) : err_(err), errStr_(errStr)
-    {}
-
-    ~Error() throw() {}
-
-    /*! \brief Get error string associated with exception
-     *
-     * \return A memory pointer to the error message string.
-     */
-    virtual const char * what() const throw ()
-    {
-        if (errStr_ == NULL) {
-            return "empty";
-        }
-        else {
-            return errStr_;
-        }
-    }
-
-    /*! \brief Get error code associated with exception
-     *
-     *  \return The error code.
-     */
-    const cl_int err(void) const { return err_; }
-};
-
-#define __ERR_STR(x) #x
-#else
-#define __ERR_STR(x) NULL
-#endif // __CL_ENABLE_EXCEPTIONS
-
-//! \cond DOXYGEN_DETAIL
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#define __GET_DEVICE_INFO_ERR               __ERR_STR(clgetDeviceInfo)
-#define __GET_PLATFORM_INFO_ERR             __ERR_STR(clGetPlatformInfo)
-#define __GET_DEVICE_IDS_ERR                __ERR_STR(clGetDeviceIDs)
-#define __GET_PLATFORM_IDS_ERR              __ERR_STR(clGetPlatformIDs)
-#define __GET_CONTEXT_INFO_ERR              __ERR_STR(clGetContextInfo)
-#define __GET_EVENT_INFO_ERR                __ERR_STR(clGetEventInfo)
-#define __GET_EVENT_PROFILE_INFO_ERR        __ERR_STR(clGetEventProfileInfo)
-#define __GET_MEM_OBJECT_INFO_ERR           __ERR_STR(clGetMemObjectInfo)
-#define __GET_IMAGE_INFO_ERR                __ERR_STR(clGetImageInfo)
-#define __GET_SAMPLER_INFO_ERR              __ERR_STR(clGetSamplerInfo)
-#define __GET_KERNEL_INFO_ERR               __ERR_STR(clGetKernelInfo)
-#define __GET_KERNEL_WORK_GROUP_INFO_ERR    __ERR_STR(clGetKernelWorkGroupInfo)
-#define __GET_PROGRAM_INFO_ERR              __ERR_STR(clGetProgramInfo)
-#define __GET_PROGRAM_BUILD_INFO_ERR        __ERR_STR(clGetProgramBuildInfo)
-#define __GET_COMMAND_QUEUE_INFO_ERR        __ERR_STR(clGetCommandQueueInfo)
-
-#define __CREATE_CONTEXT_FROM_TYPE_ERR      __ERR_STR(clCreateContextFromType)
-#define __GET_SUPPORTED_IMAGE_FORMATS_ERR   __ERR_STR(clGetSupportedImageFormats)
-
-#define __CREATE_BUFFER_ERR                 __ERR_STR(clCreateBuffer)
-#define __CREATE_SUBBUFFER_ERR              __ERR_STR(clCreateSubBuffer)
-#define __CREATE_GL_BUFFER_ERR              __ERR_STR(clCreateFromGLBuffer)
-#define __GET_GL_OBJECT_INFO_ERR            __ERR_STR(clGetGLObjectInfo)
-#define __CREATE_IMAGE2D_ERR                __ERR_STR(clCreateImage2D)
-#define __CREATE_IMAGE3D_ERR                __ERR_STR(clCreateImage3D)
-#define __CREATE_SAMPLER_ERR                __ERR_STR(clCreateSampler)
-#define __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR __ERR_STR(clSetMemObjectDestructorCallback)
-
-#define __CREATE_USER_EVENT_ERR             __ERR_STR(clCreateUserEvent)
-#define __SET_USER_EVENT_STATUS_ERR         __ERR_STR(clSetUserEventStatus)
-#define __SET_EVENT_CALLBACK_ERR            __ERR_STR(clSetEventCallback)
-#define __WAIT_FOR_EVENTS_ERR               __ERR_STR(clWaitForEvents)
-
-#define __CREATE_KERNEL_ERR                 __ERR_STR(clCreateKernel)
-#define __SET_KERNEL_ARGS_ERR               __ERR_STR(clSetKernelArg)
-#define __CREATE_PROGRAM_WITH_SOURCE_ERR    __ERR_STR(clCreateProgramWithSource)
-#define __CREATE_PROGRAM_WITH_BINARY_ERR    __ERR_STR(clCreateProgramWithBinary)
-#define __BUILD_PROGRAM_ERR                 __ERR_STR(clBuildProgram)
-#define __CREATE_KERNELS_IN_PROGRAM_ERR     __ERR_STR(clCreateKernelsInProgram)
-
-#define __CREATE_COMMAND_QUEUE_ERR          __ERR_STR(clCreateCommandQueue)
-#define __SET_COMMAND_QUEUE_PROPERTY_ERR    __ERR_STR(clSetCommandQueueProperty)
-#define __ENQUEUE_READ_BUFFER_ERR           __ERR_STR(clEnqueueReadBuffer)
-#define __ENQUEUE_READ_BUFFER_RECT_ERR      __ERR_STR(clEnqueueReadBufferRect)
-#define __ENQUEUE_WRITE_BUFFER_ERR          __ERR_STR(clEnqueueWriteBuffer)
-#define __ENQUEUE_WRITE_BUFFER_RECT_ERR     __ERR_STR(clEnqueueWriteBufferRect)
-#define __ENQEUE_COPY_BUFFER_ERR            __ERR_STR(clEnqueueCopyBuffer)
-#define __ENQEUE_COPY_BUFFER_RECT_ERR       __ERR_STR(clEnqueueCopyBufferRect)
-#define __ENQUEUE_READ_IMAGE_ERR            __ERR_STR(clEnqueueReadImage)
-#define __ENQUEUE_WRITE_IMAGE_ERR           __ERR_STR(clEnqueueWriteImage)
-#define __ENQUEUE_COPY_IMAGE_ERR            __ERR_STR(clEnqueueCopyImage)
-#define __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR  __ERR_STR(clEnqueueCopyImageToBuffer)
-#define __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR  __ERR_STR(clEnqueueCopyBufferToImage)
-#define __ENQUEUE_MAP_BUFFER_ERR            __ERR_STR(clEnqueueMapBuffer)
-#define __ENQUEUE_MAP_IMAGE_ERR             __ERR_STR(clEnqueueMapImage)
-#define __ENQUEUE_UNMAP_MEM_OBJECT_ERR      __ERR_STR(clEnqueueUnMapMemObject)
-#define __ENQUEUE_NDRANGE_KERNEL_ERR        __ERR_STR(clEnqueueNDRangeKernel)
-#define __ENQUEUE_TASK_ERR                  __ERR_STR(clEnqueueTask)
-#define __ENQUEUE_NATIVE_KERNEL             __ERR_STR(clEnqueueNativeKernel)
-#define __ENQUEUE_MARKER_ERR                __ERR_STR(clEnqueueMarker)
-#define __ENQUEUE_WAIT_FOR_EVENTS_ERR       __ERR_STR(clEnqueueWaitForEvents)
-#define __ENQUEUE_BARRIER_ERR               __ERR_STR(clEnqueueBarrier)
-
-#define __ENQUEUE_ACQUIRE_GL_ERR            __ERR_STR(clEnqueueAcquireGLObjects)
-#define __ENQUEUE_RELEASE_GL_ERR            __ERR_STR(clEnqueueReleaseGLObjects)
-
-#define __UNLOAD_COMPILER_ERR               __ERR_STR(clUnloadCompiler)
-
-#define __FLUSH_ERR                         __ERR_STR(clFlush)
-#define __FINISH_ERR                        __ERR_STR(clFinish)
-
-#define __CREATE_SUB_DEVICES                __ERR_STR(clCreateSubDevicesEXT)
-#endif // __CL_USER_OVERRIDE_ERROR_STRINGS
-//! \endcond
-
-/*! \class string
- * \brief Simple string class, that provides a limited subset of std::string
- * functionality but avoids many of the issues that come with that class.
- */
-class string
-{
-private:
-    ::size_t size_;
-    char * str_;
-public:
-    string(void) : size_(0), str_(NULL)
-    {
-    }
-
-    string(char * str, ::size_t size) :
-        size_(size),
-        str_(NULL)
-    {
-        str_ = new char[size_+1];
-        if (str_ != NULL) {
-            memcpy(str_, str, size_  * sizeof(char));
-            str_[size_] = '\0';
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string(char * str) :
-        str_(NULL)
-    {
-        size_= ::strlen(str);
-        str_ = new char[size_ + 1];
-        if (str_ != NULL) {
-            memcpy(str_, str, (size_ + 1) * sizeof(char));
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string& operator=(const string& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        if (rhs.size_ == 0 || rhs.str_ == NULL) {
-            size_ = 0;
-            str_  = NULL;
-        } 
-        else {
-            size_ = rhs.size_;
-            str_ = new char[size_ + 1];
-            if (str_ != NULL) {
-                memcpy(str_, rhs.str_, (size_ + 1) * sizeof(char));
-            }
-            else {
-                size_ = 0;
-            }
-        }
-
-        return *this;
-    }
-
-    string(const string& rhs)
-    {
-        *this = rhs;
-    }
-
-    ~string()
-    {
-        if (str_ != NULL) {
-            delete[] str_;
-        }
-    }
-
-    ::size_t size(void) const   { return size_; }
-    ::size_t length(void) const { return size(); }
-
-    const char * c_str(void) const { return (str_) ? str_ : "";}
-};
-
-#if !defined(__USE_DEV_STRING) && !defined(__NO_STD_STRING)
-#include <string>
-typedef std::string STRING_CLASS;
-#elif !defined(__USE_DEV_STRING) 
-typedef cl::string STRING_CLASS;
-#endif
-
-#if !defined(__USE_DEV_VECTOR) && !defined(__NO_STD_VECTOR)
-#include <vector>
-#define VECTOR_CLASS std::vector
-#elif !defined(__USE_DEV_VECTOR) 
-#define VECTOR_CLASS cl::vector 
-#endif
-
-#if !defined(__MAX_DEFAULT_VECTOR_SIZE)
-#define __MAX_DEFAULT_VECTOR_SIZE 10
-#endif
-
-/*! \class vector
- * \brief Fixed sized vector implementation that mirroring 
- * std::vector functionality.
- */
-template <typename T, unsigned int N = __MAX_DEFAULT_VECTOR_SIZE>
-class vector
-{
-private:
-    T data_[N];
-    unsigned int size_;
-    bool empty_;
-public:
-    vector() : 
-        size_(-1),
-        empty_(true)
-    {}
-
-    ~vector() {}
-
-    unsigned int size(void) const
-    {
-        return size_ + 1;
-    }
-
-    void clear()
-    {
-        size_ = -1;
-        empty_ = true;
-    }
-
-    void push_back (const T& x)
-    { 
-        if (size() < N) {
-            size_++;  
-            data_[size_] = x;
-            empty_ = false;
-        }
-    }
-
-    void pop_back(void)
-    {
-        if (!empty_) {
-            data_[size_].~T();
-            size_--;
-            if (size_ == -1) {
-                empty_ = true;
-            }
-        }
-    }
-  
-    vector(const vector<T, N>& vec) : 
-        size_(vec.size_),
-        empty_(vec.empty_)
-    {
-        if (!empty_) {
-            memcpy(&data_[0], &vec.data_[0], size() * sizeof(T));
-        }
-    } 
-
-    vector(unsigned int size, const T& val = T()) :
-        size_(-1),
-        empty_(true)
-    {
-        for (unsigned int i = 0; i < size; i++) {
-            push_back(val);
-        }
-    }
-
-    vector<T, N>& operator=(const vector<T, N>& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        size_  = rhs.size_;
-        empty_ = rhs.empty_;
-
-        if (!empty_) {	
-            memcpy(&data_[0], &rhs.data_[0], size() * sizeof(T));
-        }
-    
-        return *this;
-    }
-
-    bool operator==(vector<T,N> &vec)
-    {
-        if (empty_ && vec.empty_) {
-            return true;
-        }
-
-        if (size() != vec.size()) {
-            return false;
-        }
-
-        return memcmp(&data_[0], &vec.data_[0], size() * sizeof(T)) == 0 ? true : false;
-    }
-  
-    operator T* ()             { return data_; }
-    operator const T* () const { return data_; }
-   
-    bool empty (void) const
-    {
-        return empty_;
-    }
-  
-    unsigned int max_size (void) const
-    {
-        return N;
-    }
-
-    unsigned int capacity () const
-    {
-        return sizeof(T) * N;
-    }
-
-    T& operator[](int index)
-    {
-        return data_[index];
-    }
-  
-    T operator[](int index) const
-    {
-        return data_[index];
-    }
-  
-    template<class I>
-    void assign(I start, I end)
-    {
-        clear();   
-        while(start < end) {
-            push_back(*start);
-            start++;
-        }
-    }
-
-    /*! \class iterator
-     * \brief Iterator class for vectors
-     */
-    class iterator
-    {
-    private:
-        vector<T,N> vec_;
-        int index_;
-        bool initialized_;
-    public:
-        iterator(void) : 
-            index_(-1),
-            initialized_(false)
-        {
-            index_ = -1;
-            initialized_ = false;
-        }
-
-        ~iterator(void) {}
-
-        static iterator begin(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = 0;
-            }
-
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-
-        static iterator end(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = vec.size();
-            }
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-    
-        bool operator==(iterator i)
-        {
-            return ((vec_ == i.vec_) && 
-                    (index_ == i.index_) && 
-                    (initialized_ == i.initialized_));
-        }
-
-        bool operator!=(iterator i)
-        {
-            return (!(*this==i));
-        }
-
-        void operator++()
-        {
-            index_++;
-        }
-
-        void operator++(int x)
-        {
-            index_ += x;
-        }
-
-        void operator--()
-        {
-            index_--;
-        }
-
-        void operator--(int x)
-        {
-            index_ -= x;
-        }
-
-        T operator *()
-        {
-            return vec_[index_];
-        }
-    };
-
-    iterator begin(void)
-    {
-        return iterator::begin(*this);
-    }
-
-    iterator end(void)
-    {
-        return iterator::end(*this);
-    }
-
-    T& front(void)
-    {
-        return data_[0];
-    }
-
-    T& back(void)
-    {
-        return data_[size_];
-    }
-
-    const T& front(void) const
-    {
-        return data_[0];
-    }
-
-    const T& back(void) const
-    {
-        return data_[size_];
-    }
-};  
-    
-/*!
- * \brief size_t class used to interface between C++ and
- * OpenCL C calls that require arrays of size_t values, who's
- * size is known statically.
- */
-template <int N>
-struct size_t : public cl::vector< ::size_t, N> { };
-
-namespace detail {
-
-// GetInfo help struct
-template <typename Functor, typename T>
-struct GetInfoHelper
-{
-    static cl_int
-    get(Functor f, cl_uint name, T* param)
-    {
-        return f(name, sizeof(T), param, NULL);
-    }
-};
-
-// Specialized GetInfoHelper for VECTOR_CLASS params
-template <typename Func, typename T>
-struct GetInfoHelper<Func, VECTOR_CLASS<T> >
-{
-    static cl_int get(Func f, cl_uint name, VECTOR_CLASS<T>* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        T* value = (T*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        param->assign(&value[0], &value[required/sizeof(T)]);
-        return CL_SUCCESS;
-    }
-};
-
-// Specialized for getInfo<CL_PROGRAM_BINARIES>
-template <typename Func>
-struct GetInfoHelper<Func, VECTOR_CLASS<char *> >
-{
-    static cl_int
-    get(Func f, cl_uint name, VECTOR_CLASS<char *>* param)
-    {
-      cl_uint err = f(name, param->size() * sizeof(char *), &(*param)[0], NULL);
-      if (err != CL_SUCCESS) {
-        return err;
-      }
-      
-      return CL_SUCCESS;
-    }
-};
-
-// Specialized GetInfoHelper for STRING_CLASS params
-template <typename Func>
-struct GetInfoHelper<Func, STRING_CLASS>
-{
-    static cl_int get(Func f, cl_uint name, STRING_CLASS* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        char* value = (char*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        *param = value;
-        return CL_SUCCESS;
-    }
-};
-
-#define __GET_INFO_HELPER_WITH_RETAIN(CPP_TYPE) \
-namespace detail { \
-template <typename Func> \
-struct GetInfoHelper<Func, CPP_TYPE> \
-{ \
-    static cl_int get(Func f, cl_uint name, CPP_TYPE* param) \
-    { \
-      cl_uint err = f(name, sizeof(CPP_TYPE), param, NULL); \
-      if (err != CL_SUCCESS) { \
-        return err; \
-      } \
-      \
-      return ReferenceHandler<CPP_TYPE::cl_type>::retain((*param)()); \
-    } \
-}; \
-} 
-
-
-#define __PARAM_NAME_INFO_1_0(F) \
-    F(cl_platform_info, CL_PLATFORM_PROFILE, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VERSION, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_NAME, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VENDOR, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_device_info, CL_DEVICE_TYPE, cl_device_type) \
-    F(cl_device_info, CL_DEVICE_VENDOR_ID, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_COMPUTE_UNITS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_CLOCK_FREQUENCY, cl_uint) \
-    F(cl_device_info, CL_DEVICE_ADDRESS_BITS, cl_bitfield) \
-    F(cl_device_info, CL_DEVICE_MAX_READ_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WRITE_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_MEM_ALLOC_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_DEPTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE_SUPPORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_PARAMETER_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_SAMPLERS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MEM_BASE_ADDR_ALIGN, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_SINGLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_TYPE, cl_device_mem_cache_type) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE, cl_uint)\
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_TYPE, cl_device_local_mem_type) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_ERROR_CORRECTION_SUPPORT, cl_bool) \
-    F(cl_device_info, CL_DEVICE_PROFILING_TIMER_RESOLUTION, ::size_t) \
-    F(cl_device_info, CL_DEVICE_ENDIAN_LITTLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_COMPILER_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_EXECUTION_CAPABILITIES, cl_device_exec_capabilities) \
-    F(cl_device_info, CL_DEVICE_QUEUE_PROPERTIES, cl_command_queue_properties) \
-    F(cl_device_info, CL_DEVICE_PLATFORM, cl_platform_id) \
-    F(cl_device_info, CL_DEVICE_NAME, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VENDOR, STRING_CLASS) \
-    F(cl_device_info, CL_DRIVER_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_PROFILE, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_context_info, CL_CONTEXT_REFERENCE_COUNT, cl_uint) \
-    F(cl_context_info, CL_CONTEXT_DEVICES, VECTOR_CLASS<Device>) \
-    F(cl_context_info, CL_CONTEXT_PROPERTIES, VECTOR_CLASS<cl_context_properties>) \
-    \
-    F(cl_event_info, CL_EVENT_COMMAND_QUEUE, cl::CommandQueue) \
-    F(cl_event_info, CL_EVENT_COMMAND_TYPE, cl_command_type) \
-    F(cl_event_info, CL_EVENT_REFERENCE_COUNT, cl_uint) \
-    F(cl_event_info, CL_EVENT_COMMAND_EXECUTION_STATUS, cl_uint) \
-    \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_QUEUED, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_SUBMIT, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_START, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_END, cl_ulong) \
-    \
-    F(cl_mem_info, CL_MEM_TYPE, cl_mem_object_type) \
-    F(cl_mem_info, CL_MEM_FLAGS, cl_mem_flags) \
-    F(cl_mem_info, CL_MEM_SIZE, ::size_t) \
-    F(cl_mem_info, CL_MEM_HOST_PTR, void*) \
-    F(cl_mem_info, CL_MEM_MAP_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_REFERENCE_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_CONTEXT, cl::Context) \
-    \
-    F(cl_image_info, CL_IMAGE_FORMAT, cl_image_format) \
-    F(cl_image_info, CL_IMAGE_ELEMENT_SIZE, ::size_t) \
-    F(cl_image_info, CL_IMAGE_ROW_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_SLICE_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_WIDTH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_HEIGHT, ::size_t) \
-    F(cl_image_info, CL_IMAGE_DEPTH, ::size_t) \
-    \
-    F(cl_sampler_info, CL_SAMPLER_REFERENCE_COUNT, cl_uint) \
-    F(cl_sampler_info, CL_SAMPLER_CONTEXT, cl::Context) \
-    F(cl_sampler_info, CL_SAMPLER_NORMALIZED_COORDS, cl_addressing_mode) \
-    F(cl_sampler_info, CL_SAMPLER_ADDRESSING_MODE, cl_filter_mode) \
-    F(cl_sampler_info, CL_SAMPLER_FILTER_MODE, cl_bool) \
-    \
-    F(cl_program_info, CL_PROGRAM_REFERENCE_COUNT, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_CONTEXT, cl::Context) \
-    F(cl_program_info, CL_PROGRAM_NUM_DEVICES, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_DEVICES, VECTOR_CLASS<cl_device_id>) \
-    F(cl_program_info, CL_PROGRAM_SOURCE, STRING_CLASS) \
-    F(cl_program_info, CL_PROGRAM_BINARY_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_program_info, CL_PROGRAM_BINARIES, VECTOR_CLASS<char *>) \
-    \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_STATUS, cl_build_status) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_OPTIONS, STRING_CLASS) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_LOG, STRING_CLASS) \
-    \
-    F(cl_kernel_info, CL_KERNEL_FUNCTION_NAME, STRING_CLASS) \
-    F(cl_kernel_info, CL_KERNEL_NUM_ARGS, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_REFERENCE_COUNT, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_CONTEXT, cl::Context) \
-    F(cl_kernel_info, CL_KERNEL_PROGRAM, cl::Program) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_COMPILE_WORK_GROUP_SIZE, cl::size_t<3>) \
-    F(cl_kernel_work_group_info, CL_KERNEL_LOCAL_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_command_queue_info, CL_QUEUE_CONTEXT, cl::Context) \
-    F(cl_command_queue_info, CL_QUEUE_DEVICE, cl::Device) \
-    F(cl_command_queue_info, CL_QUEUE_REFERENCE_COUNT, cl_uint) \
-    F(cl_command_queue_info, CL_QUEUE_PROPERTIES, cl_command_queue_properties)
-
-#if defined(CL_VERSION_1_1)
-#define __PARAM_NAME_INFO_1_1(F) \
-    F(cl_context_info, CL_CONTEXT_NUM_DEVICES, cl_uint)\
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_DOUBLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HALF_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HOST_UNIFIED_MEMORY, cl_bool) \
-    \
-    F(cl_mem_info, CL_MEM_ASSOCIATED_MEMOBJECT, cl::Memory) \
-    F(cl_mem_info, CL_MEM_OFFSET, ::size_t) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_PRIVATE_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_event_info, CL_EVENT_CONTEXT, cl::Context)
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-#define __PARAM_NAME_DEVICE_FISSION(F) \
-    F(cl_device_info, CL_DEVICE_PARENT_DEVICE_EXT, cl_device_id) \
-	F(cl_device_info, CL_DEVICE_PARTITION_TYPES_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_AFFINITY_DOMAINS_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_REFERENCE_COUNT_EXT , cl_uint) \
-	F(cl_device_info, CL_DEVICE_PARTITION_STYLE_EXT, VECTOR_CLASS<cl_device_partition_property_ext>)
-#endif // USE_CL_DEVICE_FISSION
-
-template <typename enum_type, cl_int Name>
-struct param_traits {};
-
-#define __DECLARE_PARAM_TRAITS(token, param_name, T) \
-struct token;                                        \
-template<>                                           \
-struct param_traits<detail:: token,param_name>       \
-{                                                    \
-    enum { value = param_name };                     \
-    typedef T param_type;                            \
-};
-
-__PARAM_NAME_INFO_1_0(__DECLARE_PARAM_TRAITS);
-#if defined(CL_VERSION_1_1)
-__PARAM_NAME_INFO_1_1(__DECLARE_PARAM_TRAITS);
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-__PARAM_NAME_DEVICE_FISSION(__DECLARE_PARAM_TRAITS);
-#endif // USE_CL_DEVICE_FISSION
-
-#undef __DECLARE_PARAM_TRAITS
-
-// Convenience functions
-
-template <typename Func, typename T>
-inline cl_int
-getInfo(Func f, cl_uint name, T* param)
-{
-    return GetInfoHelper<Func, T>::get(f, name, param);
-}
-
-template <typename Func, typename Arg0>
-struct GetInfoFunctor0
-{
-    Func f_; const Arg0& arg0_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename Arg1>
-struct GetInfoFunctor1
-{
-    Func f_; const Arg0& arg0_; const Arg1& arg1_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, arg1_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, cl_uint name, T* param)
-{
-    GetInfoFunctor0<Func, Arg0> f0 = { f, arg0 };
-    return GetInfoHelper<GetInfoFunctor0<Func, Arg0>, T>
-        ::get(f0, name, param);
-}
-
-template <typename Func, typename Arg0, typename Arg1, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, const Arg1& arg1, cl_uint name, T* param)
-{
-    GetInfoFunctor1<Func, Arg0, Arg1> f0 = { f, arg0, arg1 };
-    return GetInfoHelper<GetInfoFunctor1<Func, Arg0, Arg1>, T>
-        ::get(f0, name, param);
-}
-
-template<typename T>
-struct ReferenceHandler
-{ };
-
-template <>
-struct ReferenceHandler<cl_device_id>
-{
-    // cl_device_id does not have retain().
-    static cl_int retain(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-    // cl_device_id does not have release().
-    static cl_int release(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-};
-
-template <>
-struct ReferenceHandler<cl_platform_id>
-{
-    // cl_platform_id does not have retain().
-    static cl_int retain(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-    // cl_platform_id does not have release().
-    static cl_int release(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-};
-
-template <>
-struct ReferenceHandler<cl_context>
-{
-    static cl_int retain(cl_context context)
-    { return ::clRetainContext(context); }
-    static cl_int release(cl_context context)
-    { return ::clReleaseContext(context); }
-};
-
-template <>
-struct ReferenceHandler<cl_command_queue>
-{
-    static cl_int retain(cl_command_queue queue)
-    { return ::clRetainCommandQueue(queue); }
-    static cl_int release(cl_command_queue queue)
-    { return ::clReleaseCommandQueue(queue); }
-};
-
-template <>
-struct ReferenceHandler<cl_mem>
-{
-    static cl_int retain(cl_mem memory)
-    { return ::clRetainMemObject(memory); }
-    static cl_int release(cl_mem memory)
-    { return ::clReleaseMemObject(memory); }
-};
-
-template <>
-struct ReferenceHandler<cl_sampler>
-{
-    static cl_int retain(cl_sampler sampler)
-    { return ::clRetainSampler(sampler); }
-    static cl_int release(cl_sampler sampler)
-    { return ::clReleaseSampler(sampler); }
-};
-
-template <>
-struct ReferenceHandler<cl_program>
-{
-    static cl_int retain(cl_program program)
-    { return ::clRetainProgram(program); }
-    static cl_int release(cl_program program)
-    { return ::clReleaseProgram(program); }
-};
-
-template <>
-struct ReferenceHandler<cl_kernel>
-{
-    static cl_int retain(cl_kernel kernel)
-    { return ::clRetainKernel(kernel); }
-    static cl_int release(cl_kernel kernel)
-    { return ::clReleaseKernel(kernel); }
-};
-
-template <>
-struct ReferenceHandler<cl_event>
-{
-    static cl_int retain(cl_event event)
-    { return ::clRetainEvent(event); }
-    static cl_int release(cl_event event)
-    { return ::clReleaseEvent(event); }
-};
-
-template <typename T>
-class Wrapper
-{
-public:
-    typedef T cl_type;
-
-protected:
-    cl_type object_;
-
-public:
-    Wrapper() : object_(NULL) { }
-
-    ~Wrapper()
-    {
-        if (object_ != NULL) { release(); }
-    }
-
-    Wrapper(const Wrapper<cl_type>& rhs)
-    {
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-    }
-
-    Wrapper<cl_type>& operator = (const Wrapper<cl_type>& rhs)
-    {
-        if (object_ != NULL) { release(); }
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-        return *this;
-    }
-
-    cl_type operator ()() const { return object_; }
-
-    cl_type& operator ()() { return object_; }
-
-protected:
-
-    cl_int retain() const
-    {
-        return ReferenceHandler<cl_type>::retain(object_);
-    }
-
-    cl_int release() const
-    {
-        return ReferenceHandler<cl_type>::release(object_);
-    }
-};
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-static inline cl_int errHandler (
-    cl_int err,
-    const char * errStr = NULL) throw(Error)
-{
-    if (err != CL_SUCCESS) {
-        throw Error(err, errStr);
-    }
-    return err;
-}
-#else
-static inline cl_int errHandler (cl_int err, const char * errStr = NULL)
-{
-    return err;
-}
-#endif // __CL_ENABLE_EXCEPTIONS
-
-} // namespace detail
-//! \endcond
-
-/*! \stuct ImageFormat
- * \brief ImageFormat interface fro cl_image_format.
- */
-struct ImageFormat : public cl_image_format
-{
-    ImageFormat(){}
-
-    ImageFormat(cl_channel_order order, cl_channel_type type)
-    {
-        image_channel_order = order;
-        image_channel_data_type = type;
-    }
-
-    ImageFormat& operator = (const ImageFormat& rhs)
-    {
-        if (this != &rhs) {
-            this->image_channel_data_type = rhs.image_channel_data_type;
-            this->image_channel_order     = rhs.image_channel_order;
-        }
-        return *this;
-    }
-};
-
-/*! \class Device
- * \brief Device interface for cl_device_id.
- */
-class Device : public detail::Wrapper<cl_device_id>
-{
-public:
-    Device(cl_device_id device) { object_ = device; }
-
-    Device() : detail::Wrapper<cl_type>() { }
-
-    Device(const Device& device) : detail::Wrapper<cl_type>(device) { }
-
-    Device& operator = (const Device& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_device_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetDeviceInfo, object_, name, param),
-            __GET_DEVICE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_device_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_device_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(USE_CL_DEVICE_FISSION)
-	cl_int createSubDevices(
-		const cl_device_partition_property_ext * properties,
-		VECTOR_CLASS<Device>* devices)
-	{
-		typedef CL_API_ENTRY cl_int 
-			( CL_API_CALL * PFN_clCreateSubDevicesEXT)(
-				cl_device_id /*in_device*/,
-                const cl_device_partition_property_ext * /* properties */,
-                cl_uint /*num_entries*/,
-                cl_device_id * /*out_devices*/,
-                cl_uint * /*num_devices*/ ) CL_EXT_SUFFIX__VERSION_1_1;
-
-		static PFN_clCreateSubDevicesEXT pfn_clCreateSubDevicesEXT = NULL;
-		__INIT_CL_EXT_FCN_PTR(clCreateSubDevicesEXT);
-
-		cl_uint n = 0;
-        cl_int err = pfn_clCreateSubDevicesEXT(object_, properties, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clCreateSubDevicesEXT(object_, properties, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
- 	}
-#endif
-};
-
-/*! \class Platform
- *  \brief Platform interface.
- */
-class Platform : public detail::Wrapper<cl_platform_id>
-{
-public:
-    static const Platform null();
-
-    Platform(cl_platform_id platform) { object_ = platform; }
-
-    Platform() : detail::Wrapper<cl_type>()  { }
-
-    Platform(const Platform& platform) : detail::Wrapper<cl_type>(platform) { }
-
-    Platform& operator = (const Platform& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getInfo(cl_platform_info name, STRING_CLASS* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetPlatformInfo, object_, name, param),
-            __GET_PLATFORM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_platform_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_platform_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getDevices(
-        cl_device_type type,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetDeviceIDs(object_, type, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = ::clGetDeviceIDs(object_, type, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-
-#if defined(USE_DX_INTEROP)
-   /*! \brief Get the list of available D3D10 devices.
-     *
-     *  \param d3d_device_source.
-     *
-     *  \param d3d_object.
-     *
-     *  \param d3d_device_set.
-     *
-     *  \param devices returns a vector of OpenCL D3D10 devices found. The cl::Device
-     *  values returned in devices can be used to identify a specific OpenCL
-     *  device. If \a devices argument is NULL, this argument is ignored.
-     *
-     *  \return One of the following values:
-     *    - CL_SUCCESS if the function is executed successfully.
-     *
-     *  The application can query specific capabilities of the OpenCL device(s)
-     *  returned by cl::getDevices. This can be used by the application to
-     *  determine which device(s) to use.
-     *
-     * \note In the case that exceptions are enabled and a return value
-     * other than CL_SUCCESS is generated, then cl::Error exception is
-     * generated.
-     */
-    cl_int getDevices(
-        cl_d3d10_device_source_khr d3d_device_source,
-        void *                     d3d_object,
-        cl_d3d10_device_set_khr    d3d_device_set,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clGetDeviceIDsFromD3D10KHR)(
-            cl_platform_id platform, 
-            cl_d3d10_device_source_khr d3d_device_source, 
-            void * d3d_object,
-            cl_d3d10_device_set_khr d3d_device_set,
-            cl_uint num_entries,
-            cl_device_id * devices,
-            cl_uint* num_devices);
-
-        static PFN_clGetDeviceIDsFromD3D10KHR pfn_clGetDeviceIDsFromD3D10KHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clGetDeviceIDsFromD3D10KHR);
-
-        cl_uint n = 0;
-        cl_int err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set, 
-            0, 
-            NULL, 
-            &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set,
-            n, 
-            ids, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-#endif
-
-    static cl_int get(
-        VECTOR_CLASS<Platform>* platforms)
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetPlatformIDs(0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        cl_platform_id* ids = (cl_platform_id*) alloca(
-            n * sizeof(cl_platform_id));
-        err = ::clGetPlatformIDs(n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        platforms->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-};
-
-static inline cl_int
-UnloadCompiler()
-{
-    return ::clUnloadCompiler();
-}
-
-class Context : public detail::Wrapper<cl_context>
-{
-public:
-    Context(
-        const VECTOR_CLASS<Device>& devices,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContext(
-            properties, (cl_uint) devices.size(),
-            (cl_device_id*) &devices.front(),
-            notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context(
-        cl_device_type type,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContextFromType(
-            properties, type, notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context() : detail::Wrapper<cl_type>() { }
-
-    Context(const Context& context) : detail::Wrapper<cl_type>(context) { }
-
-    Context& operator = (const Context& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_context_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetContextInfo, object_, name, param),
-            __GET_CONTEXT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_context_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_context_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getSupportedImageFormats(
-        cl_mem_flags flags,
-        cl_mem_object_type type,
-        VECTOR_CLASS<ImageFormat>* formats) const
-    {
-        cl_uint numEntries;
-        cl_int err = ::clGetSupportedImageFormats(
-           object_, 
-           flags,
-           type, 
-           0, 
-           NULL, 
-           &numEntries);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        ImageFormat* value = (ImageFormat*)
-            alloca(numEntries * sizeof(ImageFormat));
-        err = ::clGetSupportedImageFormats(
-            object_, 
-            flags, 
-            type, 
-            numEntries,
-            (cl_image_format*) value, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        formats->assign(&value[0], &value[numEntries]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Context)
-
-/*! \class Event
- * \brief Event interface for cl_event.
- */
-class Event : public detail::Wrapper<cl_event>
-{
-public:
-    Event() : detail::Wrapper<cl_type>() { }
-
-    Event(const Event& event) : detail::Wrapper<cl_type>(event) { }
-
-    Event& operator = (const Event& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_event_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetEventInfo, object_, name, param),
-            __GET_EVENT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_event_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_event_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getProfilingInfo(cl_profiling_info name, T* param) const
-    {
-        return detail::errHandler(detail::getInfo(
-            &::clGetEventProfilingInfo, object_, name, param),
-            __GET_EVENT_PROFILE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_profiling_info, name>::param_type
-    getProfilingInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_profiling_info, name>::param_type param;
-        cl_int result = getProfilingInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int wait() const
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(1, &object_),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setCallback(
-        cl_int type,
-        void (CL_CALLBACK * pfn_notify)(cl_event, cl_int, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetEventCallback(
-                object_,
-                type,
-                pfn_notify,
-                user_data), 
-            __SET_EVENT_CALLBACK_ERR);
-    }
-#endif
-
-    static cl_int
-    waitForEvents(const VECTOR_CLASS<Event>& events)
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(
-                (cl_uint) events.size(), (cl_event*)&events.front()),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Event)
-
-#if defined(CL_VERSION_1_1)
-/*! \class UserEvent
- * \brief User event interface for cl_event.
- */
-class UserEvent : public Event
-{
-public:
-    UserEvent(
-        const Context& context,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateUserEvent(
-            context(),
-            &error);
-
-        detail::errHandler(error, __CREATE_USER_EVENT_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    UserEvent() : Event() { }
-
-    UserEvent(const UserEvent& event) : Event(event) { }
-
-    UserEvent& operator = (const UserEvent& rhs)
-    {
-        if (this != &rhs) {
-            Event::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int setStatus(cl_int status)
-    {
-        return detail::errHandler(
-            ::clSetUserEventStatus(object_,status), 
-            __SET_USER_EVENT_STATUS_ERR);
-    }
-};
-#endif
-
-inline static cl_int
-WaitForEvents(const VECTOR_CLASS<Event>& events)
-{
-    return detail::errHandler(
-        ::clWaitForEvents(
-            (cl_uint) events.size(), (cl_event*)&events.front()),
-        __WAIT_FOR_EVENTS_ERR);
-}
-
-/*! \class Memory
- * \brief Memory interface for cl_mem.
- */
-class Memory : public detail::Wrapper<cl_mem>
-{
-public:
-    Memory() : detail::Wrapper<cl_type>() { }
-
-    Memory(const Memory& memory) : detail::Wrapper<cl_type>(memory) { }
-
-    Memory& operator = (const Memory& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_mem_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetMemObjectInfo, object_, name, param),
-            __GET_MEM_OBJECT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_mem_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_mem_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setDestructorCallback(
-        void (CL_CALLBACK * pfn_notify)(cl_mem, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetMemObjectDestructorCallback(
-                object_,
-                pfn_notify,
-                user_data), 
-            __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR);
-    }
-#endif
-
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Memory)
-
-/*! \class Buffer
- * \brief Memory buffer interface.
- */
-class Buffer : public Memory
-{
-public:
-    Buffer(
-        const Context& context,
-        cl_mem_flags flags,
-        ::size_t size,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateBuffer(context(), flags, size, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Buffer() : Memory() { }
-
-    Buffer(const Buffer& buffer) : Memory(buffer) { }
-
-    Buffer& operator = (const Buffer& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-
-#if defined(CL_VERSION_1_1)
-    Buffer createSubBuffer(
-        cl_mem_flags flags,
-        cl_buffer_create_type buffer_create_type,
-        const void * buffer_create_info,
-        cl_int * err = NULL)
-    {
-        Buffer result;
-        cl_int error;
-        result.object_ = ::clCreateSubBuffer(
-            object_, 
-            flags, 
-            buffer_create_type, 
-            buffer_create_info, 
-            &error);
-
-        detail::errHandler(error, __CREATE_SUBBUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-
-        return result;
-	}		
-#endif
-};
-
-#if defined (USE_DX_INTEROP)
-class BufferD3D10 : public Buffer
-{
-public:
-    typedef CL_API_ENTRY cl_mem (CL_API_CALL *PFN_clCreateFromD3D10BufferKHR)(
-    cl_context context, cl_mem_flags flags, ID3D10Buffer*  buffer,
-    cl_int* errcode_ret);
-
-    BufferD3D10(
-        const Context& context,
-        cl_mem_flags flags,
-        ID3D10Buffer* bufobj,
-        cl_int * err = NULL)
-    {
-        static PFN_clCreateFromD3D10BufferKHR pfn_clCreateFromD3D10BufferKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clCreateFromD3D10BufferKHR);
-
-        cl_int error;
-        object_ = pfn_clCreateFromD3D10BufferKHR(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferD3D10() : Buffer() { }
-
-    BufferD3D10(const BufferD3D10& buffer) : Buffer(buffer) { }
-
-    BufferD3D10& operator = (const BufferD3D10& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-};
-#endif
-
-/*! \class BufferGL
- * \brief Memory buffer interface for GL interop.
- */
-class BufferGL : public Buffer
-{
-public:
-    BufferGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLBuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferGL() : Buffer() { }
-
-    BufferGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferGL& operator = (const BufferGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class BufferRenderGL
- * \brief Memory buffer interface for GL interop with renderbuffer.
- */
-class BufferRenderGL : public Buffer
-{
-public:
-    BufferRenderGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLRenderbuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferRenderGL() : Buffer() { }
-
-    BufferRenderGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferRenderGL& operator = (const BufferRenderGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class Image
- * \brief Base class  interface for all images.
- */
-class Image : public Memory
-{
-protected:
-    Image() : Memory() { }
-
-    Image(const Image& image) : Memory(image) { }
-
-    Image& operator = (const Image& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-public:
-    template <typename T>
-    cl_int getImageInfo(cl_image_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetImageInfo, object_, name, param),
-            __GET_IMAGE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_image_info, name>::param_type
-    getImageInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_image_info, name>::param_type param;
-        cl_int result = getImageInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-/*! \class Image2D
- * \brief Image interface for 2D images.
- */
-class Image2D : public Image
-{
-public:
-    Image2D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t row_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage2D(
-            context(), flags,&format, width, height, row_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE2D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2D() { }
-
-    Image2D(const Image2D& image2D) : Image(image2D) { }
-
-    Image2D& operator = (const Image2D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image2DGL : public Image2D
-{
-public:
-    Image2DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture2D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2DGL() : Image2D() { }
-
-    Image2DGL(const Image2DGL& image) : Image2D(image) { }
-
-    Image2DGL& operator = (const Image2DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image2D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image3D
- * \brief Image interface for 3D images.
- */
-class Image3D : public Image
-{
-public:
-    Image3D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t depth,
-        ::size_t row_pitch = 0,
-        ::size_t slice_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage3D(
-            context(), flags, &format, width, height, depth, row_pitch,
-            slice_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE3D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3D() { }
-
-    Image3D(const Image3D& image3D) : Image(image3D) { }
-
-    Image3D& operator = (const Image3D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image3DGL : public Image3D
-{
-public:
-    Image3DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture3D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3DGL() : Image3D() { }
-
-    Image3DGL(const Image3DGL& image) : Image3D(image) { }
-
-    Image3DGL& operator = (const Image3DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image3D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Sampler
- * \brief Sampler interface for cl_sampler.
- */
-class Sampler : public detail::Wrapper<cl_sampler>
-{
-public:
-    Sampler() { }
-
-    Sampler(
-        const Context& context,
-        cl_bool normalized_coords,
-        cl_addressing_mode addressing_mode,
-        cl_filter_mode filter_mode,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateSampler(
-            context(), 
-            normalized_coords,
-            addressing_mode,
-            filter_mode,
-            &error);
-
-        detail::errHandler(error, __CREATE_SAMPLER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Sampler(const Sampler& sampler) : detail::Wrapper<cl_type>(sampler) { }
-
-    Sampler& operator = (const Sampler& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_sampler_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetSamplerInfo, object_, name, param),
-            __GET_SAMPLER_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_sampler_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_sampler_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Sampler)
-
-class Program;
-class CommandQueue;
-class Kernel;
-
-/*! \class NDRange
- * \brief NDRange interface
- */
-class NDRange
-{
-private:
-    size_t<3> sizes_;
-    cl_uint dimensions_;
-
-public:
-    NDRange()
-        : dimensions_(0)
-    { }
-
-    NDRange(::size_t size0)
-        : dimensions_(1)
-    {
-        sizes_.push_back(size0);
-    }
-
-    NDRange(::size_t size0, ::size_t size1)
-        : dimensions_(2)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-    }
-
-    NDRange(::size_t size0, ::size_t size1, ::size_t size2)
-        : dimensions_(3)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-        sizes_.push_back(size2);
-    }
-
-    operator const ::size_t*() const { return (const ::size_t*) sizes_; }
-    ::size_t dimensions() const { return dimensions_; }
-};
-
-static const NDRange NullRange;
-
-/*!
- * \struct LocalSpaceArg
- * \brief Local address raper for use with Kernel::setArg
- */
-struct LocalSpaceArg
-{
-    ::size_t size_;
-};
-
-namespace detail {
-
-template <typename T>
-struct KernelArgumentHandler
-{
-    static ::size_t size(const T&) { return sizeof(T); }
-    static T* ptr(T& value) { return &value; }
-};
-
-template <>
-struct KernelArgumentHandler<LocalSpaceArg>
-{
-    static ::size_t size(const LocalSpaceArg& value) { return value.size_; }
-    static void* ptr(LocalSpaceArg&) { return NULL; }
-};
-
-} 
-//! \endcond
-
-inline LocalSpaceArg
-__local(::size_t size)
-{
-    LocalSpaceArg ret = { size };
-    return ret;
-}
-
-class KernelFunctor;
-
-/*! \class Kernel
- * \brief Kernel interface that implements cl_kernel
- */
-class Kernel : public detail::Wrapper<cl_kernel>
-{
-public:
-    inline Kernel(const Program& program, const char* name, cl_int* err = NULL);
-
-    Kernel() { }
-
-    Kernel(const Kernel& kernel) : detail::Wrapper<cl_type>(kernel) { }
-
-    Kernel& operator = (const Kernel& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_kernel_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetKernelInfo, object_, name, param),
-            __GET_KERNEL_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_kernel_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getWorkGroupInfo(
-        const Device& device, cl_kernel_work_group_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetKernelWorkGroupInfo, object_, device(), name, param),
-                __GET_KERNEL_WORK_GROUP_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_work_group_info, name>::param_type
-        getWorkGroupInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-        detail::cl_kernel_work_group_info, name>::param_type param;
-        cl_int result = getWorkGroupInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int setArg(cl_uint index, T value)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(
-                object_,
-                index,
-                detail::KernelArgumentHandler<T>::size(value),
-                detail::KernelArgumentHandler<T>::ptr(value)),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    cl_int setArg(cl_uint index, ::size_t size, void* argPtr)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(object_, index, size, argPtr),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local);
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& global,
-        const NDRange& local);
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Kernel)
-
-/*! \class Program
- * \brief Program interface that implements cl_program.
- */
-class Program : public detail::Wrapper<cl_program>
-{
-public:
-    typedef VECTOR_CLASS<std::pair<const void*, ::size_t> > Binaries;
-    typedef VECTOR_CLASS<std::pair<const char*, ::size_t> > Sources;
-
-    Program(
-        const Context& context,
-        const Sources& sources,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-
-        const ::size_t n = (::size_t)sources.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const char** strings = (const char**) alloca(n * sizeof(const char*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            strings[i] = sources[(int)i].first;
-            lengths[i] = sources[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithSource(
-            context(), (cl_uint)n, strings, lengths, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_SOURCE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program(
-        const Context& context,
-        const VECTOR_CLASS<Device>& devices,
-        const Binaries& binaries,
-        VECTOR_CLASS<cl_int>* binaryStatus = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        const ::size_t n = binaries.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const unsigned char** images = (const unsigned char**) alloca(n * sizeof(const void*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            images[i] = (const unsigned char*)binaries[(int)i].first;
-            lengths[i] = binaries[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithBinary(
-            context(), (cl_uint) devices.size(),
-            (cl_device_id*)&devices.front(),
-            lengths, images, binaryStatus != NULL
-               ? (cl_int*) &binaryStatus->front()
-               : NULL, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_BINARY_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program() { }
-
-    Program(const Program& program) : detail::Wrapper<cl_type>(program) { }
-
-    Program& operator = (const Program& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int build(
-        const VECTOR_CLASS<Device>& devices,
-        const char* options = NULL,
-        void (CL_CALLBACK * notifyFptr)(cl_program, void *) = NULL,
-        void* data = NULL) const
-    {
-        return detail::errHandler(
-            ::clBuildProgram(
-                object_,
-                (cl_uint)
-                devices.size(),
-                (cl_device_id*)&devices.front(),
-                options,
-                notifyFptr,
-                data),
-                __BUILD_PROGRAM_ERR);
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_program_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetProgramInfo, object_, name, param),
-            __GET_PROGRAM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getBuildInfo(
-        const Device& device, cl_program_build_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetProgramBuildInfo, object_, device(), name, param),
-                __GET_PROGRAM_BUILD_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_build_info, name>::param_type
-    getBuildInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_build_info, name>::param_type param;
-        cl_int result = getBuildInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int createKernels(VECTOR_CLASS<Kernel>* kernels)
-    {
-        cl_uint numKernels;
-        cl_int err = ::clCreateKernelsInProgram(object_, 0, NULL, &numKernels);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        Kernel* value = (Kernel*) alloca(numKernels * sizeof(Kernel));
-        err = ::clCreateKernelsInProgram(
-            object_, numKernels, (cl_kernel*) value, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        kernels->assign(&value[0], &value[numKernels]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Program)
-
-inline Kernel::Kernel(const Program& program, const char* name, cl_int* err)
-{
-    cl_int error;
-
-    object_ = ::clCreateKernel(program(), name, &error);
-    detail::errHandler(error, __CREATE_KERNEL_ERR);
-
-    if (err != NULL) {
-        *err = error;
-    }
-
-}
-
-/*! \class CommandQueue
- * \brief CommandQueue interface for cl_command_queue.
- */
-class CommandQueue : public detail::Wrapper<cl_command_queue>
-{
-public:
-    CommandQueue(
-        const Context& context,
-        const Device& device,
-        cl_command_queue_properties properties = 0,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateCommandQueue(
-            context(), device(), properties, &error);
-
-        detail::errHandler(error, __CREATE_COMMAND_QUEUE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    CommandQueue() { }
-
-    CommandQueue(const CommandQueue& commandQueue) : detail::Wrapper<cl_type>(commandQueue) { }
-
-    CommandQueue& operator = (const CommandQueue& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_command_queue_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetCommandQueueInfo, object_, name, param),
-                __GET_COMMAND_QUEUE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_command_queue_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_command_queue_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int enqueueReadBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_BUFFER_ERR);
-    }
-
-    cl_int enqueueWriteBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        const void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBuffer(
-        const Buffer& src,
-        const Buffer& dst,
-        ::size_t src_offset,
-        ::size_t dst_offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBuffer(
-                object_, src(), dst(), src_offset, dst_offset, size,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int enqueueReadBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_READ_BUFFER_RECT_ERR);
-    }
-
-
-    cl_int enqueueWriteBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_RECT_ERR);
-    }
-
-    cl_int enqueueCopyBufferRect(
-        const Buffer& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        ::size_t src_row_pitch,
-        ::size_t src_slice_pitch,
-        ::size_t dst_row_pitch,
-        ::size_t dst_slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferRect(
-                object_, 
-                src(), 
-                dst(), 
-                (const ::size_t *)src_origin, 
-                (const ::size_t *)dst_origin, 
-                (const ::size_t *)region,
-                src_row_pitch,
-                src_slice_pitch,
-                dst_row_pitch,
-                dst_slice_pitch,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_RECT_ERR);
-    }
-#endif
-
-    cl_int enqueueReadImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_IMAGE_ERR);
-    }
-
-    cl_int enqueueWriteImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_WRITE_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImage(
-        const Image& src,
-        const Image& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImage(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *)dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImageToBuffer(
-        const Image& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& region,
-        ::size_t dst_offset,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImageToBuffer(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *) region, dst_offset,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBufferToImage(
-        const Buffer& src,
-        const Image& dst,
-        ::size_t src_offset,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferToImage(
-                object_, src(), dst(), src_offset,
-                (const ::size_t *) dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR);
-    }
-
-    void* enqueueMapBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        ::size_t offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapBuffer(
-            object_, buffer(), blocking, flags, offset, size,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-        return result;
-    }
-
-    void* enqueueMapImage(
-        const Image& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t * row_pitch,
-        ::size_t * slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapImage(
-            object_, buffer(), blocking, flags,
-            (const ::size_t *) origin, (const ::size_t *) region,
-            row_pitch, slice_pitch,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_IMAGE_ERR);
-        if (err != NULL) {
-              *err = error;
-        }
-        return result;
-    }
-
-    cl_int enqueueUnmapMemObject(
-        const Memory& memory,
-        void* mapped_ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueUnmapMemObject(
-                object_, memory(), mapped_ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_UNMAP_MEM_OBJECT_ERR);
-    }
-
-    cl_int enqueueNDRangeKernel(
-        const Kernel& kernel,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueNDRangeKernel(
-                object_, kernel(), (cl_uint) global.dimensions(),
-                offset.dimensions() != 0 ? (const ::size_t*) offset : NULL,
-                (const ::size_t*) global,
-                local.dimensions() != 0 ? (const ::size_t*) local : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NDRANGE_KERNEL_ERR);
-    }
-
-    cl_int enqueueTask(
-        const Kernel& kernel,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueTask(
-                object_, kernel(),
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_TASK_ERR);
-    }
-
-    cl_int enqueueNativeKernel(
-        void (*userFptr)(void *),
-        std::pair<void*, ::size_t> args,
-        const VECTOR_CLASS<Memory>* mem_objects = NULL,
-        const VECTOR_CLASS<const void*>* mem_locs = NULL,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        cl_mem * mems = (mem_objects != NULL && mem_objects->size() > 0) 
-            ? (cl_mem*) alloca(mem_objects->size() * sizeof(cl_mem))
-            : NULL;
-
-        if (mems != NULL) {
-            for (unsigned int i = 0; i < mem_objects->size(); i++) {
-                mems[i] = ((*mem_objects)[i])();
-            }
-        }
-
-        return detail::errHandler(
-            ::clEnqueueNativeKernel(
-                object_, userFptr, args.first, args.second,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                mems,
-                (mem_locs != NULL) ? (const void **) &mem_locs->front() : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NATIVE_KERNEL);
-    }
-
-    cl_int enqueueMarker(Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueMarker(object_, (cl_event*) event),
-            __ENQUEUE_MARKER_ERR);
-    }
-
-    cl_int enqueueWaitForEvents(const VECTOR_CLASS<Event>& events) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWaitForEvents(
-                object_,
-                (cl_uint) events.size(),
-                (const cl_event*) &events.front()),
-            __ENQUEUE_WAIT_FOR_EVENTS_ERR);
-    }
-
-    cl_int enqueueAcquireGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueAcquireGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueReleaseGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_RELEASE_GL_ERR);
-     }
-
-#if defined (USE_DX_INTEROP)
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueAcquireD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects, cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueReleaseD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects,  cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-
-    cl_int enqueueAcquireD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         static PFN_clEnqueueAcquireD3D10ObjectsKHR pfn_clEnqueueAcquireD3D10ObjectsKHR = NULL;
-         __INIT_CL_EXT_FCN_PTR(clEnqueueAcquireD3D10ObjectsKHR);
-		
-         return detail::errHandler(
-             pfn_clEnqueueAcquireD3D10ObjectsKHR(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-    {
-        static PFN_clEnqueueReleaseD3D10ObjectsKHR pfn_clEnqueueReleaseD3D10ObjectsKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clEnqueueReleaseD3D10ObjectsKHR);
-
-        return detail::errHandler(
-            pfn_clEnqueueReleaseD3D10ObjectsKHR(
-                object_,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_RELEASE_GL_ERR);
-    }
-#endif
-
-    cl_int enqueueBarrier() const
-    {
-        return detail::errHandler(
-            ::clEnqueueBarrier(object_),
-            __ENQUEUE_BARRIER_ERR);
-    }
-
-    cl_int flush() const
-    {
-        return detail::errHandler(::clFlush(object_), __FLUSH_ERR);
-    }
-
-    cl_int finish() const
-    {
-        return detail::errHandler(::clFinish(object_), __FINISH_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::CommandQueue)
-
-/*! \class KernelFunctor
- * \brief Kernel functor interface
- *
- * \note Currently only functors of zero to ten arguments are supported. It
- * is straightforward to add more and a more general solution, similar to
- * Boost.Lambda could be followed if required in the future.
- */
-class KernelFunctor
-{
-private:
-    Kernel kernel_;
-    CommandQueue queue_;
-    NDRange offset_;
-    NDRange global_;
-    NDRange local_;
-
-    cl_int err_;
-public:
-    KernelFunctor() { }
-
-    KernelFunctor(
-        const Kernel& kernel,
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local) :
-            kernel_(kernel),
-            queue_(queue),
-            offset_(offset),
-            global_(global),
-            local_(local),
-            err_(CL_SUCCESS)
-    {}
-
-    KernelFunctor& operator=(const KernelFunctor& rhs);
-
-    KernelFunctor(const KernelFunctor& rhs);
-
-    cl_int getError() { return err_; }
-
-    inline Event operator()(const VECTOR_CLASS<Event>* events = NULL);
-
-    template<typename A1>
-    inline Event operator()(
-        const A1& a1, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5, class A6>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4,
-             class A5, class A6, class A7>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12, 
-        const A13& a13,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14, class A15>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14, 
-        const A15& a15,
-        const VECTOR_CLASS<Event>* events = NULL);
-};
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& offset,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,offset,global,local);
-}
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,NullRange,global,local);
-}
-
-inline KernelFunctor& KernelFunctor::operator=(const KernelFunctor& rhs)
-{
-    if (this == &rhs) {
-        return *this;
-    }
-    
-    kernel_ = rhs.kernel_;
-    queue_  = rhs.queue_;
-    offset_ = rhs.offset_;
-    global_ = rhs.global_;
-    local_  = rhs.local_;
-    
-    return *this;
-}
-
-inline KernelFunctor::KernelFunctor(const KernelFunctor& rhs) :
-    kernel_(rhs.kernel_),
-    queue_(rhs.queue_),
-    offset_(rhs.offset_),
-    global_(rhs.global_),
-    local_(rhs.local_)
-{
-}
-
-Event KernelFunctor::operator()(const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4,
-         typename A5, typename A6, typename A7>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9, typename A10>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12, 
-    const A13& a13,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14, class A15>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14, 
-    const A15& a15,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-    kernel_.setArg(14,a15);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-#undef __ERR_STR
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#undef __GET_DEVICE_INFO_ERR
-#undef __GET_PLATFORM_INFO_ERR
-#undef __GET_DEVICE_IDS_ERR
-#undef __GET_CONTEXT_INFO_ERR
-#undef __GET_EVENT_INFO_ERR
-#undef __GET_EVENT_PROFILE_INFO_ERR
-#undef __GET_MEM_OBJECT_INFO_ERR
-#undef __GET_IMAGE_INFO_ERR
-#undef __GET_SAMPLER_INFO_ERR
-#undef __GET_KERNEL_INFO_ERR
-#undef __GET_KERNEL_WORK_GROUP_INFO_ERR
-#undef __GET_PROGRAM_INFO_ERR
-#undef __GET_PROGRAM_BUILD_INFO_ERR
-#undef __GET_COMMAND_QUEUE_INFO_ERR
-
-#undef __CREATE_CONTEXT_FROM_TYPE_ERR
-#undef __GET_SUPPORTED_IMAGE_FORMATS_ERR
-
-#undef __CREATE_BUFFER_ERR
-#undef __CREATE_SUBBUFFER_ERR
-#undef __CREATE_IMAGE2D_ERR
-#undef __CREATE_IMAGE3D_ERR
-#undef __CREATE_SAMPLER_ERR
-#undef __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR
-
-#undef __CREATE_USER_EVENT_ERR
-#undef __SET_USER_EVENT_STATUS_ERR
-#undef __SET_EVENT_CALLBACK_ERR
-
-#undef __WAIT_FOR_EVENTS_ERR
-
-#undef __CREATE_KERNEL_ERR
-#undef __SET_KERNEL_ARGS_ERR
-#undef __CREATE_PROGRAM_WITH_SOURCE_ERR
-#undef __CREATE_PROGRAM_WITH_BINARY_ERR
-#undef __BUILD_PROGRAM_ERR
-#undef __CREATE_KERNELS_IN_PROGRAM_ERR
-
-#undef __CREATE_COMMAND_QUEUE_ERR
-#undef __SET_COMMAND_QUEUE_PROPERTY_ERR
-#undef __ENQUEUE_READ_BUFFER_ERR
-#undef __ENQUEUE_WRITE_BUFFER_ERR
-#undef __ENQUEUE_READ_BUFFER_RECT_ERR
-#undef __ENQUEUE_WRITE_BUFFER_RECT_ERR
-#undef __ENQEUE_COPY_BUFFER_ERR
-#undef __ENQEUE_COPY_BUFFER_RECT_ERR
-#undef __ENQUEUE_READ_IMAGE_ERR
-#undef __ENQUEUE_WRITE_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR
-#undef __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR
-#undef __ENQUEUE_MAP_BUFFER_ERR
-#undef __ENQUEUE_MAP_IMAGE_ERR
-#undef __ENQUEUE_UNMAP_MEM_OBJECT_ERR
-#undef __ENQUEUE_NDRANGE_KERNEL_ERR
-#undef __ENQUEUE_TASK_ERR
-#undef __ENQUEUE_NATIVE_KERNEL
-
-#undef __UNLOAD_COMPILER_ERR
-#endif //__CL_USER_OVERRIDE_ERROR_STRINGS
-
-#undef __GET_INFO_HELPER_WITH_RETAIN
-
-// Extensions
-#undef __INIT_CL_EXT_FCN_PTR
-#undef __CREATE_SUB_DEVICES
-
-#if defined(USE_CL_DEVICE_FISSION)
-#undef __PARAM_NAME_DEVICE_FISSION
-#endif // USE_CL_DEVICE_FISSION
-
-} // namespace cl
-
-#endif // CL_HPP_
diff --git a/SpeedComparisons/GrayScott_OpenCL_Local/gray_scott_opencl_local.cpp b/SpeedComparisons/GrayScott_OpenCL_Local/gray_scott_opencl_local.cpp
deleted file mode 100644
index 47fc811de..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Local/gray_scott_opencl_local.cpp
+++ /dev/null
@@ -1,273 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCL:
-#define __NO_STD_VECTOR // Use cl::vector instead of STL version
-#define __CL_ENABLE_EXCEPTIONS
-
-// cl.hpp is standard but doesn't come with most SDKs, so download it from here:
-// http://www.khronos.org/registry/cl/api/1.1/cl.hpp
-#ifdef __APPLE__
-# include "cl.hpp"
-#else
-# include <CL/cl.hpp>
-#endif
-
-using namespace cl;
-
-// STL:
-#include <fstream>
-#include <iostream>
-#include <sstream>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-static int g_opt_device = 0;
-static int g_wrap = 1;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if ((i+1<argc) && (strcmp(argv[i],"-device")==0)) {
-            // select an output device
-            i++; g_opt_device = atoi(argv[i]);
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            std::cout << "Unrecognized argument: '" << argv[i] << "'\n";
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    float k, f;
-     k = 0.064f; f = 0.035f; // solitons with mitosis (spots that multiply)
-    // k = 0.06f; f = 0.035f; // stripes
-    // k = 0.065f; f = 0.056f; // long stripes
-    // k = 0.064f; f = 0.04f; // dots and stripes
-    // k = 0.0475f; f = 0.0118f; // spiral waves
-    float speed = 2.0f;
-    // ----------------
-    
-    float a[X][Y], b[X][Y];
-    const int MEM_SIZE = sizeof(float)*X*Y;
-
-    // put the initial conditions into each cell
-    init(a,b);
-
-    try { 
-        // Get available OpenCL platforms
-        vector<Platform> platforms;
-        Platform::get(&platforms);
- 
-        // Select the default platform and create a context using this platform and the GPU
-        cl_context_properties cps[3] = { 
-            CL_CONTEXT_PLATFORM, 
-            (cl_context_properties)(platforms[0])(), 
-            0 
-        };
-        Context context( CL_DEVICE_TYPE_GPU, cps);
- 
-        // Get a list of devices on this platform
-        vector<Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
-
-        // range-check the user's selection
-        int maxdev = devices.size() - 1;
-        g_opt_device = (g_opt_device > maxdev) ? maxdev :
-                                     ((g_opt_device < 0) ? 0 : g_opt_device);
-        std::cout << (maxdev+1) << " device(s) available; using device "
-                                                    << g_opt_device << ".\n";
-
-        Device &device = devices[g_opt_device];
-        std::cout << "Global memory: " << device.getInfo<CL_DEVICE_GLOBAL_MEM_SIZE>() << " bytes\n";
-        std::cout << "Local memory: " << device.getInfo<CL_DEVICE_LOCAL_MEM_SIZE>() << " bytes\n";
-        std::cout << "Local memory type: " << std::string((device.getInfo<CL_DEVICE_LOCAL_MEM_TYPE>()==CL_LOCAL)?"local":"global") << " \n";
-
-        // Create a command queue and use the selected device
-        if (maxdev < 0) {
-          std::cerr << "error -- need at least one OpenCL capable device.\n";
-          exit(-1);
-        }
-        CommandQueue queue = CommandQueue(context, device);
-        Event event;
- 
-        // Read source file
-        std::string kfn = CL_SOURCE_DIR; // (defined in CMakeLists.txt to be the source folder)
-        kfn += "/grayscott_kernel_local.cl";
-        std::ifstream sourceFile(kfn.c_str());
-        std::string sourceCode(
-            std::istreambuf_iterator<char>(sourceFile),
-            (std::istreambuf_iterator<char>()));
-        Program::Sources source(1, std::make_pair(sourceCode.c_str(), sourceCode.length()+1));
- 
-        // Make program of the source code in the context
-        Program program = Program(context, source);
-
-        const int LOCAL_X=1;
-        const int LOCAL_Y=256;
- 
-        // Build program for these specific devices
-        std::ostringstream oss;
-        oss << "-D LOCAL_X=" << LOCAL_X << " -D LOCAL_Y=" << LOCAL_Y;
-        if(g_wrap) oss << " -D WRAP";
-        program.build(devices, oss.str().c_str(), NULL, NULL);
- 
-        // Make kernel
-        Kernel kernel(program, "grayscott_compute");
-
-        // Create memory buffers
-        Buffer bufferU = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferV = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferU2 = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferV2 = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-
-        // Copy lists A and B to the memory buffers
-        queue.enqueueWriteBuffer(bufferU, CL_TRUE, 0, MEM_SIZE, a);
-        queue.enqueueWriteBuffer(bufferV, CL_TRUE, 0, MEM_SIZE, b);
- 
-        NDRange global(X,Y);
-        NDRange local(LOCAL_X,LOCAL_Y);
-
-        kernel.setArg(4, k);
-        kernel.setArg(5, f);
-        kernel.setArg(6, r_a);
-        kernel.setArg(7, r_b);
-        kernel.setArg(8, speed);
-
-        int iteration = 0;
-        float fps_avg = 0.0; // decaying average of fps
-        const int N_FRAMES_PER_DISPLAY = 2000;  // an even number, because of our double-buffering implementation
-        while(true) 
-        {
-            struct timeval tod_record;
-            double tod_before, tod_after, tod_elap;
-
-            gettimeofday(&tod_record, 0);
-            tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            // run a few iterations (without copying the data back)
-            for(int it=0;it<N_FRAMES_PER_DISPLAY/2;it++)
-            {
-                // (buffer-switching)
-
-                kernel.setArg(0, bufferU);
-                kernel.setArg(1, bufferV);
-                kernel.setArg(2, bufferU2);
-                kernel.setArg(3, bufferV2); // output in A2,B2
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-
-                kernel.setArg(0, bufferU2);
-                kernel.setArg(1, bufferV2);
-                kernel.setArg(2, bufferU);
-                kernel.setArg(3, bufferV); // output in A,B
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-            }
-
-            // read the A pattern back into a buffer
-            queue.enqueueReadBuffer(bufferU, CL_TRUE, 0, MEM_SIZE, a);
-
-            gettimeofday(&tod_record, 0);
-            tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            tod_elap = tod_after - tod_before;
-
-            char msg[1000];
-            float fps = 0.0;     // frames per second
-            if (tod_elap > 0)
-                fps = ((float)N_FRAMES_PER_DISPLAY) / tod_elap;
-            // We display an exponential moving average of the fps measurement
-            fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-	        double Mcgs = (fps_avg * ((double)X) * ((double)Y)) / 1.0e6;
-            sprintf(msg,"GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-            // display:
-            {
-                int quitnow = display(a,a,a,iteration,false,200.0f,2,10,msg);
-                if (quitnow)
-                    break;
-            }
-        }
-    } 
-    catch(Error error) 
-    {
-       std::cout << error.what() << "(" << error.err() << ")" << std::endl;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            // start with a uniform field with an approximate circle in the middle
-            //if(hypot(i%20-10/*-X/2*/,j%20-10/*-Y/2*/)<=frand(2,5)) {
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5))
-            {
-                a[i][j] = frand(0.0f,0.1f);
-                b[i][j] = frand(0.9f,1.0f);
-            }
-            else 
-            {
-                a[i][j] = frand(0.9f,1.0f);
-                b[i][j] = frand(0.0f,0.1f);
-            }
-            /*float v = frand(0.0f,1.0f);
-            a[i][j] = v;
-            b[i][j] = 1.0f-v;*/
-        }
-    }
-}
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_Local/grayscott_kernel_local.cl b/SpeedComparisons/GrayScott_OpenCL_Local/grayscott_kernel_local.cl
deleted file mode 100644
index b5e4a3388..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_Local/grayscott_kernel_local.cl
+++ /dev/null
@@ -1,80 +0,0 @@
-//  This version loads data into a local cache, see http://www.khronos.org/message_boards/viewtopic.php?p=11081#p11081
-
-__kernel void grayscott_compute(
-    __global float *U,__global float *V,
-    __global float *U2, __global float *V2,
-    float k,float F,float D_u,float D_v,float delta_t)
-{
-    // Get the index of the current element. X and Y are oriented like
-    // in high school math class, with the origin (0,0) in the bottom-left
-    // corner.
-    const int x = get_global_id(0);   // column (0=leftmost)
-    const int y = get_global_id(1);   // row (0=bottom-most)
-    const int X = get_global_size(0); // number of columns
-    const int Y = get_global_size(1); // number of rows
-    const int i = x*Y+y; // column * number_of_rows + row
-
-    // make a local cache of nearby data, for speed    
-    const int local_x = get_local_id(0);
-    const int local_y = get_local_id(1);
-    __local float local_U[LOCAL_X][LOCAL_Y];
-    __local float local_V[LOCAL_X][LOCAL_Y];
-    local_U[local_x][local_y] = U[i];
-    local_V[local_x][local_y] = V[i];
-    barrier(CLK_LOCAL_MEM_FENCE);  // wait until all the local threads have copied their data into the local cache
-    
-    const float u = U[i]; // get U
-    const float v = V[i]; // get V
-    
-    if(local_x>0 && local_y>0 && local_x<LOCAL_X-1 && local_y<LOCAL_Y-1)
-    {
-        // faster version, can use local cache
-        
-        // compute the Laplacians of a and b
-    
-        // Standard 5-point stencil
-        const float nabla_u = local_U[local_x-1][local_y] + local_U[local_x+1][local_y] + local_U[local_x][local_y-1] + local_U[local_x][local_y+1] - 4*u;
-        const float nabla_v = local_V[local_x-1][local_y] + local_V[local_x+1][local_y] + local_V[local_x][local_y-1] + local_V[local_x][local_y+1] - 4*v;
-
-        // compute the new rate of change
-        const float delta_u = D_u * nabla_u - u*v*v + F*(1.0f-u);
-        const float delta_v = D_v * nabla_v + u*v*v - (F+k)*v;
-
-        // apply the change (to the new buffer)
-        U2[i] = u + delta_t * delta_u;
-        V2[i] = v + delta_t * delta_v;
-    }
-    else
-    {
-        // slower version, must use global data
-
-        // compute the Laplacians of a and b
-#ifdef WRAP
-        const int xm1 = ((x-1+X)%X);
-        const int xp1 = ((x+1)%X);
-        const int ym1 = ((y-1+Y)%Y);
-        const int yp1 = ((y+1)%Y);
-#else
-        const int xm1 = max(x-1,0);
-        const int xp1 = min(x+1,X-1);
-        const int ym1 = max(y-1,0);
-        const int yp1 = min(y+1,Y-1);
-#endif
-        const int iLeft = xm1*Y + y;
-        const int iRight = xp1*Y + y;
-        const int iUp = x*Y + ym1;   // Actually down: y=0 is the bottom edge
-        const int iDown = x*Y + yp1; // Actually up
-    
-        // Standard 5-point stencil
-        const float nabla_u = U[iLeft] + U[iRight] + U[iUp] + U[iDown] - 4*u;
-        const float nabla_v = V[iLeft] + V[iRight] + V[iUp] + V[iDown] - 4*v;
-
-        // compute the new rate of change
-        const float delta_u = D_u * nabla_u - u*v*v + F*(1.0f-u);
-        const float delta_v = D_v * nabla_v + u*v*v - (F+k)*v;
-
-        // apply the change (to the new buffer)
-        U2[i] = u + delta_t * delta_u;
-        V2[i] = v + delta_t * delta_v;
-    }
-}
diff --git a/SpeedComparisons/GrayScott_OpenCL_float2/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenCL_float2/CMakeLists.txt
deleted file mode 100644
index 80ed5a45c..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_float2/CMakeLists.txt
+++ /dev/null
@@ -1,27 +0,0 @@
-project(GrayScott_OpenCL_float2)
-
-set(CMAKE_MODULE_PATH ${GrayScott_OpenCL_float2_SOURCE_DIR})
-# (we include our own FindOpenCL.cmake until the time that CMake comes with its own)
-
-find_package(OpenCV REQUIRED)
-include_directories( ${OPENCV_INCLUDE_DIR})
-link_libraries( ${OpenCV_LIBS} )
-
-# only build the OpenCL version if OpenCL was found
-find_package ( OpenCL )
-if(OPENCL_FOUND)
-  include_directories ( ${OPENCL_INCLUDE_DIRS} )
-  link_libraries ( ${OPENCL_LIBRARIES} )
-
-  # tell the code where the .cl file will live
-  add_definitions(-DCL_SOURCE_DIR="${GrayScott_OpenCL_float2_SOURCE_DIR}")
-
-  INCLUDE_DIRECTORIES( "../Display" )
-
-  add_executable(GrayScott_OpenCL_float2
-    gray_scott_opencl_float2.cpp 
-    grayscott_kernel_float2.cl 
-    ../Display/defs.h
-  )
-endif()
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_float2/FindOpenCL.cmake b/SpeedComparisons/GrayScott_OpenCL_float2/FindOpenCL.cmake
deleted file mode 100644
index fde90efae..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_float2/FindOpenCL.cmake
+++ /dev/null
@@ -1,79 +0,0 @@
-# - Try to find OpenCL
-# This module tries to find an OpenCL implementation on your system. It supports
-# AMD / ATI, Apple and NVIDIA implementations, but shoudl work, too.
-#
-# Once done this will define
-#  OPENCL_FOUND        - system has OpenCL
-#  OPENCL_INCLUDE_DIRS  - the OpenCL include directory
-#  OPENCL_LIBRARIES    - link these to use OpenCL
-#
-# WIN32 should work, but is untested
-
-FIND_PACKAGE( PackageHandleStandardArgs )
-
-SET (OPENCL_VERSION_STRING "0.1.0")
-SET (OPENCL_VERSION_MAJOR 0)
-SET (OPENCL_VERSION_MINOR 1)
-SET (OPENCL_VERSION_PATCH 0)
-
-IF (APPLE)
-
-  FIND_LIBRARY(OPENCL_LIBRARIES OpenCL DOC "OpenCL lib for OSX")
-  FIND_PATH(OPENCL_INCLUDE_DIRS OpenCL/cl.h DOC "Include for OpenCL on OSX")
-  FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS OpenCL/cl.hpp DOC "Include for OpenCL CPP bindings on OSX")
-
-ELSE (APPLE)
-
-	IF (WIN32)
-	
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h)
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp)
-	
-	    # The AMD SDK currently installs both x86 and x86_64 libraries
-	    # This is only a hack to find out architecture
-	    IF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86_64")
-	    ELSE (${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64")
-	    	SET(OPENCL_LIB_DIR "$ENV{ATISTREAMSDKROOT}/lib/x86")
-	    ENDIF( ${CMAKE_SYSTEM_PROCESSOR} STREQUAL "AMD64" )
-	    FIND_LIBRARY(OPENCL_LIBRARIES OpenCL.lib ${OPENCL_LIB_DIR})
-	    
-	    GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-	    
-	    # On Win32 search relative to the library
-	    FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS "${_OPENCL_INC_CAND}")
-	    FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS "${_OPENCL_INC_CAND}")
-	
-	ELSE (WIN32)
-
-            # Unix style platforms
-            FIND_LIBRARY(OPENCL_LIBRARIES OpenCL
-              ENV LD_LIBRARY_PATH
-            )
-
-            GET_FILENAME_COMPONENT(OPENCL_LIB_DIR ${OPENCL_LIBRARIES} PATH)
-            GET_FILENAME_COMPONENT(_OPENCL_INC_CAND ${OPENCL_LIB_DIR}/../../include ABSOLUTE)
-
-            # The AMD SDK currently does not place its headers
-            # in /usr/include, therefore also search relative
-            # to the library
-            FIND_PATH(OPENCL_INCLUDE_DIRS CL/cl.h PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-            FIND_PATH(_OPENCL_CPP_INCLUDE_DIRS CL/cl.hpp PATHS ${_OPENCL_INC_CAND} "/usr/local/cuda/include")
-
-	ENDIF (WIN32)
-
-ENDIF (APPLE)
-
-FIND_PACKAGE_HANDLE_STANDARD_ARGS( OpenCL DEFAULT_MSG OPENCL_LIBRARIES OPENCL_INCLUDE_DIRS )
-
-IF( _OPENCL_CPP_INCLUDE_DIRS )
-	SET( OPENCL_HAS_CPP_BINDINGS TRUE )
-	LIST( APPEND OPENCL_INCLUDE_DIRS ${_OPENCL_CPP_INCLUDE_DIRS} )
-	# This is often the same, so clean up
-	LIST( REMOVE_DUPLICATES OPENCL_INCLUDE_DIRS )
-ENDIF( _OPENCL_CPP_INCLUDE_DIRS )
-
-MARK_AS_ADVANCED(
-  OPENCL_INCLUDE_DIRS
-)
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_float2/cl.hpp b/SpeedComparisons/GrayScott_OpenCL_float2/cl.hpp
deleted file mode 100644
index 99b86a665..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_float2/cl.hpp
+++ /dev/null
@@ -1,4011 +0,0 @@
-/*******************************************************************************
- * Copyright (c) 2008-2010 The Khronos Group Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and/or associated documentation files (the
- * "Materials"), to deal in the Materials without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sublicense, and/or sell copies of the Materials, and to
- * permit persons to whom the Materials are furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice shall be included
- * in all copies or substantial portions of the Materials.
- *
- * THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
- * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
- * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
- * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
- * CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
- * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
- * MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
- ******************************************************************************/
-
-/*! \file
- *
- *   \brief C++ bindings for OpenCL 1.0 (rev 48) and OpenCL 1.1 (rev 33)    
- *   \author Benedict R. Gaster and Laurent Morichetti
- *   
- *   Additions and fixes from Brian Cole, March 3rd 2010.
- *   
- *   \version 1.1
- *   \date June 2010
- *
- *   Optional extension support
- *
- *         cl
- *         cl_ext_device_fission
- *				#define USE_CL_DEVICE_FISSION
- */
-
-/*! \mainpage
- * \section intro Introduction
- * For many large applications C++ is the language of choice and so it seems
- * reasonable to define C++ bindings for OpenCL.
- *
- *
- * The interface is contained with a single C++ header file \em cl.hpp and all
- * definitions are contained within the namespace \em cl. There is no additional
- * requirement to include \em cl.h and to use either the C++ or original C
- * bindings it is enough to simply include \em cl.hpp.
- *
- * The bindings themselves are lightweight and correspond closely to the
- * underlying C API. Using the C++ bindings introduces no additional execution
- * overhead.
- *
- * For detail documentation on the bindings see:
- *
- * The OpenCL C++ Wrapper API 1.1 (revision 04)
- *  http://www.khronos.org/registry/cl/specs/opencl-cplusplus-1.1.pdf
- *
- * \section example Example
- *
- * The following example shows a general use case for the C++
- * bindings, including support for the optional exception feature and
- * also the supplied vector and string classes, see following sections for
- * decriptions of these features.
- *
- * \code
- * #define __CL_ENABLE_EXCEPTIONS
- * 
- * #if defined(__APPLE__) || defined(__MACOSX)
- * #include <OpenCL/cl.hpp>
- * #else
- * #include <CL/cl.hpp>
- * #endif
- * #include <cstdio>
- * #include <cstdlib>
- * #include <iostream>
- * 
- *  const char * helloStr  = "__kernel void "
- *                           "hello(void) "
- *                           "{ "
- *                           "  "
- *                           "} ";
- * 
- *  int
- *  main(void)
- *  {
- *     cl_int err = CL_SUCCESS;
- *     try {
- *
- *       std::vector<cl::Platform> platforms;
- *       cl::Platform::get(&platforms);
- *       if (platforms.size() == 0) {
- *           std::cout << "Platform size 0\n";
- *           return -1;
- *       }
- *
- *       cl_context_properties properties[] = 
- *          { CL_CONTEXT_PLATFORM, (cl_context_properties)(platforms[0])(), 0};
- *       cl::Context context(CL_DEVICE_TYPE_CPU, properties); 
- * 
- *       std::vector<cl::Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
- * 
- *       cl::Program::Sources source(1,
- *           std::make_pair(helloStr,strlen(helloStr)));
- *       cl::Program program_ = cl::Program(context, source);
- *       program_.build(devices);
- * 
- *       cl::Kernel kernel(program_, "hello", &err);
- * 
- *       cl::Event event;
- *       cl::CommandQueue queue(context, devices[0], 0, &err);
- *       queue.enqueueNDRangeKernel(
- *           kernel, 
- *           cl::NullRange, 
- *           cl::NDRange(4,4),
- *           cl::NullRange,
- *           NULL,
- *           &event); 
- * 
- *       event.wait();
- *     }
- *     catch (cl::Error err) {
- *        std::cerr 
- *           << "ERROR: "
- *           << err.what()
- *           << "("
- *           << err.err()
- *           << ")"
- *           << std::endl;
- *     }
- * 
- *    return EXIT_SUCCESS;
- *  }
- * 
- * \endcode
- *
- */
-#ifndef CL_HPP_
-#define CL_HPP_
-
-#ifdef _WIN32
-#include <windows.h>
-#include <malloc.h>
-#if defined(USE_DX_INTEROP)
-#include <CL/cl_d3d10.h>
-#endif
-#endif // _WIN32
-
-// 
-#if defined(USE_CL_DEVICE_FISSION)
-#include <CL/cl_ext.h>
-#endif
-
-#if defined(__APPLE__) || defined(__MACOSX)
-#include <OpenGL/OpenGL.h>
-#include <OpenCL/opencl.h>
-#else
-#include <GL/gl.h>
-#include <CL/opencl.h>
-#endif // !__APPLE__
-
-#if !defined(CL_CALLBACK)
-#define CL_CALLBACK
-#endif //CL_CALLBACK
-
-#include <utility>
-
-#if !defined(__NO_STD_VECTOR)
-#include <vector>
-#endif
-
-#if !defined(__NO_STD_STRING)
-#include <string>
-#endif 
-
-#if defined(linux) || defined(__APPLE__) || defined(__MACOSX)
-# include <alloca.h>
-#endif // linux
-
-#include <cstring>
-
-/*! \namespace cl
- *
- * \brief The OpenCL C++ bindings are defined within this namespace.
- *
- */
-namespace cl {
-
-#define __INIT_CL_EXT_FCN_PTR(name) \
-    if(!pfn_##name) { \
-        pfn_##name = (PFN_##name) \
-            clGetExtensionFunctionAddress(#name); \
-        if(!pfn_##name) { \
-        } \
-    }
-
-class Program;
-class Device;
-class Context;
-class CommandQueue;
-class Memory;
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-#include <exception>
-/*! \class Error
- * \brief Exception class
- */
-class Error : public std::exception
-{
-private:
-    cl_int err_;
-    const char * errStr_;
-public:
-    /*! Create a new CL error exception for a given error code
-     *  and corresponding message.
-     */
-    Error(cl_int err, const char * errStr = NULL) : err_(err), errStr_(errStr)
-    {}
-
-    ~Error() throw() {}
-
-    /*! \brief Get error string associated with exception
-     *
-     * \return A memory pointer to the error message string.
-     */
-    virtual const char * what() const throw ()
-    {
-        if (errStr_ == NULL) {
-            return "empty";
-        }
-        else {
-            return errStr_;
-        }
-    }
-
-    /*! \brief Get error code associated with exception
-     *
-     *  \return The error code.
-     */
-    const cl_int err(void) const { return err_; }
-};
-
-#define __ERR_STR(x) #x
-#else
-#define __ERR_STR(x) NULL
-#endif // __CL_ENABLE_EXCEPTIONS
-
-//! \cond DOXYGEN_DETAIL
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#define __GET_DEVICE_INFO_ERR               __ERR_STR(clgetDeviceInfo)
-#define __GET_PLATFORM_INFO_ERR             __ERR_STR(clGetPlatformInfo)
-#define __GET_DEVICE_IDS_ERR                __ERR_STR(clGetDeviceIDs)
-#define __GET_PLATFORM_IDS_ERR              __ERR_STR(clGetPlatformIDs)
-#define __GET_CONTEXT_INFO_ERR              __ERR_STR(clGetContextInfo)
-#define __GET_EVENT_INFO_ERR                __ERR_STR(clGetEventInfo)
-#define __GET_EVENT_PROFILE_INFO_ERR        __ERR_STR(clGetEventProfileInfo)
-#define __GET_MEM_OBJECT_INFO_ERR           __ERR_STR(clGetMemObjectInfo)
-#define __GET_IMAGE_INFO_ERR                __ERR_STR(clGetImageInfo)
-#define __GET_SAMPLER_INFO_ERR              __ERR_STR(clGetSamplerInfo)
-#define __GET_KERNEL_INFO_ERR               __ERR_STR(clGetKernelInfo)
-#define __GET_KERNEL_WORK_GROUP_INFO_ERR    __ERR_STR(clGetKernelWorkGroupInfo)
-#define __GET_PROGRAM_INFO_ERR              __ERR_STR(clGetProgramInfo)
-#define __GET_PROGRAM_BUILD_INFO_ERR        __ERR_STR(clGetProgramBuildInfo)
-#define __GET_COMMAND_QUEUE_INFO_ERR        __ERR_STR(clGetCommandQueueInfo)
-
-#define __CREATE_CONTEXT_FROM_TYPE_ERR      __ERR_STR(clCreateContextFromType)
-#define __GET_SUPPORTED_IMAGE_FORMATS_ERR   __ERR_STR(clGetSupportedImageFormats)
-
-#define __CREATE_BUFFER_ERR                 __ERR_STR(clCreateBuffer)
-#define __CREATE_SUBBUFFER_ERR              __ERR_STR(clCreateSubBuffer)
-#define __CREATE_GL_BUFFER_ERR              __ERR_STR(clCreateFromGLBuffer)
-#define __GET_GL_OBJECT_INFO_ERR            __ERR_STR(clGetGLObjectInfo)
-#define __CREATE_IMAGE2D_ERR                __ERR_STR(clCreateImage2D)
-#define __CREATE_IMAGE3D_ERR                __ERR_STR(clCreateImage3D)
-#define __CREATE_SAMPLER_ERR                __ERR_STR(clCreateSampler)
-#define __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR __ERR_STR(clSetMemObjectDestructorCallback)
-
-#define __CREATE_USER_EVENT_ERR             __ERR_STR(clCreateUserEvent)
-#define __SET_USER_EVENT_STATUS_ERR         __ERR_STR(clSetUserEventStatus)
-#define __SET_EVENT_CALLBACK_ERR            __ERR_STR(clSetEventCallback)
-#define __WAIT_FOR_EVENTS_ERR               __ERR_STR(clWaitForEvents)
-
-#define __CREATE_KERNEL_ERR                 __ERR_STR(clCreateKernel)
-#define __SET_KERNEL_ARGS_ERR               __ERR_STR(clSetKernelArg)
-#define __CREATE_PROGRAM_WITH_SOURCE_ERR    __ERR_STR(clCreateProgramWithSource)
-#define __CREATE_PROGRAM_WITH_BINARY_ERR    __ERR_STR(clCreateProgramWithBinary)
-#define __BUILD_PROGRAM_ERR                 __ERR_STR(clBuildProgram)
-#define __CREATE_KERNELS_IN_PROGRAM_ERR     __ERR_STR(clCreateKernelsInProgram)
-
-#define __CREATE_COMMAND_QUEUE_ERR          __ERR_STR(clCreateCommandQueue)
-#define __SET_COMMAND_QUEUE_PROPERTY_ERR    __ERR_STR(clSetCommandQueueProperty)
-#define __ENQUEUE_READ_BUFFER_ERR           __ERR_STR(clEnqueueReadBuffer)
-#define __ENQUEUE_READ_BUFFER_RECT_ERR      __ERR_STR(clEnqueueReadBufferRect)
-#define __ENQUEUE_WRITE_BUFFER_ERR          __ERR_STR(clEnqueueWriteBuffer)
-#define __ENQUEUE_WRITE_BUFFER_RECT_ERR     __ERR_STR(clEnqueueWriteBufferRect)
-#define __ENQEUE_COPY_BUFFER_ERR            __ERR_STR(clEnqueueCopyBuffer)
-#define __ENQEUE_COPY_BUFFER_RECT_ERR       __ERR_STR(clEnqueueCopyBufferRect)
-#define __ENQUEUE_READ_IMAGE_ERR            __ERR_STR(clEnqueueReadImage)
-#define __ENQUEUE_WRITE_IMAGE_ERR           __ERR_STR(clEnqueueWriteImage)
-#define __ENQUEUE_COPY_IMAGE_ERR            __ERR_STR(clEnqueueCopyImage)
-#define __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR  __ERR_STR(clEnqueueCopyImageToBuffer)
-#define __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR  __ERR_STR(clEnqueueCopyBufferToImage)
-#define __ENQUEUE_MAP_BUFFER_ERR            __ERR_STR(clEnqueueMapBuffer)
-#define __ENQUEUE_MAP_IMAGE_ERR             __ERR_STR(clEnqueueMapImage)
-#define __ENQUEUE_UNMAP_MEM_OBJECT_ERR      __ERR_STR(clEnqueueUnMapMemObject)
-#define __ENQUEUE_NDRANGE_KERNEL_ERR        __ERR_STR(clEnqueueNDRangeKernel)
-#define __ENQUEUE_TASK_ERR                  __ERR_STR(clEnqueueTask)
-#define __ENQUEUE_NATIVE_KERNEL             __ERR_STR(clEnqueueNativeKernel)
-#define __ENQUEUE_MARKER_ERR                __ERR_STR(clEnqueueMarker)
-#define __ENQUEUE_WAIT_FOR_EVENTS_ERR       __ERR_STR(clEnqueueWaitForEvents)
-#define __ENQUEUE_BARRIER_ERR               __ERR_STR(clEnqueueBarrier)
-
-#define __ENQUEUE_ACQUIRE_GL_ERR            __ERR_STR(clEnqueueAcquireGLObjects)
-#define __ENQUEUE_RELEASE_GL_ERR            __ERR_STR(clEnqueueReleaseGLObjects)
-
-#define __UNLOAD_COMPILER_ERR               __ERR_STR(clUnloadCompiler)
-
-#define __FLUSH_ERR                         __ERR_STR(clFlush)
-#define __FINISH_ERR                        __ERR_STR(clFinish)
-
-#define __CREATE_SUB_DEVICES                __ERR_STR(clCreateSubDevicesEXT)
-#endif // __CL_USER_OVERRIDE_ERROR_STRINGS
-//! \endcond
-
-/*! \class string
- * \brief Simple string class, that provides a limited subset of std::string
- * functionality but avoids many of the issues that come with that class.
- */
-class string
-{
-private:
-    ::size_t size_;
-    char * str_;
-public:
-    string(void) : size_(0), str_(NULL)
-    {
-    }
-
-    string(char * str, ::size_t size) :
-        size_(size),
-        str_(NULL)
-    {
-        str_ = new char[size_+1];
-        if (str_ != NULL) {
-            memcpy(str_, str, size_  * sizeof(char));
-            str_[size_] = '\0';
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string(char * str) :
-        str_(NULL)
-    {
-        size_= ::strlen(str);
-        str_ = new char[size_ + 1];
-        if (str_ != NULL) {
-            memcpy(str_, str, (size_ + 1) * sizeof(char));
-        }
-        else {
-            size_ = 0;
-        }
-    }
-
-    string& operator=(const string& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        if (rhs.size_ == 0 || rhs.str_ == NULL) {
-            size_ = 0;
-            str_  = NULL;
-        } 
-        else {
-            size_ = rhs.size_;
-            str_ = new char[size_ + 1];
-            if (str_ != NULL) {
-                memcpy(str_, rhs.str_, (size_ + 1) * sizeof(char));
-            }
-            else {
-                size_ = 0;
-            }
-        }
-
-        return *this;
-    }
-
-    string(const string& rhs)
-    {
-        *this = rhs;
-    }
-
-    ~string()
-    {
-        if (str_ != NULL) {
-            delete[] str_;
-        }
-    }
-
-    ::size_t size(void) const   { return size_; }
-    ::size_t length(void) const { return size(); }
-
-    const char * c_str(void) const { return (str_) ? str_ : "";}
-};
-
-#if !defined(__USE_DEV_STRING) && !defined(__NO_STD_STRING)
-#include <string>
-typedef std::string STRING_CLASS;
-#elif !defined(__USE_DEV_STRING) 
-typedef cl::string STRING_CLASS;
-#endif
-
-#if !defined(__USE_DEV_VECTOR) && !defined(__NO_STD_VECTOR)
-#include <vector>
-#define VECTOR_CLASS std::vector
-#elif !defined(__USE_DEV_VECTOR) 
-#define VECTOR_CLASS cl::vector 
-#endif
-
-#if !defined(__MAX_DEFAULT_VECTOR_SIZE)
-#define __MAX_DEFAULT_VECTOR_SIZE 10
-#endif
-
-/*! \class vector
- * \brief Fixed sized vector implementation that mirroring 
- * std::vector functionality.
- */
-template <typename T, unsigned int N = __MAX_DEFAULT_VECTOR_SIZE>
-class vector
-{
-private:
-    T data_[N];
-    unsigned int size_;
-    bool empty_;
-public:
-    vector() : 
-        size_(-1),
-        empty_(true)
-    {}
-
-    ~vector() {}
-
-    unsigned int size(void) const
-    {
-        return size_ + 1;
-    }
-
-    void clear()
-    {
-        size_ = -1;
-        empty_ = true;
-    }
-
-    void push_back (const T& x)
-    { 
-        if (size() < N) {
-            size_++;  
-            data_[size_] = x;
-            empty_ = false;
-        }
-    }
-
-    void pop_back(void)
-    {
-        if (!empty_) {
-            data_[size_].~T();
-            size_--;
-            if (size_ == -1) {
-                empty_ = true;
-            }
-        }
-    }
-  
-    vector(const vector<T, N>& vec) : 
-        size_(vec.size_),
-        empty_(vec.empty_)
-    {
-        if (!empty_) {
-            memcpy(&data_[0], &vec.data_[0], size() * sizeof(T));
-        }
-    } 
-
-    vector(unsigned int size, const T& val = T()) :
-        size_(-1),
-        empty_(true)
-    {
-        for (unsigned int i = 0; i < size; i++) {
-            push_back(val);
-        }
-    }
-
-    vector<T, N>& operator=(const vector<T, N>& rhs)
-    {
-        if (this == &rhs) {
-            return *this;
-        }
-
-        size_  = rhs.size_;
-        empty_ = rhs.empty_;
-
-        if (!empty_) {	
-            memcpy(&data_[0], &rhs.data_[0], size() * sizeof(T));
-        }
-    
-        return *this;
-    }
-
-    bool operator==(vector<T,N> &vec)
-    {
-        if (empty_ && vec.empty_) {
-            return true;
-        }
-
-        if (size() != vec.size()) {
-            return false;
-        }
-
-        return memcmp(&data_[0], &vec.data_[0], size() * sizeof(T)) == 0 ? true : false;
-    }
-  
-    operator T* ()             { return data_; }
-    operator const T* () const { return data_; }
-   
-    bool empty (void) const
-    {
-        return empty_;
-    }
-  
-    unsigned int max_size (void) const
-    {
-        return N;
-    }
-
-    unsigned int capacity () const
-    {
-        return sizeof(T) * N;
-    }
-
-    T& operator[](int index)
-    {
-        return data_[index];
-    }
-  
-    T operator[](int index) const
-    {
-        return data_[index];
-    }
-  
-    template<class I>
-    void assign(I start, I end)
-    {
-        clear();   
-        while(start < end) {
-            push_back(*start);
-            start++;
-        }
-    }
-
-    /*! \class iterator
-     * \brief Iterator class for vectors
-     */
-    class iterator
-    {
-    private:
-        vector<T,N> vec_;
-        int index_;
-        bool initialized_;
-    public:
-        iterator(void) : 
-            index_(-1),
-            initialized_(false)
-        {
-            index_ = -1;
-            initialized_ = false;
-        }
-
-        ~iterator(void) {}
-
-        static iterator begin(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = 0;
-            }
-
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-
-        static iterator end(vector<T,N> &vec)
-        {
-            iterator i;
-
-            if (!vec.empty()) {
-                i.index_ = vec.size();
-            }
-            i.vec_ = vec;
-            i.initialized_ = true;
-            return i;
-        }
-    
-        bool operator==(iterator i)
-        {
-            return ((vec_ == i.vec_) && 
-                    (index_ == i.index_) && 
-                    (initialized_ == i.initialized_));
-        }
-
-        bool operator!=(iterator i)
-        {
-            return (!(*this==i));
-        }
-
-        void operator++()
-        {
-            index_++;
-        }
-
-        void operator++(int x)
-        {
-            index_ += x;
-        }
-
-        void operator--()
-        {
-            index_--;
-        }
-
-        void operator--(int x)
-        {
-            index_ -= x;
-        }
-
-        T operator *()
-        {
-            return vec_[index_];
-        }
-    };
-
-    iterator begin(void)
-    {
-        return iterator::begin(*this);
-    }
-
-    iterator end(void)
-    {
-        return iterator::end(*this);
-    }
-
-    T& front(void)
-    {
-        return data_[0];
-    }
-
-    T& back(void)
-    {
-        return data_[size_];
-    }
-
-    const T& front(void) const
-    {
-        return data_[0];
-    }
-
-    const T& back(void) const
-    {
-        return data_[size_];
-    }
-};  
-    
-/*!
- * \brief size_t class used to interface between C++ and
- * OpenCL C calls that require arrays of size_t values, who's
- * size is known statically.
- */
-template <int N>
-struct size_t : public cl::vector< ::size_t, N> { };
-
-namespace detail {
-
-// GetInfo help struct
-template <typename Functor, typename T>
-struct GetInfoHelper
-{
-    static cl_int
-    get(Functor f, cl_uint name, T* param)
-    {
-        return f(name, sizeof(T), param, NULL);
-    }
-};
-
-// Specialized GetInfoHelper for VECTOR_CLASS params
-template <typename Func, typename T>
-struct GetInfoHelper<Func, VECTOR_CLASS<T> >
-{
-    static cl_int get(Func f, cl_uint name, VECTOR_CLASS<T>* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        T* value = (T*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        param->assign(&value[0], &value[required/sizeof(T)]);
-        return CL_SUCCESS;
-    }
-};
-
-// Specialized for getInfo<CL_PROGRAM_BINARIES>
-template <typename Func>
-struct GetInfoHelper<Func, VECTOR_CLASS<char *> >
-{
-    static cl_int
-    get(Func f, cl_uint name, VECTOR_CLASS<char *>* param)
-    {
-      cl_uint err = f(name, param->size() * sizeof(char *), &(*param)[0], NULL);
-      if (err != CL_SUCCESS) {
-        return err;
-      }
-      
-      return CL_SUCCESS;
-    }
-};
-
-// Specialized GetInfoHelper for STRING_CLASS params
-template <typename Func>
-struct GetInfoHelper<Func, STRING_CLASS>
-{
-    static cl_int get(Func f, cl_uint name, STRING_CLASS* param)
-    {
-        ::size_t required;
-        cl_int err = f(name, 0, NULL, &required);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        char* value = (char*) alloca(required);
-        err = f(name, required, value, NULL);
-        if (err != CL_SUCCESS) {
-            return err;
-        }
-
-        *param = value;
-        return CL_SUCCESS;
-    }
-};
-
-#define __GET_INFO_HELPER_WITH_RETAIN(CPP_TYPE) \
-namespace detail { \
-template <typename Func> \
-struct GetInfoHelper<Func, CPP_TYPE> \
-{ \
-    static cl_int get(Func f, cl_uint name, CPP_TYPE* param) \
-    { \
-      cl_uint err = f(name, sizeof(CPP_TYPE), param, NULL); \
-      if (err != CL_SUCCESS) { \
-        return err; \
-      } \
-      \
-      return ReferenceHandler<CPP_TYPE::cl_type>::retain((*param)()); \
-    } \
-}; \
-} 
-
-
-#define __PARAM_NAME_INFO_1_0(F) \
-    F(cl_platform_info, CL_PLATFORM_PROFILE, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VERSION, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_NAME, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_VENDOR, STRING_CLASS) \
-    F(cl_platform_info, CL_PLATFORM_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_device_info, CL_DEVICE_TYPE, cl_device_type) \
-    F(cl_device_info, CL_DEVICE_VENDOR_ID, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_COMPUTE_UNITS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_WORK_ITEM_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_CLOCK_FREQUENCY, cl_uint) \
-    F(cl_device_info, CL_DEVICE_ADDRESS_BITS, cl_bitfield) \
-    F(cl_device_info, CL_DEVICE_MAX_READ_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_WRITE_IMAGE_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_MEM_ALLOC_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE2D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_WIDTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_HEIGHT, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE3D_MAX_DEPTH, ::size_t) \
-    F(cl_device_info, CL_DEVICE_IMAGE_SUPPORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MAX_PARAMETER_SIZE, ::size_t) \
-    F(cl_device_info, CL_DEVICE_MAX_SAMPLERS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MEM_BASE_ADDR_ALIGN, cl_uint) \
-    F(cl_device_info, CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_SINGLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_TYPE, cl_device_mem_cache_type) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE, cl_uint)\
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_CACHE_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_GLOBAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_MAX_CONSTANT_ARGS, cl_uint) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_TYPE, cl_device_local_mem_type) \
-    F(cl_device_info, CL_DEVICE_LOCAL_MEM_SIZE, cl_ulong) \
-    F(cl_device_info, CL_DEVICE_ERROR_CORRECTION_SUPPORT, cl_bool) \
-    F(cl_device_info, CL_DEVICE_PROFILING_TIMER_RESOLUTION, ::size_t) \
-    F(cl_device_info, CL_DEVICE_ENDIAN_LITTLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_COMPILER_AVAILABLE, cl_bool) \
-    F(cl_device_info, CL_DEVICE_EXECUTION_CAPABILITIES, cl_device_exec_capabilities) \
-    F(cl_device_info, CL_DEVICE_QUEUE_PROPERTIES, cl_command_queue_properties) \
-    F(cl_device_info, CL_DEVICE_PLATFORM, cl_platform_id) \
-    F(cl_device_info, CL_DEVICE_NAME, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VENDOR, STRING_CLASS) \
-    F(cl_device_info, CL_DRIVER_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_PROFILE, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_VERSION, STRING_CLASS) \
-    F(cl_device_info, CL_DEVICE_EXTENSIONS, STRING_CLASS) \
-    \
-    F(cl_context_info, CL_CONTEXT_REFERENCE_COUNT, cl_uint) \
-    F(cl_context_info, CL_CONTEXT_DEVICES, VECTOR_CLASS<Device>) \
-    F(cl_context_info, CL_CONTEXT_PROPERTIES, VECTOR_CLASS<cl_context_properties>) \
-    \
-    F(cl_event_info, CL_EVENT_COMMAND_QUEUE, cl::CommandQueue) \
-    F(cl_event_info, CL_EVENT_COMMAND_TYPE, cl_command_type) \
-    F(cl_event_info, CL_EVENT_REFERENCE_COUNT, cl_uint) \
-    F(cl_event_info, CL_EVENT_COMMAND_EXECUTION_STATUS, cl_uint) \
-    \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_QUEUED, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_SUBMIT, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_START, cl_ulong) \
-    F(cl_profiling_info, CL_PROFILING_COMMAND_END, cl_ulong) \
-    \
-    F(cl_mem_info, CL_MEM_TYPE, cl_mem_object_type) \
-    F(cl_mem_info, CL_MEM_FLAGS, cl_mem_flags) \
-    F(cl_mem_info, CL_MEM_SIZE, ::size_t) \
-    F(cl_mem_info, CL_MEM_HOST_PTR, void*) \
-    F(cl_mem_info, CL_MEM_MAP_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_REFERENCE_COUNT, cl_uint) \
-    F(cl_mem_info, CL_MEM_CONTEXT, cl::Context) \
-    \
-    F(cl_image_info, CL_IMAGE_FORMAT, cl_image_format) \
-    F(cl_image_info, CL_IMAGE_ELEMENT_SIZE, ::size_t) \
-    F(cl_image_info, CL_IMAGE_ROW_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_SLICE_PITCH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_WIDTH, ::size_t) \
-    F(cl_image_info, CL_IMAGE_HEIGHT, ::size_t) \
-    F(cl_image_info, CL_IMAGE_DEPTH, ::size_t) \
-    \
-    F(cl_sampler_info, CL_SAMPLER_REFERENCE_COUNT, cl_uint) \
-    F(cl_sampler_info, CL_SAMPLER_CONTEXT, cl::Context) \
-    F(cl_sampler_info, CL_SAMPLER_NORMALIZED_COORDS, cl_addressing_mode) \
-    F(cl_sampler_info, CL_SAMPLER_ADDRESSING_MODE, cl_filter_mode) \
-    F(cl_sampler_info, CL_SAMPLER_FILTER_MODE, cl_bool) \
-    \
-    F(cl_program_info, CL_PROGRAM_REFERENCE_COUNT, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_CONTEXT, cl::Context) \
-    F(cl_program_info, CL_PROGRAM_NUM_DEVICES, cl_uint) \
-    F(cl_program_info, CL_PROGRAM_DEVICES, VECTOR_CLASS<cl_device_id>) \
-    F(cl_program_info, CL_PROGRAM_SOURCE, STRING_CLASS) \
-    F(cl_program_info, CL_PROGRAM_BINARY_SIZES, VECTOR_CLASS< ::size_t>) \
-    F(cl_program_info, CL_PROGRAM_BINARIES, VECTOR_CLASS<char *>) \
-    \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_STATUS, cl_build_status) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_OPTIONS, STRING_CLASS) \
-    F(cl_program_build_info, CL_PROGRAM_BUILD_LOG, STRING_CLASS) \
-    \
-    F(cl_kernel_info, CL_KERNEL_FUNCTION_NAME, STRING_CLASS) \
-    F(cl_kernel_info, CL_KERNEL_NUM_ARGS, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_REFERENCE_COUNT, cl_uint) \
-    F(cl_kernel_info, CL_KERNEL_CONTEXT, cl::Context) \
-    F(cl_kernel_info, CL_KERNEL_PROGRAM, cl::Program) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_WORK_GROUP_SIZE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_COMPILE_WORK_GROUP_SIZE, cl::size_t<3>) \
-    F(cl_kernel_work_group_info, CL_KERNEL_LOCAL_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_command_queue_info, CL_QUEUE_CONTEXT, cl::Context) \
-    F(cl_command_queue_info, CL_QUEUE_DEVICE, cl::Device) \
-    F(cl_command_queue_info, CL_QUEUE_REFERENCE_COUNT, cl_uint) \
-    F(cl_command_queue_info, CL_QUEUE_PROPERTIES, cl_command_queue_properties)
-
-#if defined(CL_VERSION_1_1)
-#define __PARAM_NAME_INFO_1_1(F) \
-    F(cl_context_info, CL_CONTEXT_NUM_DEVICES, cl_uint)\
-    F(cl_device_info, CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_INT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE, cl_uint) \
-    F(cl_device_info, CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF, cl_uint) \
-    F(cl_device_info, CL_DEVICE_DOUBLE_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HALF_FP_CONFIG, cl_device_fp_config) \
-    F(cl_device_info, CL_DEVICE_HOST_UNIFIED_MEMORY, cl_bool) \
-    \
-    F(cl_mem_info, CL_MEM_ASSOCIATED_MEMOBJECT, cl::Memory) \
-    F(cl_mem_info, CL_MEM_OFFSET, ::size_t) \
-    \
-    F(cl_kernel_work_group_info, CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE, ::size_t) \
-    F(cl_kernel_work_group_info, CL_KERNEL_PRIVATE_MEM_SIZE, cl_ulong) \
-    \
-    F(cl_event_info, CL_EVENT_CONTEXT, cl::Context)
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-#define __PARAM_NAME_DEVICE_FISSION(F) \
-    F(cl_device_info, CL_DEVICE_PARENT_DEVICE_EXT, cl_device_id) \
-	F(cl_device_info, CL_DEVICE_PARTITION_TYPES_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_AFFINITY_DOMAINS_EXT, VECTOR_CLASS<cl_device_partition_property_ext>) \
-	F(cl_device_info, CL_DEVICE_REFERENCE_COUNT_EXT , cl_uint) \
-	F(cl_device_info, CL_DEVICE_PARTITION_STYLE_EXT, VECTOR_CLASS<cl_device_partition_property_ext>)
-#endif // USE_CL_DEVICE_FISSION
-
-template <typename enum_type, cl_int Name>
-struct param_traits {};
-
-#define __DECLARE_PARAM_TRAITS(token, param_name, T) \
-struct token;                                        \
-template<>                                           \
-struct param_traits<detail:: token,param_name>       \
-{                                                    \
-    enum { value = param_name };                     \
-    typedef T param_type;                            \
-};
-
-__PARAM_NAME_INFO_1_0(__DECLARE_PARAM_TRAITS);
-#if defined(CL_VERSION_1_1)
-__PARAM_NAME_INFO_1_1(__DECLARE_PARAM_TRAITS);
-#endif // CL_VERSION_1_1
-
-#if defined(USE_CL_DEVICE_FISSION)
-__PARAM_NAME_DEVICE_FISSION(__DECLARE_PARAM_TRAITS);
-#endif // USE_CL_DEVICE_FISSION
-
-#undef __DECLARE_PARAM_TRAITS
-
-// Convenience functions
-
-template <typename Func, typename T>
-inline cl_int
-getInfo(Func f, cl_uint name, T* param)
-{
-    return GetInfoHelper<Func, T>::get(f, name, param);
-}
-
-template <typename Func, typename Arg0>
-struct GetInfoFunctor0
-{
-    Func f_; const Arg0& arg0_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename Arg1>
-struct GetInfoFunctor1
-{
-    Func f_; const Arg0& arg0_; const Arg1& arg1_;
-    cl_int operator ()(
-        cl_uint param, ::size_t size, void* value, ::size_t* size_ret)
-    { return f_(arg0_, arg1_, param, size, value, size_ret); }
-};
-
-template <typename Func, typename Arg0, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, cl_uint name, T* param)
-{
-    GetInfoFunctor0<Func, Arg0> f0 = { f, arg0 };
-    return GetInfoHelper<GetInfoFunctor0<Func, Arg0>, T>
-        ::get(f0, name, param);
-}
-
-template <typename Func, typename Arg0, typename Arg1, typename T>
-inline cl_int
-getInfo(Func f, const Arg0& arg0, const Arg1& arg1, cl_uint name, T* param)
-{
-    GetInfoFunctor1<Func, Arg0, Arg1> f0 = { f, arg0, arg1 };
-    return GetInfoHelper<GetInfoFunctor1<Func, Arg0, Arg1>, T>
-        ::get(f0, name, param);
-}
-
-template<typename T>
-struct ReferenceHandler
-{ };
-
-template <>
-struct ReferenceHandler<cl_device_id>
-{
-    // cl_device_id does not have retain().
-    static cl_int retain(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-    // cl_device_id does not have release().
-    static cl_int release(cl_device_id)
-    { return CL_INVALID_DEVICE; }
-};
-
-template <>
-struct ReferenceHandler<cl_platform_id>
-{
-    // cl_platform_id does not have retain().
-    static cl_int retain(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-    // cl_platform_id does not have release().
-    static cl_int release(cl_platform_id)
-    { return CL_INVALID_PLATFORM; }
-};
-
-template <>
-struct ReferenceHandler<cl_context>
-{
-    static cl_int retain(cl_context context)
-    { return ::clRetainContext(context); }
-    static cl_int release(cl_context context)
-    { return ::clReleaseContext(context); }
-};
-
-template <>
-struct ReferenceHandler<cl_command_queue>
-{
-    static cl_int retain(cl_command_queue queue)
-    { return ::clRetainCommandQueue(queue); }
-    static cl_int release(cl_command_queue queue)
-    { return ::clReleaseCommandQueue(queue); }
-};
-
-template <>
-struct ReferenceHandler<cl_mem>
-{
-    static cl_int retain(cl_mem memory)
-    { return ::clRetainMemObject(memory); }
-    static cl_int release(cl_mem memory)
-    { return ::clReleaseMemObject(memory); }
-};
-
-template <>
-struct ReferenceHandler<cl_sampler>
-{
-    static cl_int retain(cl_sampler sampler)
-    { return ::clRetainSampler(sampler); }
-    static cl_int release(cl_sampler sampler)
-    { return ::clReleaseSampler(sampler); }
-};
-
-template <>
-struct ReferenceHandler<cl_program>
-{
-    static cl_int retain(cl_program program)
-    { return ::clRetainProgram(program); }
-    static cl_int release(cl_program program)
-    { return ::clReleaseProgram(program); }
-};
-
-template <>
-struct ReferenceHandler<cl_kernel>
-{
-    static cl_int retain(cl_kernel kernel)
-    { return ::clRetainKernel(kernel); }
-    static cl_int release(cl_kernel kernel)
-    { return ::clReleaseKernel(kernel); }
-};
-
-template <>
-struct ReferenceHandler<cl_event>
-{
-    static cl_int retain(cl_event event)
-    { return ::clRetainEvent(event); }
-    static cl_int release(cl_event event)
-    { return ::clReleaseEvent(event); }
-};
-
-template <typename T>
-class Wrapper
-{
-public:
-    typedef T cl_type;
-
-protected:
-    cl_type object_;
-
-public:
-    Wrapper() : object_(NULL) { }
-
-    ~Wrapper()
-    {
-        if (object_ != NULL) { release(); }
-    }
-
-    Wrapper(const Wrapper<cl_type>& rhs)
-    {
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-    }
-
-    Wrapper<cl_type>& operator = (const Wrapper<cl_type>& rhs)
-    {
-        if (object_ != NULL) { release(); }
-        object_ = rhs.object_;
-        if (object_ != NULL) { retain(); }
-        return *this;
-    }
-
-    cl_type operator ()() const { return object_; }
-
-    cl_type& operator ()() { return object_; }
-
-protected:
-
-    cl_int retain() const
-    {
-        return ReferenceHandler<cl_type>::retain(object_);
-    }
-
-    cl_int release() const
-    {
-        return ReferenceHandler<cl_type>::release(object_);
-    }
-};
-
-#if defined(__CL_ENABLE_EXCEPTIONS)
-static inline cl_int errHandler (
-    cl_int err,
-    const char * errStr = NULL) throw(Error)
-{
-    if (err != CL_SUCCESS) {
-        throw Error(err, errStr);
-    }
-    return err;
-}
-#else
-static inline cl_int errHandler (cl_int err, const char * errStr = NULL)
-{
-    return err;
-}
-#endif // __CL_ENABLE_EXCEPTIONS
-
-} // namespace detail
-//! \endcond
-
-/*! \stuct ImageFormat
- * \brief ImageFormat interface fro cl_image_format.
- */
-struct ImageFormat : public cl_image_format
-{
-    ImageFormat(){}
-
-    ImageFormat(cl_channel_order order, cl_channel_type type)
-    {
-        image_channel_order = order;
-        image_channel_data_type = type;
-    }
-
-    ImageFormat& operator = (const ImageFormat& rhs)
-    {
-        if (this != &rhs) {
-            this->image_channel_data_type = rhs.image_channel_data_type;
-            this->image_channel_order     = rhs.image_channel_order;
-        }
-        return *this;
-    }
-};
-
-/*! \class Device
- * \brief Device interface for cl_device_id.
- */
-class Device : public detail::Wrapper<cl_device_id>
-{
-public:
-    Device(cl_device_id device) { object_ = device; }
-
-    Device() : detail::Wrapper<cl_type>() { }
-
-    Device(const Device& device) : detail::Wrapper<cl_type>(device) { }
-
-    Device& operator = (const Device& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_device_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetDeviceInfo, object_, name, param),
-            __GET_DEVICE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_device_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_device_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(USE_CL_DEVICE_FISSION)
-	cl_int createSubDevices(
-		const cl_device_partition_property_ext * properties,
-		VECTOR_CLASS<Device>* devices)
-	{
-		typedef CL_API_ENTRY cl_int 
-			( CL_API_CALL * PFN_clCreateSubDevicesEXT)(
-				cl_device_id /*in_device*/,
-                const cl_device_partition_property_ext * /* properties */,
-                cl_uint /*num_entries*/,
-                cl_device_id * /*out_devices*/,
-                cl_uint * /*num_devices*/ ) CL_EXT_SUFFIX__VERSION_1_1;
-
-		static PFN_clCreateSubDevicesEXT pfn_clCreateSubDevicesEXT = NULL;
-		__INIT_CL_EXT_FCN_PTR(clCreateSubDevicesEXT);
-
-		cl_uint n = 0;
-        cl_int err = pfn_clCreateSubDevicesEXT(object_, properties, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clCreateSubDevicesEXT(object_, properties, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_SUB_DEVICES);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
- 	}
-#endif
-};
-
-/*! \class Platform
- *  \brief Platform interface.
- */
-class Platform : public detail::Wrapper<cl_platform_id>
-{
-public:
-    static const Platform null();
-
-    Platform(cl_platform_id platform) { object_ = platform; }
-
-    Platform() : detail::Wrapper<cl_type>()  { }
-
-    Platform(const Platform& platform) : detail::Wrapper<cl_type>(platform) { }
-
-    Platform& operator = (const Platform& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getInfo(cl_platform_info name, STRING_CLASS* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetPlatformInfo, object_, name, param),
-            __GET_PLATFORM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_platform_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_platform_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getDevices(
-        cl_device_type type,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetDeviceIDs(object_, type, 0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = ::clGetDeviceIDs(object_, type, n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-
-#if defined(USE_DX_INTEROP)
-   /*! \brief Get the list of available D3D10 devices.
-     *
-     *  \param d3d_device_source.
-     *
-     *  \param d3d_object.
-     *
-     *  \param d3d_device_set.
-     *
-     *  \param devices returns a vector of OpenCL D3D10 devices found. The cl::Device
-     *  values returned in devices can be used to identify a specific OpenCL
-     *  device. If \a devices argument is NULL, this argument is ignored.
-     *
-     *  \return One of the following values:
-     *    - CL_SUCCESS if the function is executed successfully.
-     *
-     *  The application can query specific capabilities of the OpenCL device(s)
-     *  returned by cl::getDevices. This can be used by the application to
-     *  determine which device(s) to use.
-     *
-     * \note In the case that exceptions are enabled and a return value
-     * other than CL_SUCCESS is generated, then cl::Error exception is
-     * generated.
-     */
-    cl_int getDevices(
-        cl_d3d10_device_source_khr d3d_device_source,
-        void *                     d3d_object,
-        cl_d3d10_device_set_khr    d3d_device_set,
-        VECTOR_CLASS<Device>* devices) const
-    {
-        typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clGetDeviceIDsFromD3D10KHR)(
-            cl_platform_id platform, 
-            cl_d3d10_device_source_khr d3d_device_source, 
-            void * d3d_object,
-            cl_d3d10_device_set_khr d3d_device_set,
-            cl_uint num_entries,
-            cl_device_id * devices,
-            cl_uint* num_devices);
-
-        static PFN_clGetDeviceIDsFromD3D10KHR pfn_clGetDeviceIDsFromD3D10KHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clGetDeviceIDsFromD3D10KHR);
-
-        cl_uint n = 0;
-        cl_int err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set, 
-            0, 
-            NULL, 
-            &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        cl_device_id* ids = (cl_device_id*) alloca(n * sizeof(cl_device_id));
-        err = pfn_clGetDeviceIDsFromD3D10KHR(
-            object_, 
-            d3d_device_source, 
-            d3d_object,
-            d3d_device_set,
-            n, 
-            ids, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_DEVICE_IDS_ERR);
-        }
-
-        devices->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-#endif
-
-    static cl_int get(
-        VECTOR_CLASS<Platform>* platforms)
-    {
-        cl_uint n = 0;
-        cl_int err = ::clGetPlatformIDs(0, NULL, &n);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        cl_platform_id* ids = (cl_platform_id*) alloca(
-            n * sizeof(cl_platform_id));
-        err = ::clGetPlatformIDs(n, ids, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_PLATFORM_IDS_ERR);
-        }
-
-        platforms->assign(&ids[0], &ids[n]);
-        return CL_SUCCESS;
-    }
-};
-
-static inline cl_int
-UnloadCompiler()
-{
-    return ::clUnloadCompiler();
-}
-
-class Context : public detail::Wrapper<cl_context>
-{
-public:
-    Context(
-        const VECTOR_CLASS<Device>& devices,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContext(
-            properties, (cl_uint) devices.size(),
-            (cl_device_id*) &devices.front(),
-            notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context(
-        cl_device_type type,
-        cl_context_properties* properties = NULL,
-        void (CL_CALLBACK * notifyFptr)(
-            const char *,
-            const void *,
-            ::size_t,
-            void *) = NULL,
-        void* data = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateContextFromType(
-            properties, type, notifyFptr, data, &error);
-
-        detail::errHandler(error, __CREATE_CONTEXT_FROM_TYPE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Context() : detail::Wrapper<cl_type>() { }
-
-    Context(const Context& context) : detail::Wrapper<cl_type>(context) { }
-
-    Context& operator = (const Context& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_context_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetContextInfo, object_, name, param),
-            __GET_CONTEXT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_context_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_context_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int getSupportedImageFormats(
-        cl_mem_flags flags,
-        cl_mem_object_type type,
-        VECTOR_CLASS<ImageFormat>* formats) const
-    {
-        cl_uint numEntries;
-        cl_int err = ::clGetSupportedImageFormats(
-           object_, 
-           flags,
-           type, 
-           0, 
-           NULL, 
-           &numEntries);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        ImageFormat* value = (ImageFormat*)
-            alloca(numEntries * sizeof(ImageFormat));
-        err = ::clGetSupportedImageFormats(
-            object_, 
-            flags, 
-            type, 
-            numEntries,
-            (cl_image_format*) value, 
-            NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __GET_SUPPORTED_IMAGE_FORMATS_ERR);
-        }
-
-        formats->assign(&value[0], &value[numEntries]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Context)
-
-/*! \class Event
- * \brief Event interface for cl_event.
- */
-class Event : public detail::Wrapper<cl_event>
-{
-public:
-    Event() : detail::Wrapper<cl_type>() { }
-
-    Event(const Event& event) : detail::Wrapper<cl_type>(event) { }
-
-    Event& operator = (const Event& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_event_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetEventInfo, object_, name, param),
-            __GET_EVENT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_event_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_event_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getProfilingInfo(cl_profiling_info name, T* param) const
-    {
-        return detail::errHandler(detail::getInfo(
-            &::clGetEventProfilingInfo, object_, name, param),
-            __GET_EVENT_PROFILE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_profiling_info, name>::param_type
-    getProfilingInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_profiling_info, name>::param_type param;
-        cl_int result = getProfilingInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int wait() const
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(1, &object_),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setCallback(
-        cl_int type,
-        void (CL_CALLBACK * pfn_notify)(cl_event, cl_int, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetEventCallback(
-                object_,
-                type,
-                pfn_notify,
-                user_data), 
-            __SET_EVENT_CALLBACK_ERR);
-    }
-#endif
-
-    static cl_int
-    waitForEvents(const VECTOR_CLASS<Event>& events)
-    {
-        return detail::errHandler(
-            ::clWaitForEvents(
-                (cl_uint) events.size(), (cl_event*)&events.front()),
-            __WAIT_FOR_EVENTS_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Event)
-
-#if defined(CL_VERSION_1_1)
-/*! \class UserEvent
- * \brief User event interface for cl_event.
- */
-class UserEvent : public Event
-{
-public:
-    UserEvent(
-        const Context& context,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateUserEvent(
-            context(),
-            &error);
-
-        detail::errHandler(error, __CREATE_USER_EVENT_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    UserEvent() : Event() { }
-
-    UserEvent(const UserEvent& event) : Event(event) { }
-
-    UserEvent& operator = (const UserEvent& rhs)
-    {
-        if (this != &rhs) {
-            Event::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int setStatus(cl_int status)
-    {
-        return detail::errHandler(
-            ::clSetUserEventStatus(object_,status), 
-            __SET_USER_EVENT_STATUS_ERR);
-    }
-};
-#endif
-
-inline static cl_int
-WaitForEvents(const VECTOR_CLASS<Event>& events)
-{
-    return detail::errHandler(
-        ::clWaitForEvents(
-            (cl_uint) events.size(), (cl_event*)&events.front()),
-        __WAIT_FOR_EVENTS_ERR);
-}
-
-/*! \class Memory
- * \brief Memory interface for cl_mem.
- */
-class Memory : public detail::Wrapper<cl_mem>
-{
-public:
-    Memory() : detail::Wrapper<cl_type>() { }
-
-    Memory(const Memory& memory) : detail::Wrapper<cl_type>(memory) { }
-
-    Memory& operator = (const Memory& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_mem_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetMemObjectInfo, object_, name, param),
-            __GET_MEM_OBJECT_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_mem_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_mem_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int setDestructorCallback(
-        void (CL_CALLBACK * pfn_notify)(cl_mem, void *),		
-        void * user_data = NULL)
-    {
-        return detail::errHandler(
-            ::clSetMemObjectDestructorCallback(
-                object_,
-                pfn_notify,
-                user_data), 
-            __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR);
-    }
-#endif
-
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Memory)
-
-/*! \class Buffer
- * \brief Memory buffer interface.
- */
-class Buffer : public Memory
-{
-public:
-    Buffer(
-        const Context& context,
-        cl_mem_flags flags,
-        ::size_t size,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateBuffer(context(), flags, size, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Buffer() : Memory() { }
-
-    Buffer(const Buffer& buffer) : Memory(buffer) { }
-
-    Buffer& operator = (const Buffer& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-
-#if defined(CL_VERSION_1_1)
-    Buffer createSubBuffer(
-        cl_mem_flags flags,
-        cl_buffer_create_type buffer_create_type,
-        const void * buffer_create_info,
-        cl_int * err = NULL)
-    {
-        Buffer result;
-        cl_int error;
-        result.object_ = ::clCreateSubBuffer(
-            object_, 
-            flags, 
-            buffer_create_type, 
-            buffer_create_info, 
-            &error);
-
-        detail::errHandler(error, __CREATE_SUBBUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-
-        return result;
-	}		
-#endif
-};
-
-#if defined (USE_DX_INTEROP)
-class BufferD3D10 : public Buffer
-{
-public:
-    typedef CL_API_ENTRY cl_mem (CL_API_CALL *PFN_clCreateFromD3D10BufferKHR)(
-    cl_context context, cl_mem_flags flags, ID3D10Buffer*  buffer,
-    cl_int* errcode_ret);
-
-    BufferD3D10(
-        const Context& context,
-        cl_mem_flags flags,
-        ID3D10Buffer* bufobj,
-        cl_int * err = NULL)
-    {
-        static PFN_clCreateFromD3D10BufferKHR pfn_clCreateFromD3D10BufferKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clCreateFromD3D10BufferKHR);
-
-        cl_int error;
-        object_ = pfn_clCreateFromD3D10BufferKHR(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferD3D10() : Buffer() { }
-
-    BufferD3D10(const BufferD3D10& buffer) : Buffer(buffer) { }
-
-    BufferD3D10& operator = (const BufferD3D10& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-};
-#endif
-
-/*! \class BufferGL
- * \brief Memory buffer interface for GL interop.
- */
-class BufferGL : public Buffer
-{
-public:
-    BufferGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLBuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferGL() : Buffer() { }
-
-    BufferGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferGL& operator = (const BufferGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class BufferRenderGL
- * \brief Memory buffer interface for GL interop with renderbuffer.
- */
-class BufferRenderGL : public Buffer
-{
-public:
-    BufferRenderGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLuint bufobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLRenderbuffer(
-            context(),
-            flags,
-            bufobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    BufferRenderGL() : Buffer() { }
-
-    BufferRenderGL(const BufferGL& buffer) : Buffer(buffer) { }
-
-    BufferRenderGL& operator = (const BufferRenderGL& rhs)
-    {
-        if (this != &rhs) {
-            Buffer::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int getObjectInfo(
-        cl_gl_object_type *type,
-        GLuint * gl_object_name)
-    {
-        return detail::errHandler(
-            ::clGetGLObjectInfo(object_,type,gl_object_name),
-            __GET_GL_OBJECT_INFO_ERR);
-    }
-};
-
-/*! \class Image
- * \brief Base class  interface for all images.
- */
-class Image : public Memory
-{
-protected:
-    Image() : Memory() { }
-
-    Image(const Image& image) : Memory(image) { }
-
-    Image& operator = (const Image& rhs)
-    {
-        if (this != &rhs) {
-            Memory::operator=(rhs);
-        }
-        return *this;
-    }
-public:
-    template <typename T>
-    cl_int getImageInfo(cl_image_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetImageInfo, object_, name, param),
-            __GET_IMAGE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_image_info, name>::param_type
-    getImageInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_image_info, name>::param_type param;
-        cl_int result = getImageInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-/*! \class Image2D
- * \brief Image interface for 2D images.
- */
-class Image2D : public Image
-{
-public:
-    Image2D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t row_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage2D(
-            context(), flags,&format, width, height, row_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE2D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2D() { }
-
-    Image2D(const Image2D& image2D) : Image(image2D) { }
-
-    Image2D& operator = (const Image2D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image2DGL : public Image2D
-{
-public:
-    Image2DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture2D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image2DGL() : Image2D() { }
-
-    Image2DGL(const Image2DGL& image) : Image2D(image) { }
-
-    Image2DGL& operator = (const Image2DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image2D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image3D
- * \brief Image interface for 3D images.
- */
-class Image3D : public Image
-{
-public:
-    Image3D(
-        const Context& context,
-        cl_mem_flags flags,
-        ImageFormat format,
-        ::size_t width,
-        ::size_t height,
-        ::size_t depth,
-        ::size_t row_pitch = 0,
-        ::size_t slice_pitch = 0,
-        void* host_ptr = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateImage3D(
-            context(), flags, &format, width, height, depth, row_pitch,
-            slice_pitch, host_ptr, &error);
-
-        detail::errHandler(error, __CREATE_IMAGE3D_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3D() { }
-
-    Image3D(const Image3D& image3D) : Image(image3D) { }
-
-    Image3D& operator = (const Image3D& rhs)
-    {
-        if (this != &rhs) {
-            Image::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Image2DGL
- * \brief 2D image interface for GL interop.
- */
-class Image3DGL : public Image3D
-{
-public:
-    Image3DGL(
-        const Context& context,
-        cl_mem_flags flags,
-        GLenum target,
-        GLint  miplevel,
-        GLuint texobj,
-        cl_int * err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateFromGLTexture3D(
-            context(),
-            flags,
-            target,
-            miplevel,
-            texobj,
-            &error);
-
-        detail::errHandler(error, __CREATE_GL_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Image3DGL() : Image3D() { }
-
-    Image3DGL(const Image3DGL& image) : Image3D(image) { }
-
-    Image3DGL& operator = (const Image3DGL& rhs)
-    {
-        if (this != &rhs) {
-            Image3D::operator=(rhs);
-        }
-        return *this;
-    }
-};
-
-/*! \class Sampler
- * \brief Sampler interface for cl_sampler.
- */
-class Sampler : public detail::Wrapper<cl_sampler>
-{
-public:
-    Sampler() { }
-
-    Sampler(
-        const Context& context,
-        cl_bool normalized_coords,
-        cl_addressing_mode addressing_mode,
-        cl_filter_mode filter_mode,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateSampler(
-            context(), 
-            normalized_coords,
-            addressing_mode,
-            filter_mode,
-            &error);
-
-        detail::errHandler(error, __CREATE_SAMPLER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Sampler(const Sampler& sampler) : detail::Wrapper<cl_type>(sampler) { }
-
-    Sampler& operator = (const Sampler& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_sampler_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetSamplerInfo, object_, name, param),
-            __GET_SAMPLER_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_sampler_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_sampler_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Sampler)
-
-class Program;
-class CommandQueue;
-class Kernel;
-
-/*! \class NDRange
- * \brief NDRange interface
- */
-class NDRange
-{
-private:
-    size_t<3> sizes_;
-    cl_uint dimensions_;
-
-public:
-    NDRange()
-        : dimensions_(0)
-    { }
-
-    NDRange(::size_t size0)
-        : dimensions_(1)
-    {
-        sizes_.push_back(size0);
-    }
-
-    NDRange(::size_t size0, ::size_t size1)
-        : dimensions_(2)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-    }
-
-    NDRange(::size_t size0, ::size_t size1, ::size_t size2)
-        : dimensions_(3)
-    {
-        sizes_.push_back(size0);
-        sizes_.push_back(size1);
-        sizes_.push_back(size2);
-    }
-
-    operator const ::size_t*() const { return (const ::size_t*) sizes_; }
-    ::size_t dimensions() const { return dimensions_; }
-};
-
-static const NDRange NullRange;
-
-/*!
- * \struct LocalSpaceArg
- * \brief Local address raper for use with Kernel::setArg
- */
-struct LocalSpaceArg
-{
-    ::size_t size_;
-};
-
-namespace detail {
-
-template <typename T>
-struct KernelArgumentHandler
-{
-    static ::size_t size(const T&) { return sizeof(T); }
-    static T* ptr(T& value) { return &value; }
-};
-
-template <>
-struct KernelArgumentHandler<LocalSpaceArg>
-{
-    static ::size_t size(const LocalSpaceArg& value) { return value.size_; }
-    static void* ptr(LocalSpaceArg&) { return NULL; }
-};
-
-} 
-//! \endcond
-
-inline LocalSpaceArg
-__local(::size_t size)
-{
-    LocalSpaceArg ret = { size };
-    return ret;
-}
-
-class KernelFunctor;
-
-/*! \class Kernel
- * \brief Kernel interface that implements cl_kernel
- */
-class Kernel : public detail::Wrapper<cl_kernel>
-{
-public:
-    inline Kernel(const Program& program, const char* name, cl_int* err = NULL);
-
-    Kernel() { }
-
-    Kernel(const Kernel& kernel) : detail::Wrapper<cl_type>(kernel) { }
-
-    Kernel& operator = (const Kernel& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_kernel_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetKernelInfo, object_, name, param),
-            __GET_KERNEL_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_kernel_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getWorkGroupInfo(
-        const Device& device, cl_kernel_work_group_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetKernelWorkGroupInfo, object_, device(), name, param),
-                __GET_KERNEL_WORK_GROUP_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_kernel_work_group_info, name>::param_type
-        getWorkGroupInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-        detail::cl_kernel_work_group_info, name>::param_type param;
-        cl_int result = getWorkGroupInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int setArg(cl_uint index, T value)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(
-                object_,
-                index,
-                detail::KernelArgumentHandler<T>::size(value),
-                detail::KernelArgumentHandler<T>::ptr(value)),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    cl_int setArg(cl_uint index, ::size_t size, void* argPtr)
-    {
-        return detail::errHandler(
-            ::clSetKernelArg(object_, index, size, argPtr),
-            __SET_KERNEL_ARGS_ERR);
-    }
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local);
-
-    KernelFunctor bind(
-        const CommandQueue& queue,
-        const NDRange& global,
-        const NDRange& local);
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Kernel)
-
-/*! \class Program
- * \brief Program interface that implements cl_program.
- */
-class Program : public detail::Wrapper<cl_program>
-{
-public:
-    typedef VECTOR_CLASS<std::pair<const void*, ::size_t> > Binaries;
-    typedef VECTOR_CLASS<std::pair<const char*, ::size_t> > Sources;
-
-    Program(
-        const Context& context,
-        const Sources& sources,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-
-        const ::size_t n = (::size_t)sources.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const char** strings = (const char**) alloca(n * sizeof(const char*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            strings[i] = sources[(int)i].first;
-            lengths[i] = sources[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithSource(
-            context(), (cl_uint)n, strings, lengths, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_SOURCE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program(
-        const Context& context,
-        const VECTOR_CLASS<Device>& devices,
-        const Binaries& binaries,
-        VECTOR_CLASS<cl_int>* binaryStatus = NULL,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        const ::size_t n = binaries.size();
-        ::size_t* lengths = (::size_t*) alloca(n * sizeof(::size_t));
-        const unsigned char** images = (const unsigned char**) alloca(n * sizeof(const void*));
-
-        for (::size_t i = 0; i < n; ++i) {
-            images[i] = (const unsigned char*)binaries[(int)i].first;
-            lengths[i] = binaries[(int)i].second;
-        }
-
-        object_ = ::clCreateProgramWithBinary(
-            context(), (cl_uint) devices.size(),
-            (cl_device_id*)&devices.front(),
-            lengths, images, binaryStatus != NULL
-               ? (cl_int*) &binaryStatus->front()
-               : NULL, &error);
-
-        detail::errHandler(error, __CREATE_PROGRAM_WITH_BINARY_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    Program() { }
-
-    Program(const Program& program) : detail::Wrapper<cl_type>(program) { }
-
-    Program& operator = (const Program& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    cl_int build(
-        const VECTOR_CLASS<Device>& devices,
-        const char* options = NULL,
-        void (CL_CALLBACK * notifyFptr)(cl_program, void *) = NULL,
-        void* data = NULL) const
-    {
-        return detail::errHandler(
-            ::clBuildProgram(
-                object_,
-                (cl_uint)
-                devices.size(),
-                (cl_device_id*)&devices.front(),
-                options,
-                notifyFptr,
-                data),
-                __BUILD_PROGRAM_ERR);
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_program_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(&::clGetProgramInfo, object_, name, param),
-            __GET_PROGRAM_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    template <typename T>
-    cl_int getBuildInfo(
-        const Device& device, cl_program_build_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetProgramBuildInfo, object_, device(), name, param),
-                __GET_PROGRAM_BUILD_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_program_build_info, name>::param_type
-    getBuildInfo(const Device& device, cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_program_build_info, name>::param_type param;
-        cl_int result = getBuildInfo(device, name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int createKernels(VECTOR_CLASS<Kernel>* kernels)
-    {
-        cl_uint numKernels;
-        cl_int err = ::clCreateKernelsInProgram(object_, 0, NULL, &numKernels);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        Kernel* value = (Kernel*) alloca(numKernels * sizeof(Kernel));
-        err = ::clCreateKernelsInProgram(
-            object_, numKernels, (cl_kernel*) value, NULL);
-        if (err != CL_SUCCESS) {
-            return detail::errHandler(err, __CREATE_KERNELS_IN_PROGRAM_ERR);
-        }
-
-        kernels->assign(&value[0], &value[numKernels]);
-        return CL_SUCCESS;
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::Program)
-
-inline Kernel::Kernel(const Program& program, const char* name, cl_int* err)
-{
-    cl_int error;
-
-    object_ = ::clCreateKernel(program(), name, &error);
-    detail::errHandler(error, __CREATE_KERNEL_ERR);
-
-    if (err != NULL) {
-        *err = error;
-    }
-
-}
-
-/*! \class CommandQueue
- * \brief CommandQueue interface for cl_command_queue.
- */
-class CommandQueue : public detail::Wrapper<cl_command_queue>
-{
-public:
-    CommandQueue(
-        const Context& context,
-        const Device& device,
-        cl_command_queue_properties properties = 0,
-        cl_int* err = NULL)
-    {
-        cl_int error;
-        object_ = ::clCreateCommandQueue(
-            context(), device(), properties, &error);
-
-        detail::errHandler(error, __CREATE_COMMAND_QUEUE_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-    }
-
-    CommandQueue() { }
-
-    CommandQueue(const CommandQueue& commandQueue) : detail::Wrapper<cl_type>(commandQueue) { }
-
-    CommandQueue& operator = (const CommandQueue& rhs)
-    {
-        if (this != &rhs) {
-            detail::Wrapper<cl_type>::operator=(rhs);
-        }
-        return *this;
-    }
-
-    template <typename T>
-    cl_int getInfo(cl_command_queue_info name, T* param) const
-    {
-        return detail::errHandler(
-            detail::getInfo(
-                &::clGetCommandQueueInfo, object_, name, param),
-                __GET_COMMAND_QUEUE_INFO_ERR);
-    }
-
-    template <cl_int name> typename
-    detail::param_traits<detail::cl_command_queue_info, name>::param_type
-    getInfo(cl_int* err = NULL) const
-    {
-        typename detail::param_traits<
-            detail::cl_command_queue_info, name>::param_type param;
-        cl_int result = getInfo(name, &param);
-        if (err != NULL) {
-            *err = result;
-        }
-        return param;
-    }
-
-    cl_int enqueueReadBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_BUFFER_ERR);
-    }
-
-    cl_int enqueueWriteBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        ::size_t offset,
-        ::size_t size,
-        const void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBuffer(
-                object_, buffer(), blocking, offset, size,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBuffer(
-        const Buffer& src,
-        const Buffer& dst,
-        ::size_t src_offset,
-        ::size_t dst_offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBuffer(
-                object_, src(), dst(), src_offset, dst_offset, size,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_ERR);
-    }
-
-#if defined(CL_VERSION_1_1)
-    cl_int enqueueReadBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_READ_BUFFER_RECT_ERR);
-    }
-
-
-    cl_int enqueueWriteBufferRect(
-        const Buffer& buffer,
-        cl_bool blocking,
-        const size_t<3>& buffer_offset,
-        const size_t<3>& host_offset,
-        const size_t<3>& region,
-        ::size_t buffer_row_pitch,
-        ::size_t buffer_slice_pitch,
-        ::size_t host_row_pitch,
-        ::size_t host_slice_pitch,
-        void *ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteBufferRect(
-                object_, 
-                buffer(), 
-                blocking, 
-                (const ::size_t *)buffer_offset,
-                (const ::size_t *)host_offset,
-                (const ::size_t *)region,
-                buffer_row_pitch,
-                buffer_slice_pitch,
-                host_row_pitch,
-                host_slice_pitch,
-                ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-                __ENQUEUE_WRITE_BUFFER_RECT_ERR);
-    }
-
-    cl_int enqueueCopyBufferRect(
-        const Buffer& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        ::size_t src_row_pitch,
-        ::size_t src_slice_pitch,
-        ::size_t dst_row_pitch,
-        ::size_t dst_slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferRect(
-                object_, 
-                src(), 
-                dst(), 
-                (const ::size_t *)src_origin, 
-                (const ::size_t *)dst_origin, 
-                (const ::size_t *)region,
-                src_row_pitch,
-                src_slice_pitch,
-                dst_row_pitch,
-                dst_slice_pitch,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQEUE_COPY_BUFFER_RECT_ERR);
-    }
-#endif
-
-    cl_int enqueueReadImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueReadImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_READ_IMAGE_ERR);
-    }
-
-    cl_int enqueueWriteImage(
-        const Image& image,
-        cl_bool blocking,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t row_pitch,
-        ::size_t slice_pitch,
-        void* ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWriteImage(
-                object_, image(), blocking, (const ::size_t *) origin,
-                (const ::size_t *) region, row_pitch, slice_pitch, ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_WRITE_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImage(
-        const Image& src,
-        const Image& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImage(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *)dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_ERR);
-    }
-
-    cl_int enqueueCopyImageToBuffer(
-        const Image& src,
-        const Buffer& dst,
-        const size_t<3>& src_origin,
-        const size_t<3>& region,
-        ::size_t dst_offset,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyImageToBuffer(
-                object_, src(), dst(), (const ::size_t *) src_origin,
-                (const ::size_t *) region, dst_offset,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR);
-    }
-
-    cl_int enqueueCopyBufferToImage(
-        const Buffer& src,
-        const Image& dst,
-        ::size_t src_offset,
-        const size_t<3>& dst_origin,
-        const size_t<3>& region,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueCopyBufferToImage(
-                object_, src(), dst(), src_offset,
-                (const ::size_t *) dst_origin, (const ::size_t *) region,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR);
-    }
-
-    void* enqueueMapBuffer(
-        const Buffer& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        ::size_t offset,
-        ::size_t size,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapBuffer(
-            object_, buffer(), blocking, flags, offset, size,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_BUFFER_ERR);
-        if (err != NULL) {
-            *err = error;
-        }
-        return result;
-    }
-
-    void* enqueueMapImage(
-        const Image& buffer,
-        cl_bool blocking,
-        cl_map_flags flags,
-        const size_t<3>& origin,
-        const size_t<3>& region,
-        ::size_t * row_pitch,
-        ::size_t * slice_pitch,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL,
-        cl_int* err = NULL) const
-    {
-        cl_int error;
-        void * result = ::clEnqueueMapImage(
-            object_, buffer(), blocking, flags,
-            (const ::size_t *) origin, (const ::size_t *) region,
-            row_pitch, slice_pitch,
-            (events != NULL) ? (cl_uint) events->size() : 0,
-            (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-            (cl_event*) event,
-            &error);
-
-        detail::errHandler(error, __ENQUEUE_MAP_IMAGE_ERR);
-        if (err != NULL) {
-              *err = error;
-        }
-        return result;
-    }
-
-    cl_int enqueueUnmapMemObject(
-        const Memory& memory,
-        void* mapped_ptr,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueUnmapMemObject(
-                object_, memory(), mapped_ptr,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_UNMAP_MEM_OBJECT_ERR);
-    }
-
-    cl_int enqueueNDRangeKernel(
-        const Kernel& kernel,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueNDRangeKernel(
-                object_, kernel(), (cl_uint) global.dimensions(),
-                offset.dimensions() != 0 ? (const ::size_t*) offset : NULL,
-                (const ::size_t*) global,
-                local.dimensions() != 0 ? (const ::size_t*) local : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NDRANGE_KERNEL_ERR);
-    }
-
-    cl_int enqueueTask(
-        const Kernel& kernel,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueTask(
-                object_, kernel(),
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_TASK_ERR);
-    }
-
-    cl_int enqueueNativeKernel(
-        void (*userFptr)(void *),
-        std::pair<void*, ::size_t> args,
-        const VECTOR_CLASS<Memory>* mem_objects = NULL,
-        const VECTOR_CLASS<const void*>* mem_locs = NULL,
-        const VECTOR_CLASS<Event>* events = NULL,
-        Event* event = NULL) const
-    {
-        cl_mem * mems = (mem_objects != NULL && mem_objects->size() > 0) 
-            ? (cl_mem*) alloca(mem_objects->size() * sizeof(cl_mem))
-            : NULL;
-
-        if (mems != NULL) {
-            for (unsigned int i = 0; i < mem_objects->size(); i++) {
-                mems[i] = ((*mem_objects)[i])();
-            }
-        }
-
-        return detail::errHandler(
-            ::clEnqueueNativeKernel(
-                object_, userFptr, args.first, args.second,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                mems,
-                (mem_locs != NULL) ? (const void **) &mem_locs->front() : NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_NATIVE_KERNEL);
-    }
-
-    cl_int enqueueMarker(Event* event = NULL) const
-    {
-        return detail::errHandler(
-            ::clEnqueueMarker(object_, (cl_event*) event),
-            __ENQUEUE_MARKER_ERR);
-    }
-
-    cl_int enqueueWaitForEvents(const VECTOR_CLASS<Event>& events) const
-    {
-        return detail::errHandler(
-            ::clEnqueueWaitForEvents(
-                object_,
-                (cl_uint) events.size(),
-                (const cl_event*) &events.front()),
-            __ENQUEUE_WAIT_FOR_EVENTS_ERR);
-    }
-
-    cl_int enqueueAcquireGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueAcquireGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseGLObjects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         return detail::errHandler(
-             ::clEnqueueReleaseGLObjects(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL && events->size() > 0) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_RELEASE_GL_ERR);
-     }
-
-#if defined (USE_DX_INTEROP)
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueAcquireD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects, cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-typedef CL_API_ENTRY cl_int (CL_API_CALL *PFN_clEnqueueReleaseD3D10ObjectsKHR)(
-    cl_command_queue command_queue, cl_uint num_objects,
-    const cl_mem* mem_objects,  cl_uint num_events_in_wait_list,
-    const cl_event* event_wait_list, cl_event* event);
-
-    cl_int enqueueAcquireD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-     {
-         static PFN_clEnqueueAcquireD3D10ObjectsKHR pfn_clEnqueueAcquireD3D10ObjectsKHR = NULL;
-         __INIT_CL_EXT_FCN_PTR(clEnqueueAcquireD3D10ObjectsKHR);
-		
-         return detail::errHandler(
-             pfn_clEnqueueAcquireD3D10ObjectsKHR(
-                 object_,
-                 (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                 (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                 (events != NULL) ? (cl_uint) events->size() : 0,
-                 (events != NULL) ? (cl_event*) &events->front() : NULL,
-                 (cl_event*) event),
-             __ENQUEUE_ACQUIRE_GL_ERR);
-     }
-
-    cl_int enqueueReleaseD3D10Objects(
-         const VECTOR_CLASS<Memory>* mem_objects = NULL,
-         const VECTOR_CLASS<Event>* events = NULL,
-         Event* event = NULL) const
-    {
-        static PFN_clEnqueueReleaseD3D10ObjectsKHR pfn_clEnqueueReleaseD3D10ObjectsKHR = NULL;
-        __INIT_CL_EXT_FCN_PTR(clEnqueueReleaseD3D10ObjectsKHR);
-
-        return detail::errHandler(
-            pfn_clEnqueueReleaseD3D10ObjectsKHR(
-                object_,
-                (mem_objects != NULL) ? (cl_uint) mem_objects->size() : 0,
-                (mem_objects != NULL) ? (const cl_mem *) &mem_objects->front(): NULL,
-                (events != NULL) ? (cl_uint) events->size() : 0,
-                (events != NULL) ? (cl_event*) &events->front() : NULL,
-                (cl_event*) event),
-            __ENQUEUE_RELEASE_GL_ERR);
-    }
-#endif
-
-    cl_int enqueueBarrier() const
-    {
-        return detail::errHandler(
-            ::clEnqueueBarrier(object_),
-            __ENQUEUE_BARRIER_ERR);
-    }
-
-    cl_int flush() const
-    {
-        return detail::errHandler(::clFlush(object_), __FLUSH_ERR);
-    }
-
-    cl_int finish() const
-    {
-        return detail::errHandler(::clFinish(object_), __FINISH_ERR);
-    }
-};
-
-__GET_INFO_HELPER_WITH_RETAIN(cl::CommandQueue)
-
-/*! \class KernelFunctor
- * \brief Kernel functor interface
- *
- * \note Currently only functors of zero to ten arguments are supported. It
- * is straightforward to add more and a more general solution, similar to
- * Boost.Lambda could be followed if required in the future.
- */
-class KernelFunctor
-{
-private:
-    Kernel kernel_;
-    CommandQueue queue_;
-    NDRange offset_;
-    NDRange global_;
-    NDRange local_;
-
-    cl_int err_;
-public:
-    KernelFunctor() { }
-
-    KernelFunctor(
-        const Kernel& kernel,
-        const CommandQueue& queue,
-        const NDRange& offset,
-        const NDRange& global,
-        const NDRange& local) :
-            kernel_(kernel),
-            queue_(queue),
-            offset_(offset),
-            global_(global),
-            local_(local),
-            err_(CL_SUCCESS)
-    {}
-
-    KernelFunctor& operator=(const KernelFunctor& rhs);
-
-    KernelFunctor(const KernelFunctor& rhs);
-
-    cl_int getError() { return err_; }
-
-    inline Event operator()(const VECTOR_CLASS<Event>* events = NULL);
-
-    template<typename A1>
-    inline Event operator()(
-        const A1& a1, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5, class A6>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4,
-             class A5, class A6, class A7>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8,
-        const VECTOR_CLASS<Event>* events = NULL);
-
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6, 
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11, 
-        const A12& a12, 
-        const A13& a13,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14,
-        const VECTOR_CLASS<Event>* events = NULL);
-    
-    template<class A1, class A2, class A3, class A4, class A5,
-             class A6, class A7, class A8, class A9, class A10,
-             class A11, class A12, class A13, class A14, class A15>
-    inline Event operator()(
-        const A1& a1, 
-        const A2& a2, 
-        const A3& a3, 
-        const A4& a4, 
-        const A5& a5, 
-        const A6& a6,
-        const A7& a7, 
-        const A8& a8, 
-        const A9& a9, 
-        const A10& a10, 
-        const A11& a11,
-        const A12& a12, 
-        const A13& a13, 
-        const A14& a14, 
-        const A15& a15,
-        const VECTOR_CLASS<Event>* events = NULL);
-};
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& offset,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,offset,global,local);
-}
-
-inline KernelFunctor Kernel::bind(
-    const CommandQueue& queue,
-    const NDRange& global,
-    const NDRange& local)
-{
-    return KernelFunctor(*this,queue,NullRange,global,local);
-}
-
-inline KernelFunctor& KernelFunctor::operator=(const KernelFunctor& rhs)
-{
-    if (this == &rhs) {
-        return *this;
-    }
-    
-    kernel_ = rhs.kernel_;
-    queue_  = rhs.queue_;
-    offset_ = rhs.offset_;
-    global_ = rhs.global_;
-    local_  = rhs.local_;
-    
-    return *this;
-}
-
-inline KernelFunctor::KernelFunctor(const KernelFunctor& rhs) :
-    kernel_(rhs.kernel_),
-    queue_(rhs.queue_),
-    offset_(rhs.offset_),
-    global_(rhs.global_),
-    local_(rhs.local_)
-{
-}
-
-Event KernelFunctor::operator()(const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4,
-         typename A5, typename A6, typename A7>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<typename A1, typename A2, typename A3, typename A4, typename A5,
-         typename A6, typename A7, typename A8, typename A9, typename A10>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11, 
-    const A12& a12, 
-    const A13& a13,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5, 
-    const A6& a6,
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-template<class A1, class A2, class A3, class A4, class A5,
-         class A6, class A7, class A8, class A9, class A10,
-         class A11, class A12, class A13, class A14, class A15>
-Event KernelFunctor::operator()(
-    const A1& a1, 
-    const A2& a2, 
-    const A3& a3, 
-    const A4& a4, 
-    const A5& a5,
-    const A6& a6, 
-    const A7& a7, 
-    const A8& a8, 
-    const A9& a9, 
-    const A10& a10, 
-    const A11& a11,
-    const A12& a12, 
-    const A13& a13, 
-    const A14& a14, 
-    const A15& a15,
-    const VECTOR_CLASS<Event>* events)
-{
-    Event event;
-    
-    kernel_.setArg(0,a1);
-    kernel_.setArg(1,a2);
-    kernel_.setArg(2,a3);
-    kernel_.setArg(3,a4);
-    kernel_.setArg(4,a5);
-    kernel_.setArg(5,a6);
-    kernel_.setArg(6,a7);
-    kernel_.setArg(7,a8);
-    kernel_.setArg(8,a9);
-    kernel_.setArg(9,a10);
-    kernel_.setArg(10,a11);
-    kernel_.setArg(11,a12);
-    kernel_.setArg(12,a13);
-    kernel_.setArg(13,a14);
-    kernel_.setArg(14,a15);
-
-    err_ = queue_.enqueueNDRangeKernel(
-        kernel_,
-        offset_,
-        global_,
-        local_,
-        NULL,    // bgaster_fixme - do we want to allow wait event lists?
-        &event);
-
-    return event;
-}
-
-#undef __ERR_STR
-#if !defined(__CL_USER_OVERRIDE_ERROR_STRINGS)
-#undef __GET_DEVICE_INFO_ERR
-#undef __GET_PLATFORM_INFO_ERR
-#undef __GET_DEVICE_IDS_ERR
-#undef __GET_CONTEXT_INFO_ERR
-#undef __GET_EVENT_INFO_ERR
-#undef __GET_EVENT_PROFILE_INFO_ERR
-#undef __GET_MEM_OBJECT_INFO_ERR
-#undef __GET_IMAGE_INFO_ERR
-#undef __GET_SAMPLER_INFO_ERR
-#undef __GET_KERNEL_INFO_ERR
-#undef __GET_KERNEL_WORK_GROUP_INFO_ERR
-#undef __GET_PROGRAM_INFO_ERR
-#undef __GET_PROGRAM_BUILD_INFO_ERR
-#undef __GET_COMMAND_QUEUE_INFO_ERR
-
-#undef __CREATE_CONTEXT_FROM_TYPE_ERR
-#undef __GET_SUPPORTED_IMAGE_FORMATS_ERR
-
-#undef __CREATE_BUFFER_ERR
-#undef __CREATE_SUBBUFFER_ERR
-#undef __CREATE_IMAGE2D_ERR
-#undef __CREATE_IMAGE3D_ERR
-#undef __CREATE_SAMPLER_ERR
-#undef __SET_MEM_OBJECT_DESTRUCTOR_CALLBACK_ERR
-
-#undef __CREATE_USER_EVENT_ERR
-#undef __SET_USER_EVENT_STATUS_ERR
-#undef __SET_EVENT_CALLBACK_ERR
-
-#undef __WAIT_FOR_EVENTS_ERR
-
-#undef __CREATE_KERNEL_ERR
-#undef __SET_KERNEL_ARGS_ERR
-#undef __CREATE_PROGRAM_WITH_SOURCE_ERR
-#undef __CREATE_PROGRAM_WITH_BINARY_ERR
-#undef __BUILD_PROGRAM_ERR
-#undef __CREATE_KERNELS_IN_PROGRAM_ERR
-
-#undef __CREATE_COMMAND_QUEUE_ERR
-#undef __SET_COMMAND_QUEUE_PROPERTY_ERR
-#undef __ENQUEUE_READ_BUFFER_ERR
-#undef __ENQUEUE_WRITE_BUFFER_ERR
-#undef __ENQUEUE_READ_BUFFER_RECT_ERR
-#undef __ENQUEUE_WRITE_BUFFER_RECT_ERR
-#undef __ENQEUE_COPY_BUFFER_ERR
-#undef __ENQEUE_COPY_BUFFER_RECT_ERR
-#undef __ENQUEUE_READ_IMAGE_ERR
-#undef __ENQUEUE_WRITE_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_ERR
-#undef __ENQUEUE_COPY_IMAGE_TO_BUFFER_ERR
-#undef __ENQUEUE_COPY_BUFFER_TO_IMAGE_ERR
-#undef __ENQUEUE_MAP_BUFFER_ERR
-#undef __ENQUEUE_MAP_IMAGE_ERR
-#undef __ENQUEUE_UNMAP_MEM_OBJECT_ERR
-#undef __ENQUEUE_NDRANGE_KERNEL_ERR
-#undef __ENQUEUE_TASK_ERR
-#undef __ENQUEUE_NATIVE_KERNEL
-
-#undef __UNLOAD_COMPILER_ERR
-#endif //__CL_USER_OVERRIDE_ERROR_STRINGS
-
-#undef __GET_INFO_HELPER_WITH_RETAIN
-
-// Extensions
-#undef __INIT_CL_EXT_FCN_PTR
-#undef __CREATE_SUB_DEVICES
-
-#if defined(USE_CL_DEVICE_FISSION)
-#undef __PARAM_NAME_DEVICE_FISSION
-#endif // USE_CL_DEVICE_FISSION
-
-} // namespace cl
-
-#endif // CL_HPP_
diff --git a/SpeedComparisons/GrayScott_OpenCL_float2/gray_scott_opencl_float2.cpp b/SpeedComparisons/GrayScott_OpenCL_float2/gray_scott_opencl_float2.cpp
deleted file mode 100644
index 567fefc02..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_float2/gray_scott_opencl_float2.cpp
+++ /dev/null
@@ -1,383 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCL:
-#define __NO_STD_VECTOR // Use cl::vector instead of STL version
-#define __CL_ENABLE_EXCEPTIONS
-
-// cl.hpp is standard but doesn't come with most SDKs, so download it from here:
-// http://www.khronos.org/registry/cl/api/1.1/cl.hpp
-#ifdef __APPLE__
-# include "cl.hpp"
-#else
-# include <CL/cl.hpp>
-#endif
-
-using namespace cl;
-
-// STL:
-#include <fstream>
-#include <iostream>
-
-// local:
-#include "defs.h"
-
-void init(float ab[X][Y][2]);
-bool display(float r[X][Y][2],float g[X][Y][2],float b[X][Y][2],
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
-
-static int g_opt_device = 0;
-static int g_wrap = 1;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if ((i+1<argc) && (strcmp(argv[i],"-device")==0)) {
-            // select an output device
-            i++; g_opt_device = atoi(argv[i]);
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            std::cout << "Unrecognized argument: '" << argv[i] << "'\n";
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    float k, f;
-     k = 0.064f; f = 0.035f; // solitons with mitosis (spots that multiply)
-    // k = 0.06f; f = 0.035f; // stripes
-    // k = 0.065f; f = 0.056f; // long stripes
-    // k = 0.064f; f = 0.04f; // dots and stripes
-    // k = 0.0475f; f = 0.0118f; // spiral waves
-    float speed = 2.0f;
-    // ----------------
-    
-    float ab[X][Y][2];
-    const int MEM_SIZE = sizeof(float)*X*Y*2;
-
-    // put the initial conditions into each cell
-    init(ab);
-
-    try { 
-        // Get available OpenCL platforms
-        cl::vector<Platform> platforms;
-        Platform::get(&platforms);
- 
-        // Select the default platform and create a context using this platform and the GPU
-        cl_context_properties cps[3] = { 
-            CL_CONTEXT_PLATFORM, 
-            (cl_context_properties)(platforms[0])(), 
-            0 
-        };
-        Context context( CL_DEVICE_TYPE_GPU, cps);
- 
-        // Get a list of devices on this platform
-        cl::vector<Device> devices = context.getInfo<CL_CONTEXT_DEVICES>();
-
-        // range-check the user's selection
-        int maxdev = devices.size() - 1;
-        g_opt_device = (g_opt_device > maxdev) ? maxdev :
-                                     ((g_opt_device < 0) ? 0 : g_opt_device);
-        std::cout << (maxdev+1) << " device(s) available; using device "
-                                                    << g_opt_device << ".\n";
-
-        Device &device = devices[g_opt_device];
-        std::cout << "Global memory: " << device.getInfo<CL_DEVICE_GLOBAL_MEM_SIZE>() << " bytes\n";
-        std::cout << "Local memory: " << device.getInfo<CL_DEVICE_LOCAL_MEM_SIZE>() << " bytes\n";
-        std::cout << "Local memory type: " << std::string((device.getInfo<CL_DEVICE_LOCAL_MEM_TYPE>()==CL_LOCAL)?"local":"global") << " \n";
-        std::cout << "CL_DEVICE_MAX_WORK_GROUP_SIZE: " << device.getInfo<CL_DEVICE_MAX_WORK_GROUP_SIZE>() << "\n";
-
-        // Create a command queue and use the selected device
-        if (maxdev < 0) {
-          std::cerr << "error -- need at least one OpenCL capable device.\n";
-          exit(-1);
-        }
-        CommandQueue queue = CommandQueue(context, device);
-        Event event;
- 
-        // Read source file
-        std::string kfn = CL_SOURCE_DIR; // (defined in CMakeLists.txt to be the source folder)
-        kfn += "/grayscott_kernel_float2.cl";
-        std::ifstream sourceFile(kfn.c_str());
-        std::string sourceCode(
-            std::istreambuf_iterator<char>(sourceFile),
-            (std::istreambuf_iterator<char>()));
-        Program::Sources source(1, std::make_pair(sourceCode.c_str(), sourceCode.length()+1));
- 
-        // Make program of the source code in the context
-        Program program = Program(context, source);
- 
-        // Build program for these specific devices
-        // If wrap (toroidal) option is selected, we define a preprocessor flag
-        // that controls how the xm1, xp1, etc. are computed.
-        program.build(devices, g_wrap ? "-D WRAP" : NULL, NULL, NULL);
- 
-        // Make kernel
-        Kernel kernel(program, "grayscott_compute");
-
-        std::cout << "CL_KERNEL_WORK_GROUP_SIZE: " << kernel.getWorkGroupInfo<CL_KERNEL_WORK_GROUP_SIZE>(device) << "\n";
-
-        // Create memory buffers
-        Buffer bufferUV = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-        Buffer bufferUV2 = Buffer(context, CL_MEM_READ_ONLY, MEM_SIZE);
-
-        // Copy lists A and B to the memory buffers
-        queue.enqueueWriteBuffer(bufferUV, CL_TRUE, 0, MEM_SIZE, ab);
- 
-        NDRange global(X,Y);
-        NDRange local(1,256);
-
-        kernel.setArg(2, k);
-        kernel.setArg(3, f);
-        kernel.setArg(4, r_a);
-        kernel.setArg(5, r_b);
-        kernel.setArg(6, speed);
-
-        int iteration = 0;
-        float fps_avg = 0.0; // decaying average of fps
-        const int N_FRAMES_PER_DISPLAY = 2000;  // an even number, because of our double-buffering implementation
-        while(true) 
-        {
-            struct timeval tod_record;
-            double tod_before, tod_after, tod_elap;
-
-            gettimeofday(&tod_record, 0);
-            tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            // run a few iterations (without copying the data back)
-            for(int it=0;it<N_FRAMES_PER_DISPLAY/2;it++)
-            {
-                // (buffer-switching)
-
-                kernel.setArg(0, bufferUV);
-                kernel.setArg(1, bufferUV2); // output in UV2
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-
-                kernel.setArg(0, bufferUV2);
-                kernel.setArg(1, bufferUV); // output in UV
-                queue.enqueueNDRangeKernel(kernel, NullRange, global, local);
-                iteration++;
-            }
-
-            queue.enqueueReadBuffer(bufferUV, CL_TRUE, 0, MEM_SIZE, ab);
-
-            gettimeofday(&tod_record, 0);
-            tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-            tod_elap = tod_after - tod_before;
-
-            char msg[1000];
-            float fps = 0.0;     // frames per second
-            if (tod_elap > 0)
-                fps = ((float)N_FRAMES_PER_DISPLAY) / tod_elap;
-            // We display an exponential moving average of the fps measurement
-            fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-	        double Mcgs = (fps_avg * ((double)X) * ((double)Y)) / 1.0e6;
-            sprintf(msg,"GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-            // display:
-            {
-                int quitnow = display(ab,ab,ab,iteration,false,200.0f,2,10,msg);
-                if (quitnow)
-                    break;
-            }
-        }
-    } 
-    catch(Error error) 
-    {
-       std::cout << error.what() << "(" << error.err() << ")" << std::endl;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float ab[X][Y][2])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            // start with a uniform field with an approximate circle in the middle
-            //if(hypot(i%20-10/*-X/2*/,j%20-10/*-Y/2*/)<=frand(2,5)) {
-            if(hypot(i-X/3,(j-Y/2)/1.5)<=frand(2,5))
-            {
-                ab[i][j][0] = frand(0.0f,0.1f);
-                ab[i][j][1] = frand(0.9f,1.0f);
-            }
-            else 
-            {
-                ab[i][j][0] = frand(0.9f,1.0f);
-                ab[i][j][1] = frand(0.0f,0.1f);
-            }
-        }
-    }
-}
-
-bool display(float r[X][Y][2],float g[X][Y][2],float b[X][Y][2],
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-        
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-        
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-        if(write_video)
-        {
-            video = cvCreateVideoWriter(title,CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im2),1);
-            border = 20;
-        }
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<X;i++)
-        {
-            for(int j=0;j<Y;j++)
-            {
-                if(r) {
-                    val = r[i][j][0];
-                    if(val<minR) minR=val; if(val>maxR) maxR=val;
-                }
-                if(g) {
-                    val = g[i][j][0];
-                    if(val<minG) minG=val; if(val>maxG) maxG=val;
-                }
-                if(b) {
-                    val = b[i][j][0];
-                    if(val<minB) minB=val; if(val>maxB) maxB=val;
-                }
-            }
-        }
-    }
-    for(int i=0;i<X;i++)
-    {
-        for(int j=0;j<Y;j++)
-        {
-            if(r) {
-                float val = r[i][Y-j-1][0];
-                if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 2] = (uchar)val;
-            }
-            if(g) {
-                float val = g[i][Y-j-1][0];
-                if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 1] = (uchar)val;
-            }
-            if(b) {
-                float val = b[i][Y-j-1][0];
-                if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 0] = (uchar)val;
-            }
-        }
-    }
-
-    cvResize(im,im2);
-
-    char txt[100];
-    if(!write_video)
-    {
-        sprintf(txt,"%d",iteration);
-        cvPutText(im2,txt,cvPoint(20,20),&font,white);
-    }
-
-    cvPutText(im2,message,cvPoint(20,40),&font,white);
-
-    if(write_video)
-        cvWriteFrame(video,im2);
-
-    cvShowImage(title,im2);
-    
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        if(write_video)
-            cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
-
diff --git a/SpeedComparisons/GrayScott_OpenCL_float2/grayscott_kernel_float2.cl b/SpeedComparisons/GrayScott_OpenCL_float2/grayscott_kernel_float2.cl
deleted file mode 100644
index 364749c31..000000000
--- a/SpeedComparisons/GrayScott_OpenCL_float2/grayscott_kernel_float2.cl
+++ /dev/null
@@ -1,38 +0,0 @@
-__kernel void grayscott_compute(
-    __global float2 *UV,
-    __global float2 *UV2,
-    float k,float F,float D_u,float D_v,float delta_t)
-{
-    // Get the index of the current element
-    const int x = get_global_id(0);
-    const int y = get_global_id(1);
-    const int X = get_global_size(0);
-    const int Y = get_global_size(1);
-    const int i = x*Y+y;
-
-    const float2 uv = UV[i];
-    
-    // compute the Laplacians of a and b
-#ifdef WRAP
-    // speedy modulo operator for when X and Y are powers of 2
-    // http://forums.amd.com/devforum/messageview.cfm?catid=390&threadid=143648
-    const int xm1 = ((x-1+X) & (X-1));    // = ((x-1+X)%X);
-    const int xp1 = ((x+1) & (X-1));      // = ((x+1)%X);
-    const int ym1 = ((y-1+Y) & (Y-1));    // = ((y-1+Y)%Y);
-    const int yp1 = ((y+1) & (Y-1));      // = ((y+1)%Y);
-#else
-    const int xm1 = max(x-1,0);
-    const int xp1 = min(x+1,X-1);
-    const int ym1 = max(y-1,0);
-    const int yp1 = min(y+1,Y-1);
-#endif
-    const int iLeft = xm1*Y + y;
-    const int iRight = xp1*Y + y;
-    const int iUp = x*Y + ym1;
-    const int iDown = x*Y + yp1;
-    // Standard 5-point stencil
-    const float2 nabla_uv = UV[iLeft] + UV[iRight] + UV[iUp] + UV[iDown] - 4.0f*uv;
-
-    // compute the new value
-    UV2[i] = uv + delta_t * (float2)(D_u * nabla_uv.x - uv.x*uv.y*uv.y + F*(1.0f-uv.x), D_v * nabla_uv.y + uv.x*uv.y*uv.y - (F+k)*uv.y);
-}
diff --git a/SpeedComparisons/GrayScott_OpenCV/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenCV/CMakeLists.txt
deleted file mode 100644
index 9a34c1102..000000000
--- a/SpeedComparisons/GrayScott_OpenCV/CMakeLists.txt
+++ /dev/null
@@ -1,13 +0,0 @@
-project(GrayScott_OpenCV)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(GrayScott_OpenCV 
-  gray_scott_OpenCV.cpp 
-  ../Display/defs.h
-)
-
-TARGET_LINK_LIBRARIES(GrayScott_OpenCV ${OpenCV_LIBS} )
diff --git a/SpeedComparisons/GrayScott_OpenCV/gray_scott_OpenCV.cpp b/SpeedComparisons/GrayScott_OpenCV/gray_scott_OpenCV.cpp
deleted file mode 100644
index 75808efff..000000000
--- a/SpeedComparisons/GrayScott_OpenCV/gray_scott_OpenCV.cpp
+++ /dev/null
@@ -1,268 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(IplImage *a,IplImage *b);
-
-void compute(IplImage *a,IplImage *b,
-             IplImage *t1,IplImage *t2,
-             IplImage *abb,
-             float r_a,float r_b,float f,float k,
-             float speed);
-
-bool display(IplImage *r,IplImage *g,IplImage *b,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
-
-static int g_wrap = 0;
-static bool g_paramspace = 0;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            fprintf(stderr, "Unrecognized argument: '%s'\n", argv[i]);
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    // for spots:
-    float k = 0.064f;
-    float f = 0.035f;
-    // for stripes:
-    //float k = 0.06f;
-    //float f = 0.035f;
-    // for long stripes
-    //float k = 0.065f;
-    //float f = 0.056f;
-    // for dots and stripes
-    //float k = 0.064f;
-    //float f = 0.04f;
-    // for spiral waves:
-    //float k = 0.0475f;
-    //float f = 0.0118f;
-    float speed = 1.0f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    IplImage *a,*b;
-
-    // temp images:
-    IplImage *t1,*t2,*abb;
-
-    a = cvCreateImage(cvSize(X,Y),IPL_DEPTH_32F,1);
-    b = cvCreateImage(cvSize(X,Y),IPL_DEPTH_32F,1);
-    t1 = cvCreateImage(cvSize(X,Y),IPL_DEPTH_32F,1);
-    t2 = cvCreateImage(cvSize(X,Y),IPL_DEPTH_32F,1);
-    abb = cvCreateImage(cvSize(X,Y),IPL_DEPTH_32F,1);
-
-    // put the initial conditions into each cell
-    init(a,b);
-
-    const int N_FRAMES_PER_DISPLAY = 200;
-    int iteration = 0;
-    double fps_avg = 0.0; // decaying average of fps
-    while(true) 
-    {
-        struct timeval tod_record;
-        double tod_before, tod_after, tod_elapsed;
-        double fps = 0.0;     // frames per second
-
-        gettimeofday(&tod_record, 0);
-        tod_before = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        // compute:
-        for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-        {
-            compute(a,b,t1,t2,abb,r_a,r_b,f,k,speed);
-            iteration++;
-        }
-
-        gettimeofday(&tod_record, 0);
-        tod_after = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        tod_elapsed = tod_after - tod_before;
-        fps = ((double)N_FRAMES_PER_DISPLAY) / tod_elapsed;
-        // We display an exponential moving average of the fps measurement
-        fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-		double Mcgs = fps_avg * ((double) X) * ((double) Y) / 1.0e6;
-
-        char msg[1000];
-        sprintf(msg, "GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-        // display:
-        if(display(a,a,a,iteration,false,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-
-    cvReleaseImage(&a);
-    cvReleaseImage(&b);
-    cvReleaseImage(&t1);
-    cvReleaseImage(&t2);
-    cvReleaseImage(&abb);
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(IplImage *a,IplImage *b)
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            
-            if(hypot(i%50-25/*-X/2*/,j%50-25/*-Y/2*/)<=frand(2,5))
-            //if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5)) // start with a uniform field with an approximate circle in the middle
-            {
-                cvSet2D(a,i,j,cvScalar(0.0f));
-                cvSet2D(b,i,j,cvScalar(1.0f));
-            }
-            else 
-            {
-                cvSet2D(a,i,j,cvScalar(1.0f));
-                cvSet2D(b,i,j,cvScalar(0.0f));
-            }
-        }
-    }
-}
-
-// a += speed * ( r_a * dda - aval*bval*bval + f*(1-aval) );
-// b += speed * ( r_b * ddb + aval*bval*bval - (f+k)*bval );
-void compute(IplImage *a,IplImage *b,
-             IplImage *t1,IplImage *t2,
-             IplImage *abb,
-             float r_a,float r_b,float f,float k,float speed)
-{
-    cvMul(a,b,abb);
-    cvMul(abb,b,abb);
-    cvLaplace(a,t1,1);
-    cvLaplace(b,t2,1);
-    cvAddWeighted(t1,r_a,abb,-1.0,0.0,t1);
-    cvAddWeighted(t2,r_b,abb,1.0,0.0,t2);
-    cvAddWeighted(t1,1.0,a,-f,f,t1);
-    cvAddWeighted(t2,1.0,b,-(f+k),0.0,t2);
-    cvAddWeighted(a,1.0,t1,speed,1e-10f,a); // adding a teeny-tiny bit to avoid denormals
-    cvAddWeighted(b,1.0,t2,speed,1e-10f,b);
-}
-
-bool display(IplImage *r,IplImage *g,IplImage *b,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2,*float_im;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        float_im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_32F,3);
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-        
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-        
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-        if(write_video)
-        {
-            video = cvCreateVideoWriter(title,CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im2),1);
-        }
-    }
-
-    cvMerge(r,g,b,NULL,float_im);
-    cvConvertScale(float_im,im,manual_brighten);
-    cvResize(im,im2);
-
-    char txt[100];
-    if(!write_video)
-    {
-        sprintf(txt,"%d",iteration);
-        cvPutText(im2,txt,cvPoint(20,20),&font,white);
-    }
-
-    cvPutText(im2,message,cvPoint(20,40),&font,white);
-
-    if(write_video)
-        cvWriteFrame(video,im2);
-
-    cvShowImage(title,im2);
-    
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        if(write_video)
-            cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
-
diff --git a/SpeedComparisons/GrayScott_OpenMP/CMakeLists.txt b/SpeedComparisons/GrayScott_OpenMP/CMakeLists.txt
deleted file mode 100644
index 11889dfc9..000000000
--- a/SpeedComparisons/GrayScott_OpenMP/CMakeLists.txt
+++ /dev/null
@@ -1,23 +0,0 @@
-project(GrayScott_OpenMP)
-
-find_package( OpenCV REQUIRED )
-include_directories( ${OPENCV_INCLUDE_DIR} )
-link_libraries( ${OpenCV_LIBS} )
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-# only build the OpenMP version if OpenMP was found
-include(FindOpenMP)
-if(OPENMP_FOUND)
-  set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${OpenMP_C_FLAGS}")
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
-  set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${OpenMP_EXE_LINKER_FLAGS}")
-
-  add_executable(GrayScott_OpenMP
-    gray_scott_openmp.cpp 
-    ../Display/display.cpp
-    ../Display/display.h
-    ../Display/defs.h
-  )
-endif()
-
diff --git a/SpeedComparisons/GrayScott_OpenMP/gray_scott_openmp.cpp b/SpeedComparisons/GrayScott_OpenMP/gray_scott_openmp.cpp
deleted file mode 100644
index 2f1d4894b..000000000
--- a/SpeedComparisons/GrayScott_OpenMP/gray_scott_openmp.cpp
+++ /dev/null
@@ -1,247 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenMP:
-#include <omp.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y]);
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float r_a,float r_b,float f,float k,
-             float speed, bool parameter_space);
-
-static int g_nthreads;
-static int g_wrap = 1;
-static bool g_paramspace = 0;
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if (strcmp(argv[i],"-paramspace")==0) {
-            // do a parameter space plot, like in the Pearson paper
-            g_paramspace = true;
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            fprintf(stderr, "Unrecognized argument: '%s'\n", argv[i]);
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    // for spots:
-    float k = 0.064f;
-    float f = 0.035f;
-    // for stripes:
-    //float k = 0.06f;
-    //float f = 0.035f;
-    // for long stripes
-    //float k = 0.065f;
-    //float f = 0.056f;
-    // for dots and stripes
-    //float k = 0.064f;
-    //float f = 0.04f;
-    // for spiral waves:
-    //float k = 0.0475f;
-    //float f = 0.0118f;
-    float speed = 1.0f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-
-    const int N_FRAMES_PER_DISPLAY = 1000;
-    int iteration = 0;
-    double fps_avg = 0.0; // decaying average of fps
-    while(true) 
-    {
-        struct timeval tod_record;
-        double tod_before, tod_after, tod_elap;
-        double fps = 0.0;     // frames per second
-
-        gettimeofday(&tod_record, 0);
-        tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        // compute:
-        for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-        {
-            compute(a,b,da,db,r_a,r_b,f,k,speed,g_paramspace);
-            iteration++;
-        }
-
-        gettimeofday(&tod_record, 0);
-        tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        tod_elap = tod_after - tod_before;
-
-        char msg[1000];
-        fps = ((double)N_FRAMES_PER_DISPLAY) / tod_elap;
-        // We display an exponential moving average of the fps measurement
-        fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0); 
-		double Mcgs = fps_avg * ((double) X) * ((double) Y) / 1.0e6;
-        sprintf(msg,"GrayScott %d threads %0.2f fps %0.2f Mcgs", g_nthreads, fps_avg, Mcgs);
-
-        // display:
-        if(display(a,a,a,iteration,false,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void init(float a[X][Y],float b[X][Y])
-{
-  #pragma omp parallel default(shared)
-  {
-    g_nthreads = omp_get_num_threads();
-  }
-
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            // start with a uniform field with an approximate circle in the middle
-            //if(hypot(i%20-10/*-X/2*/,j%20-10/*-Y/2*/)<=frand(2,5)) {
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5))
-            {
-                a[i][j] = 0.0f;
-                b[i][j] = 1.0f;
-            }
-            else {
-                a[i][j] = 1;
-                b[i][j] = 0;
-            }
-            /*float v = frand(0.0f,1.0f);
-            a[i][j] = v;
-            b[i][j] = 1.0f-v;*/
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],
-             float da[X][Y],float db[X][Y],
-             float r_a,float r_b,float par_f,float par_k,float speed,
-             bool parameter_space)
-{
-    // compute change in each cell
-    #pragma omp parallel for
-    for(int i = 0; i < X; i++) 
-    {
-        int iprev,inext;
-        float f = par_f;
-        float k = par_k;
-        if (g_wrap) {
-            iprev = (i + X - 1) % X;
-            inext = (i + 1) % X;
-        } else {
-            iprev = max(0,i-1);
-            inext = min(X-1,i+1);
-        }
-
-        for(int j = 0; j < Y; j++) 
-        {
-            int jprev,jnext;
-            if (g_wrap) {
-                // toroidal
-                jprev = (j + Y - 1) % Y;
-                jnext = (j + 1) % Y;
-            } else {
-                jprev = max(0,j-1);
-                jnext = min(Y-1,j+1);
-            }
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            if (parameter_space) {
-                const float kmin=0.045f,kmax=0.07f,fmin=0.01f,fmax=0.09f;
-                // set f and k for this location (ignore the provided values of f and k)
-                k = kmin + i*(kmax-kmin)/X;
-                f = fmin + j*(fmax-fmin)/Y;
-            }
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = r_a * dda - aval*bval*bval + f*(1-aval);
-            db[i][j] = r_b * ddb + aval*bval*bval - (f+k)*bval;
-        }
-    }
-
-    // effect change
-    #pragma omp parallel for
-    for(int i = 0; i < X; i++) 
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] += speed * da[i][j];
-            b[i][j] += speed * db[i][j];
-            // kill denormals by adding a teeny tiny something (http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.20.1348&rank=4)
-            a[i][j] += 1e-10f;
-            b[i][j] += 1e-10f;
-        }
-    }
-}
-
diff --git a/SpeedComparisons/GrayScott_SSE/CMakeLists.txt b/SpeedComparisons/GrayScott_SSE/CMakeLists.txt
deleted file mode 100644
index 01b08353f..000000000
--- a/SpeedComparisons/GrayScott_SSE/CMakeLists.txt
+++ /dev/null
@@ -1,19 +0,0 @@
-project(GrayScott_SSE)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-if(MSVC)
-  add_definitions(/arch:SSE2)
-else()
-  add_definitions(-msse2)
-endif()
-
-add_executable(GrayScott_SSE
-  gray_scott_SSE.cpp 
-  ../Display/defs.h
-)
-
-TARGET_LINK_LIBRARIES(GrayScott_SSE ${OpenCV_LIBS} )
diff --git a/SpeedComparisons/GrayScott_SSE/gray_scott_SSE.cpp b/SpeedComparisons/GrayScott_SSE/gray_scott_SSE.cpp
deleted file mode 100644
index 63d3ab750..000000000
--- a/SpeedComparisons/GrayScott_SSE/gray_scott_SSE.cpp
+++ /dev/null
@@ -1,496 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// SSE:
-#include <xmmintrin.h>
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// If SIZE_OPTIONS is declared, the user can specify an image size with -width N and -height N options
-//#define SIZE_OPTIONS
-
-#ifndef SIZE_OPTIONS
-// local:
-#include "defs.h"
-#endif
-
-// consecutive horizontal SSE blocks lie end to end, to enable easy use of _mm_loadu_ps() for left and right
-
-const int SSE_BITS_PER_BLOCK = 128;
-const int FLOATS_PER_BLOCK = (SSE_BITS_PER_BLOCK/8) / sizeof(float);
-
-#ifndef SIZE_OPTIONS
-const int PADDED_X = X + 2*FLOATS_PER_BLOCK; // our toroidal wrap-around scheme uses a border that is copied from the other side each time
-const int PADDED_Y = Y + 2;
-const int X_BLOCKS = PADDED_X / FLOATS_PER_BLOCK;
-const int Y_BLOCKS = PADDED_Y;
-const int TOTAL_BLOCKS = X_BLOCKS * Y_BLOCKS;
-#else
-static int X = 256;
-static int Y = 256;
-static int PADDED_X = X + 2*FLOATS_PER_BLOCK; // our toroidal wrap-around scheme uses a border that is copied from the other side each time
-static int PADDED_Y = Y + 2;
-static int X_BLOCKS = PADDED_X / FLOATS_PER_BLOCK;
-static int Y_BLOCKS = PADDED_Y;
-static int TOTAL_BLOCKS = X_BLOCKS * Y_BLOCKS;
-#endif
-
-inline int at(int x,int y) { return y*PADDED_X+x; } 
-inline int block_at(int x,int y) { return y*X_BLOCKS+x; }
-
-void init(float *a,float *b,float *da,float *db);
-void compute(float *a,float *b,float *da,float *db,
-             const float r_a,const float r_b,const float f,const float k,
-             const float speed);
-bool display(float *r,float *g,float *b,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
-
-int main(int argc, char * * argv)
-{
-#ifdef SIZE_OPTIONS
-  for (int i = 1; i < argc; i++) {
-    if (0) {
-    } else if ((i+1<argc) && (strcmp(argv[i],"-height")==0)) {
-      // set height
-      i++; Y = atoi(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-size")==0)) {
-      // set both width and height
-      i++; Y = X = atoi(argv[i]);
-    } else if ((i+1<argc) && (strcmp(argv[i],"-width")==0)) {
-      // set width
-      i++; X = atoi(argv[i]);
-    } else {
-      fprintf(stderr, "Unrecognized argument, or parameter needed: '%s'\n", argv[i]);
-      exit(-1);
-    }
-  }
-  PADDED_X = X + 2*FLOATS_PER_BLOCK; // our toroidal wrap-around scheme uses a border that is copied from the other side each time
-  PADDED_Y = Y + 2;
-  X_BLOCKS = PADDED_X / FLOATS_PER_BLOCK;
-  Y_BLOCKS = PADDED_Y;
-  TOTAL_BLOCKS = X_BLOCKS * Y_BLOCKS;
-#endif
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    // for spots:
-    float k = 0.064f;
-    float f = 0.035f;
-    // for stripes:
-    //float k = 0.06f;
-    //float f = 0.035f;
-    // for long stripes
-    //float k = 0.065f;
-    //float f = 0.056f;
-    // for dots and stripes
-    //float k = 0.064f;
-    //float f = 0.04f;
-    // for spiral waves:
-    //float k = 0.0475f;
-    //float f = 0.0118f;
-    float speed = 1.0f;
-    // ----------------
-
-    const int n_cells = PADDED_X*PADDED_Y;
-    float *a = (float*)_mm_malloc(n_cells*sizeof(float),16);
-    float *b = (float*)_mm_malloc(n_cells*sizeof(float),16);
-    float *da = (float*)_mm_malloc(n_cells*sizeof(float),16);
-    float *db = (float*)_mm_malloc(n_cells*sizeof(float),16);
-
-    init(a,b,da,db);
-
-    const int N_FRAMES_PER_DISPLAY = 1000;
-    int iteration = 0;
-    double fps_avg = 0.0; // decaying average of fps
-    while(true)
-    {
-        struct timeval tod_record;
-        double tod_before, tod_after, tod_elapsed;
-        double fps = 0.0;     // frames per second
-
-        gettimeofday(&tod_record, 0);
-        tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        // compute:
-        for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-        {
-            compute(a,b,da,db,r_a,r_b,f,k,speed);
-            iteration++;
-        }
-
-        gettimeofday(&tod_record, 0);
-        tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        tod_elapsed = tod_after - tod_before;
-        fps = ((double)N_FRAMES_PER_DISPLAY) / tod_elapsed;
-        // We display an exponential moving average of the fps measurement
-        fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-        float Mcgs = (fps_avg * ((float)X) * ((float)Y)) / 1.0e6;
-
-        char msg[1000];
-        sprintf(msg,"GrayScott - %0.2f fps %0.3f Mcgs",fps_avg, Mcgs);
-
-        // display:
-        if(display(a,a,a,iteration,false,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-
-    _mm_free(a);
-    _mm_free(b);
-    _mm_free(da);
-    _mm_free(db);
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float *a,float *b,float *da,float *db)
-{
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-  /* On Intel we disable accurate handling of denorms and zeros. This is an
-     important speed optimization. */
-  int oldMXCSR = _mm_getcsr(); //read the old MXCSR setting
-  int newMXCSR = oldMXCSR | 0x8040; // set DAZ and FZ bits
-  _mm_setcsr( newMXCSR ); //write the new MXCSR setting to the MXCSR
-#endif
-
-    srand((unsigned int)time(NULL));
-
-    // figure the values
-    float val=1.0f;
-    for(int i = 0; i < PADDED_X; i++)
-    {
-        for(int j = 0; j < PADDED_Y; j++)
-        {
-            if(hypot(i-PADDED_X/3,(j-PADDED_Y/4)/1.5)<=frand(2,5)) // start with a uniform field with an approximate circle in the middle
-            {
-                a[at(i,j)] = 0.0f;
-                b[at(i,j)] = 1.0f;
-            }
-            else {
-                a[at(i,j)] = 1;
-                b[at(i,j)] = 0;
-            }
-            da[at(i,j)]=0.0f;
-            db[at(i,j)]=0.0f;
-        }
-    }
-}
-
-// some macros to make the next macros slightly more readable
-#define ADD(a,b) _mm_add_ps(a,b)
-#define MUL(a,b) _mm_mul_ps(a,b)
-#define SUB(a,b) _mm_sub_ps(a,b)
-#define SET(a) _mm_set1_ps(a)
-
-// we write the Gray-Scott computation as macros because functions and __m128 don't mix well
-
-#define LAPLACIAN(_a,_a_left,_a_right,_a_above,_a_below) ADD(ADD(ADD(ADD(MUL(_a,SET(-4.0f)),_a_left),_a_right),_a_above),_a_below)
-
-#define GRAYSCOTT_DA(_a,_abb,_a_left,_a_right,_a_above,_a_below,_r_a,_f) ADD(SUB(MUL(_r_a,LAPLACIAN(_a,_a_left,_a_right,_a_above,_a_below)),_abb),MUL(_f,SUB(SET(1.0f),_a)));
-// da = r_a*dda - aval*bval*bval + f*(1-aval)
-
-#define GRAYSCOTT_DB(_b,_abb,_b_left,_b_right,_b_above,_b_below,_r_b,_f_plus_k) SUB(ADD(MUL(_r_b,LAPLACIAN(_b,_b_left,_b_right,_b_above,_b_below)),_abb),MUL(_f_plus_k,_b));
-// db = r_b*ddb + aval*bval*bval - (f+k)*bval
-
-
-#if 0
-
-// we tried using 'raise' and 'lower' macros to reduce the amount of loading but it ended up slower
-
-// the same as _mm_loadu_ps(((float*)(&_here))-1) when _here follows _left in memory
-#define FLOAT_LEFT(_left,_here) _mm_shuffle_ps(_mm_shuffle_ps(_left,_here,_MM_SHUFFLE(0,0,3,3)),_here,_MM_SHUFFLE(2,1,2,0))
-// (Robert Munafo's RAISE function)
-
-// the same as _mm_loadu_ps(((float*)(&_here))+1) when _right follows _here in memory
-#define FLOAT_RIGHT(_here,_right) _mm_shuffle_ps(_here,_mm_move_ss(_here,_right),_MM_SHUFFLE(0,3,2,1))
-
-void compute(float *a,float *b,float *da,float *db,
-             const float r_a,const float r_b,const float f,const float k,
-             const float speed)
-{
-    /* 
-    {
-        // test our new FLOAT_LEFT and FLOAT_RIGHT macros:
-        float arr1[12] = {1,2,3,4,5,6,7,8,9,10,11,12};
-        float *arr2 = (float*)_mm_malloc(12*sizeof(float),16);
-        memcpy(arr2,arr1,12*sizeof(float));
-        __m128 *a = (__m128*)(arr2+0);
-        __m128 *b = (__m128*)(arr2+4);
-        __m128 *c = (__m128*)(arr2+8);
-        __m128 left = FLOAT_LEFT(*a,*b);
-        __m128 right = FLOAT_RIGHT(*b,*c);
-        printf("done"); // put a breakpoint here and look at left and right
-    }
-    */
-
-    const __m128 r_a_SSE = SET(r_a);
-    const __m128 r_b_SSE = SET(r_b);
-    const __m128 f_SSE = SET(f);
-    const __m128 f_plus_k_SSE = SET(f+k);
-    const __m128 speed_SSE = SET(speed);
-
-    // compute the rate of change
-    for(int j=1;j<Y_BLOCKS-1;j++) // we skip the top- and bottom-most blocks
-    {
-        __m128 *a1,*b1;
-        __m128 *a2 = ((__m128*)a)+block_at(0,j);
-        __m128 *a3 = ((__m128*)a)+block_at(1,j);
-        __m128 *b2 = ((__m128*)b)+block_at(0,j);
-        __m128 *b3 = ((__m128*)b)+block_at(1,j);
-        // (TODO: wrap-around)
-        for(int i=1;i<X_BLOCKS-1;i++) // we skip the left- and right-most blocks
-        {
-            // load the next blocks
-            a1=a2; a2=a3; a3 = ((__m128*)a)+block_at(i+1,j);
-            b1=b2; b2=b3; b3 = ((__m128*)b)+block_at(i+1,j);
-            __m128 *da_SSE = ((__m128*)da)+block_at(i,j);
-            __m128 *db_SSE = ((__m128*)db)+block_at(i,j);
-            // retrieve the neighboring cells
-            __m128 a_left = FLOAT_LEFT(*a1,*a2);
-            __m128 a_right = FLOAT_RIGHT(*a2,*a3);
-            __m128 *a_above = a2-X_BLOCKS;
-            __m128 *a_below = a2+X_BLOCKS;
-            __m128 b_left = FLOAT_LEFT(*b1,*b2);
-            __m128 b_right = FLOAT_RIGHT(*b2,*b3);
-            __m128 *b_above = b2-X_BLOCKS;
-            __m128 *b_below = b2+X_BLOCKS;
-            __m128 abb = _mm_mul_ps(_mm_mul_ps(*a2,*b2),*b2);
-            // compute the new rates of change of each chemical
-            *da_SSE = GRAYSCOTT_DA(*a2,abb,a_left,a_right,*a_above,*a_below,r_a_SSE,f_SSE);
-            *db_SSE = GRAYSCOTT_DB(*b2,abb,b_left,b_right,*b_above,*b_below,r_b_SSE,f_plus_k_SSE);
-        }
-    }
-
-    // apply the rate of change
-    for(int j=1;j<Y_BLOCKS-1;j++) // we skip the top- and bottom-most blocks
-    {
-        for(int i=1;i<X_BLOCKS-1;i++) // we skip the left- and right-most blocks
-        {
-            __m128 *a_SSE = ((__m128*)a)+block_at(i,j);
-            __m128 *b_SSE = ((__m128*)b)+block_at(i,j);
-            __m128 *da_SSE = ((__m128*)da)+block_at(i,j);
-            __m128 *db_SSE = ((__m128*)db)+block_at(i,j);
-            *a_SSE = ADD(*a_SSE,MUL(*da_SSE,speed_SSE)); // a[i] += speed * da[i];
-            *b_SSE = ADD(*b_SSE,MUL(*db_SSE,speed_SSE)); // b[i] += speed * db[i];
-        }
-    }
-}
-
-#else
-
-// original code (faster)
-
-void compute(float *a,float *b,float *da,float *db,
-             const float r_a,const float r_b,const float f,const float k,
-             const float speed)
-{
-    const __m128 r_a_SSE = SET(r_a);
-    const __m128 r_b_SSE = SET(r_b);
-    const __m128 f_SSE = SET(f);
-    const __m128 f_plus_k_SSE = SET(f+k);
-    const __m128 speed_SSE = SET(speed);
-
-    // compute the rate of change
-    for(int j=1;j<Y_BLOCKS-1;j++) // we skip the top- and bottom-most blocks
-    {
-        for(int i=1;i<X_BLOCKS-1;i++) // we skip the left- and right-most blocks
-        {
-            __m128 *a_SSE = ((__m128*)a)+block_at(i,j);
-            __m128 *b_SSE = ((__m128*)b)+block_at(i,j);
-            __m128 *da_SSE = ((__m128*)da)+block_at(i,j);
-            __m128 *db_SSE = ((__m128*)db)+block_at(i,j);
-            // retrieve the neighboring cells
-            __m128 a_left = _mm_loadu_ps(((float*)a_SSE)-1);
-            __m128 a_right = _mm_loadu_ps(((float*)a_SSE)+1);
-            __m128 *a_above = a_SSE-X_BLOCKS;
-            __m128 *a_below = a_SSE+X_BLOCKS;
-            __m128 b_left = _mm_loadu_ps(((float*)b_SSE)-1);
-            __m128 b_right = _mm_loadu_ps(((float*)b_SSE)+1);
-            __m128 *b_above = b_SSE-X_BLOCKS;
-            __m128 *b_below = b_SSE+X_BLOCKS;
-            __m128 abb = _mm_mul_ps(_mm_mul_ps(*a_SSE,*b_SSE),*b_SSE);
-            // compute the new rates of change of each chemical
-            *da_SSE = GRAYSCOTT_DA(*a_SSE,abb,a_left,a_right,*a_above,*a_below,r_a_SSE,f_SSE);
-            *db_SSE = GRAYSCOTT_DB(*b_SSE,abb,b_left,b_right,*b_above,*b_below,r_b_SSE,f_plus_k_SSE);
-        }
-    }
-
-    // apply the rate of change
-    for(int j=1;j<Y_BLOCKS-1;j++) // we skip the top- and bottom-most blocks
-    {
-        for(int i=1;i<X_BLOCKS-1;i++) // we skip the left- and right-most blocks
-        {
-            __m128 *a_SSE = ((__m128*)a)+block_at(i,j);
-            __m128 *b_SSE = ((__m128*)b)+block_at(i,j);
-            __m128 *da_SSE = ((__m128*)da)+block_at(i,j);
-            __m128 *db_SSE = ((__m128*)db)+block_at(i,j);
-            *a_SSE = ADD(*a_SSE,MUL(*da_SSE,speed_SSE)); // a[i] += speed * da[i];
-            *b_SSE = ADD(*b_SSE,MUL(*db_SSE,speed_SSE)); // b[i] += speed * db[i];
-        }
-    }
-
-    // copy the top and bottom rows of the active area to the other side
-    for(int i=1;i<X_BLOCKS-1;i++) 
-    {
-        *(((__m128*)a)+block_at(i,Y_BLOCKS-1)) = *(((__m128*)a)+block_at(i,1));
-        *(((__m128*)a)+block_at(i,0)) = *(((__m128*)a)+block_at(i,Y_BLOCKS-2));
-        *(((__m128*)b)+block_at(i,Y_BLOCKS-1)) = *(((__m128*)b)+block_at(i,1));
-        *(((__m128*)b)+block_at(i,0)) = *(((__m128*)b)+block_at(i,Y_BLOCKS-2));
-    }
-    // likewise for the left- and right-most columns
-    for(int j=1;j<Y_BLOCKS-1;j++) 
-    {
-        *(((__m128*)a)+block_at(X_BLOCKS-1,j)) = *(((__m128*)a)+block_at(1,j));
-        *(((__m128*)a)+block_at(0,j)) = *(((__m128*)a)+block_at(X_BLOCKS-2,j));
-        *(((__m128*)b)+block_at(X_BLOCKS-1,j)) = *(((__m128*)b)+block_at(1,j));
-        *(((__m128*)b)+block_at(0,j)) = *(((__m128*)b)+block_at(X_BLOCKS-2,j));
-    }
-}
-
-#endif
-
-bool display(float *r,float *g,float *b,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2,*im3;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(PADDED_X,PADDED_Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(PADDED_X*scale,PADDED_Y*scale),IPL_DEPTH_8U,3);
-
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<PADDED_X;i++)
-        {
-            for(int j=0;j<PADDED_Y;j++)
-            {
-                if(r) {
-                    val = r[at(i,j)];
-                    if(val<minR) minR=val; if(val>maxR) maxR=val;
-                }
-                if(g) {
-                    val = g[at(i,j)];
-                    if(val<minG) minG=val; if(val>maxG) maxG=val;
-                }
-                if(b) {
-                    val = b[at(i,j)];
-                    if(val<minB) minB=val; if(val>maxB) maxB=val;
-                }
-            }
-        }
-    }
-    for(int col=0;col<PADDED_X;col++)
-    {
-        for(int row=0;row<PADDED_Y;row++)
-        {
-            if(r) {
-                float val = r[at(col,row)];
-                if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 2] = (uchar)val;
-            }
-            if(g) {
-                float val = g[at(col,row)];
-                if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 1] = (uchar)val;
-            }
-            if(b) {
-                float val = b[at(col,row)];
-                if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + row*im->widthStep))[col*im->nChannels + 0] = (uchar)val;
-            }
-        }
-    }
-
-    cvResize(im,im2);
-
-    {
-        char txt[100];
-        sprintf(txt,"%d",iteration);
-        cvPutText(im2,txt,cvPoint(20,20),&font,white);
-        cvPutText(im2,message,cvPoint(20,40),&font,white);
-    }
-
-    cvShowImage(title,im2);
-
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        return true;
-    }
-    return false;
-}
diff --git a/SpeedComparisons/GrayScott_SSE_OpenMP/CMakeLists.txt b/SpeedComparisons/GrayScott_SSE_OpenMP/CMakeLists.txt
deleted file mode 100644
index a03fbb916..000000000
--- a/SpeedComparisons/GrayScott_SSE_OpenMP/CMakeLists.txt
+++ /dev/null
@@ -1,29 +0,0 @@
-project(GrayScott_SSE_OpenMP)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-if(MSVC)
-  add_definitions(/arch:SSE2)
-else()
-  add_definitions(-msse)
-endif()
-
-# only build the OpenMP version if OpenMP was found
-include(FindOpenMP)
-if(OPENMP_FOUND)
-  set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${OpenMP_C_FLAGS}")
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
-  set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${OpenMP_EXE_LINKER_FLAGS}")
-
-  add_executable(GrayScott_SSE_OpenMP
-    gray_scott_SSE_OpenMP.cpp 
-    ../Display/display.cpp
-    ../Display/display.h
-    ../Display/defs.h
-  )
-
-  TARGET_LINK_LIBRARIES(GrayScott_SSE_OpenMP ${OpenCV_LIBS} )
-endif()
diff --git a/SpeedComparisons/GrayScott_SSE_OpenMP/gray_scott_SSE_OpenMP.cpp b/SpeedComparisons/GrayScott_SSE_OpenMP/gray_scott_SSE_OpenMP.cpp
deleted file mode 100644
index 9387e6dfd..000000000
--- a/SpeedComparisons/GrayScott_SSE_OpenMP/gray_scott_SSE_OpenMP.cpp
+++ /dev/null
@@ -1,388 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// SSE:
-#include <xmmintrin.h>
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// local:
-#include "defs.h"
-//#include "display.h"
-
-// OpenMP:
-#include <omp.h>
-
-// consecutive horizontal SSE blocks lie end to end, to enable easy use of _mm_loadu_ps() for left and right
-
-const int SSE_BITS_PER_BLOCK = 128;
-const int FLOATS_PER_BLOCK = (SSE_BITS_PER_BLOCK/8) / sizeof(float);
-const int PADDED_X = X + 2*FLOATS_PER_BLOCK; // our toroidal wrap-around scheme uses a border that is copied from the other side each time
-const int PADDED_Y = Y + 2;
-const int X_BLOCKS = PADDED_X / FLOATS_PER_BLOCK;
-const int Y_BLOCKS = PADDED_Y;
-const int TOTAL_BLOCKS = X_BLOCKS * Y_BLOCKS;
-
-inline int at(int x,int y) { return y*PADDED_X+x; } 
-inline int block_at(int x,int y) { return y*X_BLOCKS+x; }
-
-void init(float *a,float *b,float *da,float *db);
-void compute(float *a,float *b,float *da,float *db,
-             const float r_a,const float r_b,const float f,const float k,
-             const float speed);
-bool display(float *r,float *g,float *b,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message);
-
-int main()
-{
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    float r_a = 0.082f;
-    float r_b = 0.041f;
-
-    // for spots:
-    float k = 0.064f;
-    float f = 0.035f;
-    // for stripes:
-    //float k = 0.06f;
-    //float f = 0.035f;
-    // for long stripes
-    //float k = 0.065f;
-    //float f = 0.056f;
-    // for dots and stripes
-    //float k = 0.064f;
-    //float f = 0.04f;
-    // for spiral waves:
-    //float k = 0.0475f;
-    //float f = 0.0118f;
-    float speed = 1.0f;
-    // ----------------
-
-    const int n_cells = PADDED_X*PADDED_Y;
-    float *a = (float*)_mm_malloc(n_cells*sizeof(float),16);
-    float *b = (float*)_mm_malloc(n_cells*sizeof(float),16);
-    float *da = (float*)_mm_malloc(n_cells*sizeof(float),16);
-    float *db = (float*)_mm_malloc(n_cells*sizeof(float),16);
-
-    init(a,b,da,db);
-
-    const int N_FRAMES_PER_DISPLAY = 1000;
-    int iteration = 0;
-    double fps_avg = 0.0; // decaying average of fps
-    while(true)
-    {
-        struct timeval tod_record;
-        double tod_before, tod_after, tod_elapsed;
-        double fps = 0.0;     // frames per second
-
-        gettimeofday(&tod_record, 0);
-        tod_before = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        // compute:
-        for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-        {
-            compute(a,b,da,db,r_a,r_b,f,k,speed);
-            iteration++;
-        }
-
-        gettimeofday(&tod_record, 0);
-        tod_after = ((double) (tod_record.tv_sec))
-                                    + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        tod_elapsed = tod_after - tod_before;
-        fps = ((double)N_FRAMES_PER_DISPLAY) / tod_elapsed;
-        // We display an exponential moving average of the fps measurement
-        fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-		double Mcgs = fps_avg * ((double) X) * ((double) Y) / 1.0e6;
-
-        char msg[1000];
-        sprintf(msg,"GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-        // display:
-        if(display(a,a,a,iteration,false,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-
-    _mm_free(a);
-    _mm_free(b);
-    _mm_free(da);
-    _mm_free(db);
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float *a,float *b,float *da,float *db)
-{
-#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__) || 	defined(_M_X64) || defined(_M_IX86))
-  /* On Intel we disable accurate handling of denorms and zeros. This is an
-     important speed optimization. */
-  int oldMXCSR = _mm_getcsr(); //read the old MXCSR setting
-  int newMXCSR = oldMXCSR | 0x8040; // set DAZ and FZ bits
-  _mm_setcsr( newMXCSR ); //write the new MXCSR setting to the MXCSR
-#endif
-
-    srand((unsigned int)time(NULL));
-
-    // figure the values
-    float val=1.0f;
-    for(int i = 0; i < PADDED_X; i++)
-    {
-        for(int j = 0; j < PADDED_Y; j++)
-        {
-            if(hypot(i-PADDED_X/2,(j-PADDED_Y/2)/1.5)<=frand(2,5)) // start with a uniform field with an approximate circle in the middle
-            {
-                a[at(i,j)] = 0.0f;
-                b[at(i,j)] = 1.0f;
-            }
-            else {
-                a[at(i,j)] = 1;
-                b[at(i,j)] = 0;
-            }
-            da[at(i,j)]=0.0f;
-            db[at(i,j)]=0.0f;
-        }
-    }
-}
-
-// some macros to make the next macros slightly more readable
-#define ADD(a,b) _mm_add_ps(a,b)
-#define MUL(a,b) _mm_mul_ps(a,b)
-#define SUB(a,b) _mm_sub_ps(a,b)
-#define SET(a) _mm_set1_ps(a)
-
-// we write the Gray-Scott computation as macros because functions and __m128 don't mix well
-
-#define LAPLACIAN(_a,_a_left,_a_right,_a_above,_a_below) ADD(ADD(ADD(ADD(MUL(_a,SET(-4.0f)),_a_left),_a_right),_a_above),_a_below)
-
-#define GRAYSCOTT_DA(_a,_abb,_a_left,_a_right,_a_above,_a_below,_r_a,_f) ADD(SUB(MUL(_r_a,LAPLACIAN(_a,_a_left,_a_right,_a_above,_a_below)),_abb),MUL(_f,SUB(SET(1.0f),_a)));
-// da = r_a*dda - aval*bval*bval + f*(1-aval)
-
-#define GRAYSCOTT_DB(_b,_abb,_b_left,_b_right,_b_above,_b_below,_r_b,_f_plus_k) SUB(ADD(MUL(_r_b,LAPLACIAN(_b,_b_left,_b_right,_b_above,_b_below)),_abb),MUL(_f_plus_k,_b));
-// db = r_b*ddb + aval*bval*bval - (f+k)*bval
-
-void compute(float *a,float *b,float *da,float *db,
-             const float r_a,const float r_b,const float f,const float k,
-             const float speed)
-{
-    const __m128 r_a_SSE = SET(r_a);
-    const __m128 r_b_SSE = SET(r_b);
-    const __m128 f_SSE = SET(f);
-    const __m128 f_plus_k_SSE = SET(f+k);
-    const __m128 speed_SSE = SET(speed);
-
-    // compute the rates of change
-    #pragma omp parallel for
-    for(int j=1;j<Y_BLOCKS-1;j++) // we skip the top- and bottom-most blocks
-    {
-        for(int i=1;i<X_BLOCKS-1;i++) // we skip the left- and right-most blocks
-        {
-            __m128 *a_SSE = ((__m128*)a)+block_at(i,j);
-            __m128 *b_SSE = ((__m128*)b)+block_at(i,j);
-            __m128 *da_SSE = ((__m128*)da)+block_at(i,j);
-            __m128 *db_SSE = ((__m128*)db)+block_at(i,j);
-            // retrieve the neighboring cells
-            __m128 a_left = _mm_loadu_ps(((float*)a_SSE)-1);
-            __m128 a_right = _mm_loadu_ps(((float*)a_SSE)+1);
-            __m128 *a_above = a_SSE-X_BLOCKS;
-            __m128 *a_below = a_SSE+X_BLOCKS;
-            __m128 b_left = _mm_loadu_ps(((float*)b_SSE)-1);
-            __m128 b_right = _mm_loadu_ps(((float*)b_SSE)+1);
-            __m128 *b_above = b_SSE-X_BLOCKS;
-            __m128 *b_below = b_SSE+X_BLOCKS;
-            __m128 abb = _mm_mul_ps(_mm_mul_ps(*a_SSE,*b_SSE),*b_SSE);
-            // compute the new rates of change of each chemical
-            *da_SSE = GRAYSCOTT_DA(*a_SSE,abb,a_left,a_right,*a_above,*a_below,r_a_SSE,f_SSE);
-            *db_SSE = GRAYSCOTT_DB(*b_SSE,abb,b_left,b_right,*b_above,*b_below,r_b_SSE,f_plus_k_SSE);
-        }
-    }
-
-    // apply the rate of change
-    #pragma omp parallel for
-    for(int j=1;j<Y_BLOCKS-1;j++)
-    {
-        for(int i=1;i<X_BLOCKS-1;i++)
-        {
-            __m128 *a_SSE = ((__m128*)a)+block_at(i,j);
-            __m128 *b_SSE = ((__m128*)b)+block_at(i,j);
-            __m128 *da_SSE = ((__m128*)da)+block_at(i,j);
-            __m128 *db_SSE = ((__m128*)db)+block_at(i,j);
-            *a_SSE = ADD(*a_SSE,MUL(*da_SSE,speed_SSE)); // a[i] += speed * da[i];
-            *b_SSE = ADD(*b_SSE,MUL(*db_SSE,speed_SSE)); // b[i] += speed * db[i];
-        }
-    }
-
-    // copy the top and bottom rows of the active area to the other side
-    for(int i=1;i<X_BLOCKS-1;i++) 
-    {
-        *(((__m128*)a)+block_at(i,Y_BLOCKS-1)) = *(((__m128*)a)+block_at(i,1));
-        *(((__m128*)a)+block_at(i,0)) = *(((__m128*)a)+block_at(i,Y_BLOCKS-2));
-        *(((__m128*)b)+block_at(i,Y_BLOCKS-1)) = *(((__m128*)b)+block_at(i,1));
-        *(((__m128*)b)+block_at(i,0)) = *(((__m128*)b)+block_at(i,Y_BLOCKS-2));
-    }
-    // likewise for the left- and right-most columns
-    #pragma omp parallel for
-    for(int j=1;j<Y_BLOCKS-1;j++) 
-    {
-        *(((__m128*)a)+block_at(X_BLOCKS-1,j)) = *(((__m128*)a)+block_at(1,j));
-        *(((__m128*)a)+block_at(0,j)) = *(((__m128*)a)+block_at(X_BLOCKS-2,j));
-        *(((__m128*)b)+block_at(X_BLOCKS-1,j)) = *(((__m128*)b)+block_at(1,j));
-        *(((__m128*)b)+block_at(0,j)) = *(((__m128*)b)+block_at(X_BLOCKS-2,j));
-    }
-}
-
-bool display(float *r,float *g,float *b,
-             int iteration,bool auto_brighten,float manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2,*im3;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(PADDED_X,PADDED_Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(PADDED_X*scale,PADDED_Y*scale),IPL_DEPTH_8U,3);
-        im3 = cvCreateImage(cvSize(PADDED_X*scale+border*2,PADDED_Y*scale+border),IPL_DEPTH_8U,3);
-
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-        if(write_video)
-        {
-            video = cvCreateVideoWriter(title,CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im3),1);
-            border = 20;
-        }
-    }
-
-    // convert float arrays to IplImage for OpenCV to display
-    float val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<PADDED_X;i++)
-        {
-            for(int j=0;j<PADDED_Y;j++)
-            {
-                if(r) {
-                    val = r[at(i,j)];
-                    if(val<minR) minR=val; if(val>maxR) maxR=val;
-                }
-                if(g) {
-                    val = g[at(i,j)];
-                    if(val<minG) minG=val; if(val>maxG) maxG=val;
-                }
-                if(b) {
-                    val = b[at(i,j)];
-                    if(val<minB) minB=val; if(val>maxB) maxB=val;
-                }
-            }
-        }
-    }
-    for(int i=0;i<PADDED_X;i++)
-    {
-        for(int j=0;j<PADDED_Y;j++)
-        {
-            if(r) {
-                float val = r[at(i,PADDED_Y-j-1)];
-                if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 2] = (uchar)val;
-            }
-            if(g) {
-                float val = g[at(i,PADDED_Y-j-1)];
-                if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 1] = (uchar)val;
-            }
-            if(b) {
-                float val = b[at(i,PADDED_Y-j-1)];
-                if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 0] = (uchar)val;
-            }
-        }
-    }
-
-    cvResize(im,im2);
-    cvCopyMakeBorder(im2,im3,cvPoint(border*2,0),IPL_BORDER_CONSTANT);
-
-    char txt[100];
-    if(!write_video)
-    {
-        sprintf(txt,"%d",iteration);
-        cvPutText(im3,txt,cvPoint(20,20),&font,white);
-    }
-
-    cvPutText(im3,message,cvPoint(20,40),&font,white);
-
-    if(write_video)
-        cvWriteFrame(video,im3);
-
-    cvShowImage(title,im3);
-
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        if(write_video)
-            cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
diff --git a/SpeedComparisons/GrayScott_double/CMakeLists.txt b/SpeedComparisons/GrayScott_double/CMakeLists.txt
deleted file mode 100644
index 8a6e0449d..000000000
--- a/SpeedComparisons/GrayScott_double/CMakeLists.txt
+++ /dev/null
@@ -1,13 +0,0 @@
-project(GrayScott_double)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(GrayScott_double 
-  gray_scott_double.cpp 
-  ../Display/defs.h
-)
-
-TARGET_LINK_LIBRARIES(GrayScott_double ${OpenCV_LIBS} )
diff --git a/SpeedComparisons/GrayScott_double/gray_scott_double.cpp b/SpeedComparisons/GrayScott_double/gray_scott_double.cpp
deleted file mode 100644
index 7dc5d884c..000000000
--- a/SpeedComparisons/GrayScott_double/gray_scott_double.cpp
+++ /dev/null
@@ -1,375 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
-#include <math.h>
-
-#ifdef _WIN32
-    #include <sys/timeb.h>
-    #include <sys/types.h>
-    #include <winsock.h>
-    // http://www.linuxjournal.com/article/5574
-    void gettimeofday(struct timeval* t,void* timezone)
-    {       struct _timeb timebuffer;
-          _ftime( &timebuffer );
-          t->tv_sec=timebuffer.time;
-          t->tv_usec=1000*timebuffer.millitm;
-    }
-#else
-    #include <sys/time.h>
-#endif
-
-// OpenCV:
-#include <cv.h>
-#include <highgui.h>
-
-// stdlib
-#include <stdio.h>
-
-// local:
-#include "defs.h"
-
-static int g_wrap = 1;
-static bool g_paramspace = 0;
-
-bool display_gsd(double r[X][Y],double g[X][Y],double b[X][Y],
-             int iteration,bool auto_brighten,double manual_brighten,
-             int scale,int delay_ms,const char* message)
-{
-    static bool need_init = true;
-    static bool write_video = false;
-
-    static IplImage *im,*im2,*im3;
-    static int border = 0;
-    static CvFont font;
-    static CvVideoWriter *video;
-    static const CvScalar white = cvScalar(255,255,255);
-
-    const char *title = "Press ESC to quit";
-
-    if(need_init)
-    {
-        need_init = false;
-
-        im = cvCreateImage(cvSize(X,Y),IPL_DEPTH_8U,3);
-        cvSet(im,cvScalar(0,0,0));
-        im2 = cvCreateImage(cvSize(X*scale,Y*scale),IPL_DEPTH_8U,3);
-        im3 = cvCreateImage(cvSize(X*scale+border*2,Y*scale+border),IPL_DEPTH_8U,3);
-        
-        cvNamedWindow(title,CV_WINDOW_AUTOSIZE);
-        
-        double hScale=0.4;
-        double vScale=0.4;
-        int lineWidth=1;
-        cvInitFont(&font,CV_FONT_HERSHEY_COMPLEX,hScale,vScale,0,lineWidth,CV_AA);
-
-        if(write_video)
-        {
-            video = cvCreateVideoWriter(title,CV_FOURCC('D','I','V','X'),25.0,cvGetSize(im3),1);
-            border = 20;
-        }
-    }
-
-    // convert double arrays to IplImage for OpenCV to display
-    double val,minR=FLT_MAX,maxR=-FLT_MAX,minG=FLT_MAX,maxG=-FLT_MAX,minB=FLT_MAX,maxB=-FLT_MAX;
-    if(auto_brighten)
-    {
-        for(int i=0;i<X;i++)
-        {
-            for(int j=0;j<Y;j++)
-            {
-                if(r) {
-                    val = r[i][j];
-                    if(val<minR) minR=val; if(val>maxR) maxR=val;
-                }
-                if(g) {
-                    val = g[i][j];
-                    if(val<minG) minG=val; if(val>maxG) maxG=val;
-                }
-                if(b) {
-                    val = b[i][j];
-                    if(val<minB) minB=val; if(val>maxB) maxB=val;
-                }
-            }
-        }
-    }
-    #pragma omp parallel for
-    for(int i=0;i<X;i++)
-    {
-        for(int j=0;j<Y;j++)
-        {
-            if(r) {
-                double val = r[i][Y-j-1];
-                if(auto_brighten) val = 255.0f * (val-minR) / (maxR-minR);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 2] = (uchar)val;
-            }
-            if(g) {
-                double val = g[i][Y-j-1];
-                if(auto_brighten) val = 255.0f * (val-minG) / (maxG-minG);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 1] = (uchar)val;
-            }
-            if(b) {
-                double val = b[i][Y-j-1];
-                if(auto_brighten) val = 255.0f * (val-minB) / (maxB-minB);
-                else val *= manual_brighten;
-                if(val<0) val=0; if(val>255) val=255;
-                ((uchar *)(im->imageData + j*im->widthStep))[i*im->nChannels + 0] = (uchar)val;
-            }
-        }
-    }
-
-    cvResize(im,im2);
-    cvCopyMakeBorder(im2,im3,cvPoint(border*2,0),IPL_BORDER_CONSTANT);
-
-    char txt[100];
-    if(!write_video)
-    {
-        sprintf(txt,"%d",iteration);
-        cvPutText(im3,txt,cvPoint(20,20),&font,white);
-
-        // DEBUG:
-        sprintf(txt,"%.4f,%.4f,%.4f",r[0][0],g[0][0],b[0][0]);
-        //cvPutText(im3,txt,cvPoint(20,40),&font,white);
-    }
-
-    // DEBUG:
-    if(write_video)
-    {
-        cvPutText(im3,"0.06",cvPoint(5,15),&font,white);
-        cvPutText(im3,"F",cvPoint(5,im2->height/2),&font,white);
-        cvPutText(im3,"0.00",cvPoint(5,im2->height),&font,white);
-        cvPutText(im3,"0.03",cvPoint(border*2-10,im2->height+15),&font,white);
-        cvPutText(im3,"K",cvPoint(border*2+im2->width/2,im2->height+15),&font,white);
-        cvPutText(im3,"0.07",cvPoint(im3->width-35,im2->height+15),&font,white);
-    }
-
-    cvPutText(im3,message,cvPoint(20,40),&font,white);
-
-    if(write_video)
-        cvWriteFrame(video,im3);
-
-    cvShowImage(title,im3);
-    
-    int key = cvWaitKey(delay_ms); // allow time for the image to be drawn
-    if(key==27) // did user ask to quit?
-    {
-        cvDestroyWindow(title);
-        cvReleaseImage(&im);
-        cvReleaseImage(&im2);
-        if(write_video)
-            cvReleaseVideoWriter(&video);
-        return true;
-    }
-    return false;
-}
-
-void init(double a[X][Y],double b[X][Y]);
-
-void compute(double a[X][Y],double b[X][Y],
-             double da[X][Y],double db[X][Y],
-             double r_a,double r_b,double f,double k,
-             double speed,
-             bool parameter_space);
-
-int main(int argc, char * * argv)
-{
-    for (int i = 1; i < argc; i++) {
-        if (0) {
-        } else if (strcmp(argv[i],"-paramspace")==0) {
-            // do a parameter space plot, like in the Pearson paper
-            g_paramspace = true;
-        } else if (strcmp(argv[i],"-wrap")==0) {
-            // patterns wrap around ("torus", also called "continuous boundary
-            // condition")
-            g_wrap = 1;
-        } else {
-            fprintf(stderr, "Unrecognized argument: '%s'\n", argv[i]);
-            exit(-1);
-        }
-    }
-
-    // Here we implement the Gray-Scott model, as described here:
-    // http://www.cc.gatech.edu/~turk/bio_sim/hw3.html
-    // http://arxiv.org/abs/patt-sol/9304003
-
-    // -- parameters --
-    double r_a = 0.082f;
-    double r_b = 0.041f;
-
-    // for spots:
-    double k = 0.064f;
-    double f = 0.035f;
-    // for stripes:
-    //double k = 0.06f;
-    //double f = 0.035f;
-    // for long stripes
-    //double k = 0.065f;
-    //double f = 0.056f;
-    // for dots and stripes
-    //double k = 0.064f;
-    //double f = 0.04f;
-    // for spiral waves:
-    //double k = 0.0475f;
-    //double f = 0.0118f;
-    double speed = 1.0f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    double a[X][Y], b[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    double da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b);
-    
-    const int N_FRAMES_PER_DISPLAY = 200;
-    int iteration = 0;
-    double fps_avg = 0.0; // decaying average of fps
-    while(true) 
-    {
-        struct timeval tod_record;
-        double tod_before, tod_after, tod_elapsed;
-        double fps = 0.0;     // frames per second
-
-        gettimeofday(&tod_record, 0);
-        tod_before = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        // compute:
-        for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-        {
-            compute(a,b,da,db,r_a,r_b,f,k,speed,g_paramspace);
-            iteration++;
-        }
-
-        gettimeofday(&tod_record, 0);
-        tod_after = ((double) (tod_record.tv_sec))
-                                + ((double) (tod_record.tv_usec)) / 1.0e6;
-
-        tod_elapsed = tod_after - tod_before;
-        fps = ((double)N_FRAMES_PER_DISPLAY) / tod_elapsed;
-        // We display an exponential moving average of the fps measurement
-        fps_avg = (fps_avg == 0) ? fps : (((fps_avg * 10.0) + fps) / 11.0);
-		double Mcgs = fps_avg * ((double) X) * ((double) Y) / 1.0e6;
-
-        char msg[1000];
-        sprintf(msg,"GrayScott - %0.2f fps %0.2f Mcgs", fps_avg, Mcgs);
-
-        // display:
-        if(display_gsd(a,a,a,iteration,false,200.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-double frand(double lower,double upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-#ifndef max
-#define max(a,b) (((a) > (b)) ? (a) : (b))
-#define min(a,b) (((a) < (b)) ? (a) : (b))
-#endif
-
-void init(double a[X][Y],double b[X][Y])
-{
-    srand((unsigned int)time(NULL));
-    
-    // figure the values
-    for(int i = 0; i < X; i++) {
-        for(int j = 0; j < Y; j++) {
-            
-            //if(hypot(i%10-5/*-X/2*/,j%10-5/*-Y/2*/)<=2.0f)//frand(2,3))
-            if(hypot(i-X/2,(j-Y/2)/1.5)<=frand(2,5)) // start with a uniform field with an approximate circle in the middle
-            {
-                a[i][j] = 0.0f;
-                b[i][j] = 1.0f;
-            }
-            else {
-                a[i][j] = 1;
-                b[i][j] = 0;
-            }
-            //double v = frand(0.0f,1.0f);
-            //a[i][j] = v;
-            //b[i][j] = 1.0f-v;
-            //a[i][j] += frand(-0.01f,0.01f);
-            //b[i][j] += frand(-0.01f,0.01f);
-        }
-    }
-}
-
-void compute(double a[X][Y],double b[X][Y],
-             double da[X][Y],double db[X][Y],
-             double r_a,double r_b,double f,double k,double speed,
-             bool parameter_space)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) {
-        int iprev,inext;
-        if (g_wrap) {
-            iprev = (i + X - 1) % X;
-            inext = (i + 1) % X;
-        } else {
-            iprev = max(0,i-1);
-            inext = min(X-1,i+1);
-        }
-
-        for(int j = 0; j < Y; j++) {
-            int jprev,jnext;
-            if (g_wrap) {
-                jprev = (j + Y - 1) % Y;
-                jnext = (j + 1) % Y;
-            } else {
-                jprev = max(0,j-1);
-                jnext = min(Y-1,j+1);
-            }
-
-            double aval = a[i][j];
-            double bval = b[i][j];
-
-            if (parameter_space) {
-                const double kmin=0.045f,kmax=0.07f,fmin=0.01f,fmax=0.09f;
-                // set f and k for this location (ignore the provided values of f and k)
-                k = kmin + i*(kmax-kmin)/X;
-                f = fmin + j*(fmax-fmin)/Y;
-            }
-
-            // compute the Laplacians of a and b
-            double dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4*aval;
-            double ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4*bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = r_a * dda - aval*bval*bval + f*(1-aval);
-            db[i][j] = r_b * ddb + aval*bval*bval - (f+k)*bval;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++) 
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] += speed * da[i][j];
-            b[i][j] += speed * db[i][j];
-            // kill denormals by adding a teeny tiny something (http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.20.1348&rank=4)
-            a[i][j] += 1e-20;
-            b[i][j] += 1e-20;
-        }
-    }
-}
-
diff --git a/Ready/TODO.txt b/TODO.txt
similarity index 100%
rename from Ready/TODO.txt
rename to TODO.txt
diff --git a/TuringSpots/CMakeLists.txt b/TuringSpots/CMakeLists.txt
deleted file mode 100644
index bc1ceaa79..000000000
--- a/TuringSpots/CMakeLists.txt
+++ /dev/null
@@ -1,14 +0,0 @@
-project(TuringSpots)
-
-FIND_PACKAGE(OpenCV REQUIRED)
-INCLUDE_DIRECTORIES( ${OPENCV_INCLUDE_DIR})
-
-INCLUDE_DIRECTORIES( "../Display" )
-
-add_executable(TuringSpots 
-  turing_spots.cpp 
-  ../Display/display.cpp
-  ../Display/display.h
-)
-
-TARGET_LINK_LIBRARIES(TuringSpots ${OpenCV_LIBS} )
diff --git a/TuringSpots/turing_spots.cpp b/TuringSpots/turing_spots.cpp
deleted file mode 100644
index 0c6a6a53e..000000000
--- a/TuringSpots/turing_spots.cpp
+++ /dev/null
@@ -1,138 +0,0 @@
-/*
-
-A port of part of Greg Turk's reaction-diffusion code, from:
-http://www.cc.gatech.edu/~turk/reaction_diffusion/reaction_diffusion.html
-
-See README.txt for more details.
-
-See also:
-http://www1.cse.wustl.edu/~faanly/materials/Sketching_RD_Texture.pdf
-
-*/
-
-// stdlib:
-#include <time.h>
-#include <stdlib.h>
-#include <stdio.h>
-
-// local:
-#include "defs.h"
-#include "display.h"
-
-void init(float a[X][Y],float b[X][Y],float beta[X][Y],
-          float a_steady,float b_steady,float beta_init,float beta_rand);
-
-void compute(float a[X][Y],float b[X][Y],float beta[X][Y],
-             float da[X][Y],float db[X][Y],
-             float diff1,float diff2,float p1,float speed);
-
-int main()
-{
-    // -- parameters --
-    float diff1 = 0.125f;
-    float diff2 = 0.03125f;
-    float s = 0.0125f;
-    float speed = 1.0f;
-    float a_steady = 4.0f;
-    float b_steady = 4.0f;
-    float beta_init = 12.0f;
-    float beta_rand = 0.1f;
-    // ----------------
-    
-    // these arrays store the chemical concentrations:
-    float a[X][Y], b[X][Y], beta[X][Y];
-    // these arrays store the rate of change of those chemicals:
-    float da[X][Y], db[X][Y];
-
-    // put the initial conditions into each cell
-    init(a,b,beta,a_steady,b_steady,beta_init,beta_rand);
-    
-	clock_t start,end;
-
-	const int N_FRAMES_PER_DISPLAY = 100;
-    int iteration = 0;
-    while(true)
-    {
-		start = clock();
-
-        // compute:
-		for(int it=0;it<N_FRAMES_PER_DISPLAY;it++)
-		{
-            compute(a,b,beta,da,db,diff1,diff2,s,speed);
-            iteration++;
-        }
-
-		end = clock();
-
-		char msg[1000];
-		sprintf(msg,"Turing spots - %0.2f fps",N_FRAMES_PER_DISPLAY / ((end-start)/(float)CLOCKS_PER_SEC));
-
-        // display:
-        if(display(a,a,b,iteration,false,30.0f,2,10,msg)) // did user ask to quit?
-            break;
-    }
-}
-
-// return a random value between lower and upper
-float frand(float lower,float upper)
-{
-    return lower + rand()*(upper-lower)/RAND_MAX;
-}
-
-void init(float a[X][Y],float b[X][Y],float beta[X][Y],
-          float a_steady,float b_steady,float beta_init,float beta_rand)
-{
-    srand((unsigned int)time(NULL));
-    for(int i = 0; i < X; i++)
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] = a_steady;
-            b[i][j] = b_steady;
-            //if(i<X/2)
-                beta[i][j] = beta_init + frand(-beta_rand, beta_rand);
-            /*else
-                beta[i][j] = beta_init;*/
-        }
-    }
-}
-
-void compute(float a[X][Y],float b[X][Y],float beta[X][Y],
-             float da[X][Y],float db[X][Y],
-             float diff1,float diff2,float s,float speed)
-{
-    // compute change in each cell
-    for(int i = 0; i < X; i++) 
-    {
-        int iprev = (i + X - 1) % X;
-        int inext = (i + 1) % X;
-        for(int j = 0; j < Y; j++) 
-        {
-            int jprev = (j + Y - 1) % Y;
-            int jnext = (j + 1) % Y;
-
-            float aval = a[i][j];
-            float bval = b[i][j];
-
-            // compute the Laplacians of a and b
-            float dda = a[i][jprev] + a[i][jnext] + a[iprev][j] + a[inext][j] - 4 * aval;
-            float ddb = b[i][jprev] + b[i][jnext] + b[iprev][j] + b[inext][j] - 4 * bval;
-
-            // compute the new rate of change of a and b
-            da[i][j] = s * (16 - aval * bval) + diff1 * dda;
-            db[i][j] = s * (aval * bval - bval - beta[i][j]) + diff2 * ddb;
-        }
-    }
-
-    // effect change
-    for(int i = 0; i < X; i++)
-    {
-        for(int j = 0; j < Y; j++) 
-        {
-            a[i][j] += (speed * da[i][j]);
-            b[i][j] += (speed * db[i][j]);
-            if (b[i][j] < 0)
-                b[i][j] = 0;
-        }
-    }
-}
diff --git a/Ready/resources/Cursors/brush-cursor.png b/resources/Cursors/brush-cursor.png
similarity index 100%
rename from Ready/resources/Cursors/brush-cursor.png
rename to resources/Cursors/brush-cursor.png
diff --git a/Ready/resources/Cursors/pencil-cursor.png b/resources/Cursors/pencil-cursor.png
similarity index 100%
rename from Ready/resources/Cursors/pencil-cursor.png
rename to resources/Cursors/pencil-cursor.png
diff --git a/Ready/resources/Cursors/picker-cursor.png b/resources/Cursors/picker-cursor.png
similarity index 100%
rename from Ready/resources/Cursors/picker-cursor.png
rename to resources/Cursors/picker-cursor.png
diff --git a/Ready/resources/Icons/22px/camera-photo.png b/resources/Icons/22px/camera-photo.png
similarity index 100%
rename from Ready/resources/Icons/22px/camera-photo.png
rename to resources/Icons/22px/camera-photo.png
diff --git a/Ready/resources/Icons/22px/color-picker.png b/resources/Icons/22px/color-picker.png
similarity index 100%
rename from Ready/resources/Icons/22px/color-picker.png
rename to resources/Icons/22px/color-picker.png
diff --git a/Ready/resources/Icons/22px/document-new.png b/resources/Icons/22px/document-new.png
similarity index 100%
rename from Ready/resources/Icons/22px/document-new.png
rename to resources/Icons/22px/document-new.png
diff --git a/Ready/resources/Icons/22px/document-open.png b/resources/Icons/22px/document-open.png
similarity index 100%
rename from Ready/resources/Icons/22px/document-open.png
rename to resources/Icons/22px/document-open.png
diff --git a/Ready/resources/Icons/22px/document-revert.png b/resources/Icons/22px/document-revert.png
similarity index 100%
rename from Ready/resources/Icons/22px/document-revert.png
rename to resources/Icons/22px/document-revert.png
diff --git a/Ready/resources/Icons/22px/document-save.png b/resources/Icons/22px/document-save.png
similarity index 100%
rename from Ready/resources/Icons/22px/document-save.png
rename to resources/Icons/22px/document-save.png
diff --git a/Ready/resources/Icons/22px/draw-brush.png b/resources/Icons/22px/draw-brush.png
similarity index 100%
rename from Ready/resources/Icons/22px/draw-brush.png
rename to resources/Icons/22px/draw-brush.png
diff --git a/Ready/resources/Icons/22px/draw-freehand.png b/resources/Icons/22px/draw-freehand.png
similarity index 100%
rename from Ready/resources/Icons/22px/draw-freehand.png
rename to resources/Icons/22px/draw-freehand.png
diff --git a/Ready/resources/Icons/22px/icon-pointer.png b/resources/Icons/22px/icon-pointer.png
similarity index 100%
rename from Ready/resources/Icons/22px/icon-pointer.png
rename to resources/Icons/22px/icon-pointer.png
diff --git a/Ready/resources/Icons/22px/list-add_gray.png b/resources/Icons/22px/list-add_gray.png
similarity index 100%
rename from Ready/resources/Icons/22px/list-add_gray.png
rename to resources/Icons/22px/list-add_gray.png
diff --git a/Ready/resources/Icons/22px/media-playback-pause_red.png b/resources/Icons/22px/media-playback-pause_red.png
similarity index 100%
rename from Ready/resources/Icons/22px/media-playback-pause_red.png
rename to resources/Icons/22px/media-playback-pause_red.png
diff --git a/Ready/resources/Icons/22px/media-playback-start_green.png b/resources/Icons/22px/media-playback-start_green.png
similarity index 100%
rename from Ready/resources/Icons/22px/media-playback-start_green.png
rename to resources/Icons/22px/media-playback-start_green.png
diff --git a/Ready/resources/Icons/22px/media-record.png b/resources/Icons/22px/media-record.png
similarity index 100%
rename from Ready/resources/Icons/22px/media-record.png
rename to resources/Icons/22px/media-record.png
diff --git a/Ready/resources/Icons/22px/media-seek-backward.png b/resources/Icons/22px/media-seek-backward.png
similarity index 100%
rename from Ready/resources/Icons/22px/media-seek-backward.png
rename to resources/Icons/22px/media-seek-backward.png
diff --git a/Ready/resources/Icons/22px/media-seek-forward.png b/resources/Icons/22px/media-seek-forward.png
similarity index 100%
rename from Ready/resources/Icons/22px/media-seek-forward.png
rename to resources/Icons/22px/media-seek-forward.png
diff --git a/Ready/resources/Icons/22px/media-skip-backward_modified.png b/resources/Icons/22px/media-skip-backward_modified.png
similarity index 100%
rename from Ready/resources/Icons/22px/media-skip-backward_modified.png
rename to resources/Icons/22px/media-skip-backward_modified.png
diff --git a/Ready/resources/Icons/22px/system-run.png b/resources/Icons/22px/system-run.png
similarity index 100%
rename from Ready/resources/Icons/22px/system-run.png
rename to resources/Icons/22px/system-run.png
diff --git a/Ready/resources/Icons/32px/camera-photo.png b/resources/Icons/32px/camera-photo.png
similarity index 100%
rename from Ready/resources/Icons/32px/camera-photo.png
rename to resources/Icons/32px/camera-photo.png
diff --git a/Ready/resources/Icons/32px/color-picker.png b/resources/Icons/32px/color-picker.png
similarity index 100%
rename from Ready/resources/Icons/32px/color-picker.png
rename to resources/Icons/32px/color-picker.png
diff --git a/Ready/resources/Icons/32px/document-new.png b/resources/Icons/32px/document-new.png
similarity index 100%
rename from Ready/resources/Icons/32px/document-new.png
rename to resources/Icons/32px/document-new.png
diff --git a/Ready/resources/Icons/32px/document-open.png b/resources/Icons/32px/document-open.png
similarity index 100%
rename from Ready/resources/Icons/32px/document-open.png
rename to resources/Icons/32px/document-open.png
diff --git a/Ready/resources/Icons/32px/document-revert.png b/resources/Icons/32px/document-revert.png
similarity index 100%
rename from Ready/resources/Icons/32px/document-revert.png
rename to resources/Icons/32px/document-revert.png
diff --git a/Ready/resources/Icons/32px/document-save.png b/resources/Icons/32px/document-save.png
similarity index 100%
rename from Ready/resources/Icons/32px/document-save.png
rename to resources/Icons/32px/document-save.png
diff --git a/Ready/resources/Icons/32px/draw-brush.png b/resources/Icons/32px/draw-brush.png
similarity index 100%
rename from Ready/resources/Icons/32px/draw-brush.png
rename to resources/Icons/32px/draw-brush.png
diff --git a/Ready/resources/Icons/32px/draw-freehand.png b/resources/Icons/32px/draw-freehand.png
similarity index 100%
rename from Ready/resources/Icons/32px/draw-freehand.png
rename to resources/Icons/32px/draw-freehand.png
diff --git a/Ready/resources/Icons/32px/icon-pointer.png b/resources/Icons/32px/icon-pointer.png
similarity index 100%
rename from Ready/resources/Icons/32px/icon-pointer.png
rename to resources/Icons/32px/icon-pointer.png
diff --git a/Ready/resources/Icons/32px/list-add_gray.png b/resources/Icons/32px/list-add_gray.png
similarity index 100%
rename from Ready/resources/Icons/32px/list-add_gray.png
rename to resources/Icons/32px/list-add_gray.png
diff --git a/Ready/resources/Icons/32px/media-playback-pause_red.png b/resources/Icons/32px/media-playback-pause_red.png
similarity index 100%
rename from Ready/resources/Icons/32px/media-playback-pause_red.png
rename to resources/Icons/32px/media-playback-pause_red.png
diff --git a/Ready/resources/Icons/32px/media-playback-start_green.png b/resources/Icons/32px/media-playback-start_green.png
similarity index 100%
rename from Ready/resources/Icons/32px/media-playback-start_green.png
rename to resources/Icons/32px/media-playback-start_green.png
diff --git a/Ready/resources/Icons/32px/media-record.png b/resources/Icons/32px/media-record.png
similarity index 100%
rename from Ready/resources/Icons/32px/media-record.png
rename to resources/Icons/32px/media-record.png
diff --git a/Ready/resources/Icons/32px/media-seek-backward.png b/resources/Icons/32px/media-seek-backward.png
similarity index 100%
rename from Ready/resources/Icons/32px/media-seek-backward.png
rename to resources/Icons/32px/media-seek-backward.png
diff --git a/Ready/resources/Icons/32px/media-seek-forward.png b/resources/Icons/32px/media-seek-forward.png
similarity index 100%
rename from Ready/resources/Icons/32px/media-seek-forward.png
rename to resources/Icons/32px/media-seek-forward.png
diff --git a/Ready/resources/Icons/32px/media-skip-backward_modified.png b/resources/Icons/32px/media-skip-backward_modified.png
similarity index 100%
rename from Ready/resources/Icons/32px/media-skip-backward_modified.png
rename to resources/Icons/32px/media-skip-backward_modified.png
diff --git a/Ready/resources/Icons/32px/system-run.png b/resources/Icons/32px/system-run.png
similarity index 100%
rename from Ready/resources/Icons/32px/system-run.png
rename to resources/Icons/32px/system-run.png
diff --git a/Ready/resources/Info.plist.in b/resources/Info.plist.in
similarity index 100%
rename from Ready/resources/Info.plist.in
rename to resources/Info.plist.in
diff --git a/Ready/resources/app.icns b/resources/app.icns
similarity index 100%
rename from Ready/resources/app.icns
rename to resources/app.icns
diff --git a/Ready/resources/appicon.ico b/resources/appicon.ico
similarity index 100%
rename from Ready/resources/appicon.ico
rename to resources/appicon.ico
diff --git a/Ready/resources/appicon.xpm b/resources/appicon.xpm
similarity index 100%
rename from Ready/resources/appicon.xpm
rename to resources/appicon.xpm
diff --git a/Ready/resources/appicon16.ico b/resources/appicon16.ico
similarity index 100%
rename from Ready/resources/appicon16.ico
rename to resources/appicon16.ico
diff --git a/Ready/resources/appicon16.xpm b/resources/appicon16.xpm
similarity index 100%
rename from Ready/resources/appicon16.xpm
rename to resources/appicon16.xpm
diff --git a/Ready/resources/appicon32.ico b/resources/appicon32.ico
similarity index 100%
rename from Ready/resources/appicon32.ico
rename to resources/appicon32.ico
diff --git a/Ready/resources/appicon48.ico b/resources/appicon48.ico
similarity index 100%
rename from Ready/resources/appicon48.ico
rename to resources/appicon48.ico
diff --git a/Ready/resources/file.icns b/resources/file.icns
similarity index 100%
rename from Ready/resources/file.icns
rename to resources/file.icns
diff --git a/Ready/resources/logo.png b/resources/logo.png
similarity index 100%
rename from Ready/resources/logo.png
rename to resources/logo.png
diff --git a/Ready/resources/ready.rc b/resources/ready.rc
similarity index 100%
rename from Ready/resources/ready.rc
rename to resources/ready.rc
diff --git a/Ready/src/Doxyfile.in b/src/Doxyfile.in
similarity index 100%
rename from Ready/src/Doxyfile.in
rename to src/Doxyfile.in
diff --git a/Ready/src/FindOpenCL.cmake b/src/FindOpenCL.cmake
similarity index 100%
rename from Ready/src/FindOpenCL.cmake
rename to src/FindOpenCL.cmake
diff --git a/Ready/src/cmd/main.cpp b/src/cmd/main.cpp
similarity index 100%
rename from Ready/src/cmd/main.cpp
rename to src/cmd/main.cpp
diff --git a/Ready/src/gui/HelpPanel.cpp b/src/gui/HelpPanel.cpp
similarity index 100%
rename from Ready/src/gui/HelpPanel.cpp
rename to src/gui/HelpPanel.cpp
diff --git a/Ready/src/gui/HelpPanel.hpp b/src/gui/HelpPanel.hpp
similarity index 100%
rename from Ready/src/gui/HelpPanel.hpp
rename to src/gui/HelpPanel.hpp
diff --git a/Ready/src/gui/IDs.hpp b/src/gui/IDs.hpp
similarity index 100%
rename from Ready/src/gui/IDs.hpp
rename to src/gui/IDs.hpp
diff --git a/Ready/src/gui/InfoPanel.cpp b/src/gui/InfoPanel.cpp
similarity index 100%
rename from Ready/src/gui/InfoPanel.cpp
rename to src/gui/InfoPanel.cpp
diff --git a/Ready/src/gui/InfoPanel.hpp b/src/gui/InfoPanel.hpp
similarity index 100%
rename from Ready/src/gui/InfoPanel.hpp
rename to src/gui/InfoPanel.hpp
diff --git a/Ready/src/gui/InteractorStylePainter.cpp b/src/gui/InteractorStylePainter.cpp
similarity index 100%
rename from Ready/src/gui/InteractorStylePainter.cpp
rename to src/gui/InteractorStylePainter.cpp
diff --git a/Ready/src/gui/InteractorStylePainter.hpp b/src/gui/InteractorStylePainter.hpp
similarity index 100%
rename from Ready/src/gui/InteractorStylePainter.hpp
rename to src/gui/InteractorStylePainter.hpp
diff --git a/Ready/src/gui/PatternsPanel.cpp b/src/gui/PatternsPanel.cpp
similarity index 100%
rename from Ready/src/gui/PatternsPanel.cpp
rename to src/gui/PatternsPanel.cpp
diff --git a/Ready/src/gui/PatternsPanel.hpp b/src/gui/PatternsPanel.hpp
similarity index 100%
rename from Ready/src/gui/PatternsPanel.hpp
rename to src/gui/PatternsPanel.hpp
diff --git a/Ready/src/gui/RecordingDialog.cpp b/src/gui/RecordingDialog.cpp
similarity index 100%
rename from Ready/src/gui/RecordingDialog.cpp
rename to src/gui/RecordingDialog.cpp
diff --git a/Ready/src/gui/RecordingDialog.hpp b/src/gui/RecordingDialog.hpp
similarity index 100%
rename from Ready/src/gui/RecordingDialog.hpp
rename to src/gui/RecordingDialog.hpp
diff --git a/Ready/src/gui/app.cpp b/src/gui/app.cpp
similarity index 100%
rename from Ready/src/gui/app.cpp
rename to src/gui/app.cpp
diff --git a/Ready/src/gui/app.hpp b/src/gui/app.hpp
similarity index 100%
rename from Ready/src/gui/app.hpp
rename to src/gui/app.hpp
diff --git a/Ready/src/gui/dialogs.cpp b/src/gui/dialogs.cpp
similarity index 100%
rename from Ready/src/gui/dialogs.cpp
rename to src/gui/dialogs.cpp
diff --git a/Ready/src/gui/dialogs.hpp b/src/gui/dialogs.hpp
similarity index 100%
rename from Ready/src/gui/dialogs.hpp
rename to src/gui/dialogs.hpp
diff --git a/Ready/src/gui/frame.cpp b/src/gui/frame.cpp
similarity index 100%
rename from Ready/src/gui/frame.cpp
rename to src/gui/frame.cpp
diff --git a/Ready/src/gui/frame.hpp b/src/gui/frame.hpp
similarity index 100%
rename from Ready/src/gui/frame.hpp
rename to src/gui/frame.hpp
diff --git a/Ready/src/gui/prefs.cpp b/src/gui/prefs.cpp
similarity index 100%
rename from Ready/src/gui/prefs.cpp
rename to src/gui/prefs.cpp
diff --git a/Ready/src/gui/prefs.hpp b/src/gui/prefs.hpp
similarity index 100%
rename from Ready/src/gui/prefs.hpp
rename to src/gui/prefs.hpp
diff --git a/Ready/src/gui/vtk_pipeline.cpp b/src/gui/vtk_pipeline.cpp
similarity index 100%
rename from Ready/src/gui/vtk_pipeline.cpp
rename to src/gui/vtk_pipeline.cpp
diff --git a/Ready/src/gui/vtk_pipeline.hpp b/src/gui/vtk_pipeline.hpp
similarity index 100%
rename from Ready/src/gui/vtk_pipeline.hpp
rename to src/gui/vtk_pipeline.hpp
diff --git a/Ready/src/gui/wxVTKRenderWindowInteractor.cxx b/src/gui/wxVTKRenderWindowInteractor.cxx
similarity index 100%
rename from Ready/src/gui/wxVTKRenderWindowInteractor.cxx
rename to src/gui/wxVTKRenderWindowInteractor.cxx
diff --git a/Ready/src/gui/wxVTKRenderWindowInteractor.h b/src/gui/wxVTKRenderWindowInteractor.h
similarity index 100%
rename from Ready/src/gui/wxVTKRenderWindowInteractor.h
rename to src/gui/wxVTKRenderWindowInteractor.h
diff --git a/Ready/src/gui/wxutils.cpp b/src/gui/wxutils.cpp
similarity index 100%
rename from Ready/src/gui/wxutils.cpp
rename to src/gui/wxutils.cpp
diff --git a/Ready/src/gui/wxutils.hpp b/src/gui/wxutils.hpp
similarity index 100%
rename from Ready/src/gui/wxutils.hpp
rename to src/gui/wxutils.hpp
diff --git a/Ready/src/readybase/AbstractRD.cpp b/src/readybase/AbstractRD.cpp
similarity index 100%
rename from Ready/src/readybase/AbstractRD.cpp
rename to src/readybase/AbstractRD.cpp
diff --git a/Ready/src/readybase/AbstractRD.hpp b/src/readybase/AbstractRD.hpp
similarity index 100%
rename from Ready/src/readybase/AbstractRD.hpp
rename to src/readybase/AbstractRD.hpp
diff --git a/Ready/src/readybase/FormulaOpenCLImageRD.cpp b/src/readybase/FormulaOpenCLImageRD.cpp
similarity index 100%
rename from Ready/src/readybase/FormulaOpenCLImageRD.cpp
rename to src/readybase/FormulaOpenCLImageRD.cpp
diff --git a/Ready/src/readybase/FormulaOpenCLImageRD.hpp b/src/readybase/FormulaOpenCLImageRD.hpp
similarity index 100%
rename from Ready/src/readybase/FormulaOpenCLImageRD.hpp
rename to src/readybase/FormulaOpenCLImageRD.hpp
diff --git a/Ready/src/readybase/FormulaOpenCLMeshRD.cpp b/src/readybase/FormulaOpenCLMeshRD.cpp
similarity index 100%
rename from Ready/src/readybase/FormulaOpenCLMeshRD.cpp
rename to src/readybase/FormulaOpenCLMeshRD.cpp
diff --git a/Ready/src/readybase/FormulaOpenCLMeshRD.hpp b/src/readybase/FormulaOpenCLMeshRD.hpp
similarity index 100%
rename from Ready/src/readybase/FormulaOpenCLMeshRD.hpp
rename to src/readybase/FormulaOpenCLMeshRD.hpp
diff --git a/Ready/src/readybase/FullKernelOpenCLImageRD.cpp b/src/readybase/FullKernelOpenCLImageRD.cpp
similarity index 100%
rename from Ready/src/readybase/FullKernelOpenCLImageRD.cpp
rename to src/readybase/FullKernelOpenCLImageRD.cpp
diff --git a/Ready/src/readybase/FullKernelOpenCLImageRD.hpp b/src/readybase/FullKernelOpenCLImageRD.hpp
similarity index 100%
rename from Ready/src/readybase/FullKernelOpenCLImageRD.hpp
rename to src/readybase/FullKernelOpenCLImageRD.hpp
diff --git a/Ready/src/readybase/FullKernelOpenCLMeshRD.cpp b/src/readybase/FullKernelOpenCLMeshRD.cpp
similarity index 100%
rename from Ready/src/readybase/FullKernelOpenCLMeshRD.cpp
rename to src/readybase/FullKernelOpenCLMeshRD.cpp
diff --git a/Ready/src/readybase/FullKernelOpenCLMeshRD.hpp b/src/readybase/FullKernelOpenCLMeshRD.hpp
similarity index 100%
rename from Ready/src/readybase/FullKernelOpenCLMeshRD.hpp
rename to src/readybase/FullKernelOpenCLMeshRD.hpp
diff --git a/Ready/src/readybase/GrayScottImageRD.cpp b/src/readybase/GrayScottImageRD.cpp
similarity index 100%
rename from Ready/src/readybase/GrayScottImageRD.cpp
rename to src/readybase/GrayScottImageRD.cpp
diff --git a/Ready/src/readybase/GrayScottImageRD.hpp b/src/readybase/GrayScottImageRD.hpp
similarity index 100%
rename from Ready/src/readybase/GrayScottImageRD.hpp
rename to src/readybase/GrayScottImageRD.hpp
diff --git a/Ready/src/readybase/GrayScottMeshRD.cpp b/src/readybase/GrayScottMeshRD.cpp
similarity index 100%
rename from Ready/src/readybase/GrayScottMeshRD.cpp
rename to src/readybase/GrayScottMeshRD.cpp
diff --git a/Ready/src/readybase/GrayScottMeshRD.hpp b/src/readybase/GrayScottMeshRD.hpp
similarity index 100%
rename from Ready/src/readybase/GrayScottMeshRD.hpp
rename to src/readybase/GrayScottMeshRD.hpp
diff --git a/Ready/src/readybase/IO_XML.cpp b/src/readybase/IO_XML.cpp
similarity index 100%
rename from Ready/src/readybase/IO_XML.cpp
rename to src/readybase/IO_XML.cpp
diff --git a/Ready/src/readybase/IO_XML.hpp b/src/readybase/IO_XML.hpp
similarity index 100%
rename from Ready/src/readybase/IO_XML.hpp
rename to src/readybase/IO_XML.hpp
diff --git a/Ready/src/readybase/ImageRD.cpp b/src/readybase/ImageRD.cpp
similarity index 100%
rename from Ready/src/readybase/ImageRD.cpp
rename to src/readybase/ImageRD.cpp
diff --git a/Ready/src/readybase/ImageRD.hpp b/src/readybase/ImageRD.hpp
similarity index 100%
rename from Ready/src/readybase/ImageRD.hpp
rename to src/readybase/ImageRD.hpp
diff --git a/Ready/src/readybase/MeshGenerators.cpp b/src/readybase/MeshGenerators.cpp
similarity index 100%
rename from Ready/src/readybase/MeshGenerators.cpp
rename to src/readybase/MeshGenerators.cpp
diff --git a/Ready/src/readybase/MeshGenerators.hpp b/src/readybase/MeshGenerators.hpp
similarity index 100%
rename from Ready/src/readybase/MeshGenerators.hpp
rename to src/readybase/MeshGenerators.hpp
diff --git a/Ready/src/readybase/MeshRD.cpp b/src/readybase/MeshRD.cpp
similarity index 100%
rename from Ready/src/readybase/MeshRD.cpp
rename to src/readybase/MeshRD.cpp
diff --git a/Ready/src/readybase/MeshRD.hpp b/src/readybase/MeshRD.hpp
similarity index 100%
rename from Ready/src/readybase/MeshRD.hpp
rename to src/readybase/MeshRD.hpp
diff --git a/Ready/src/readybase/OpenCLImageRD.cpp b/src/readybase/OpenCLImageRD.cpp
similarity index 100%
rename from Ready/src/readybase/OpenCLImageRD.cpp
rename to src/readybase/OpenCLImageRD.cpp
diff --git a/Ready/src/readybase/OpenCLImageRD.hpp b/src/readybase/OpenCLImageRD.hpp
similarity index 100%
rename from Ready/src/readybase/OpenCLImageRD.hpp
rename to src/readybase/OpenCLImageRD.hpp
diff --git a/Ready/src/readybase/OpenCLMeshRD.cpp b/src/readybase/OpenCLMeshRD.cpp
similarity index 100%
rename from Ready/src/readybase/OpenCLMeshRD.cpp
rename to src/readybase/OpenCLMeshRD.cpp
diff --git a/Ready/src/readybase/OpenCLMeshRD.hpp b/src/readybase/OpenCLMeshRD.hpp
similarity index 100%
rename from Ready/src/readybase/OpenCLMeshRD.hpp
rename to src/readybase/OpenCLMeshRD.hpp
diff --git a/Ready/src/readybase/OpenCL_Dyn_Load.c b/src/readybase/OpenCL_Dyn_Load.c
similarity index 100%
rename from Ready/src/readybase/OpenCL_Dyn_Load.c
rename to src/readybase/OpenCL_Dyn_Load.c
diff --git a/Ready/src/readybase/OpenCL_Dyn_Load.h b/src/readybase/OpenCL_Dyn_Load.h
similarity index 100%
rename from Ready/src/readybase/OpenCL_Dyn_Load.h
rename to src/readybase/OpenCL_Dyn_Load.h
diff --git a/Ready/src/readybase/OpenCL_MixIn.cpp b/src/readybase/OpenCL_MixIn.cpp
similarity index 100%
rename from Ready/src/readybase/OpenCL_MixIn.cpp
rename to src/readybase/OpenCL_MixIn.cpp
diff --git a/Ready/src/readybase/OpenCL_MixIn.hpp b/src/readybase/OpenCL_MixIn.hpp
similarity index 100%
rename from Ready/src/readybase/OpenCL_MixIn.hpp
rename to src/readybase/OpenCL_MixIn.hpp
diff --git a/Ready/src/readybase/OpenCL_utils.cpp b/src/readybase/OpenCL_utils.cpp
similarity index 100%
rename from Ready/src/readybase/OpenCL_utils.cpp
rename to src/readybase/OpenCL_utils.cpp
diff --git a/Ready/src/readybase/OpenCL_utils.hpp b/src/readybase/OpenCL_utils.hpp
similarity index 100%
rename from Ready/src/readybase/OpenCL_utils.hpp
rename to src/readybase/OpenCL_utils.hpp
diff --git a/Ready/src/readybase/Properties.cpp b/src/readybase/Properties.cpp
similarity index 100%
rename from Ready/src/readybase/Properties.cpp
rename to src/readybase/Properties.cpp
diff --git a/Ready/src/readybase/Properties.hpp b/src/readybase/Properties.hpp
similarity index 100%
rename from Ready/src/readybase/Properties.hpp
rename to src/readybase/Properties.hpp
diff --git a/Ready/src/readybase/SystemFactory.cpp b/src/readybase/SystemFactory.cpp
similarity index 100%
rename from Ready/src/readybase/SystemFactory.cpp
rename to src/readybase/SystemFactory.cpp
diff --git a/Ready/src/readybase/SystemFactory.hpp b/src/readybase/SystemFactory.hpp
similarity index 100%
rename from Ready/src/readybase/SystemFactory.hpp
rename to src/readybase/SystemFactory.hpp
diff --git a/Ready/src/readybase/overlays.cpp b/src/readybase/overlays.cpp
similarity index 100%
rename from Ready/src/readybase/overlays.cpp
rename to src/readybase/overlays.cpp
diff --git a/Ready/src/readybase/overlays.hpp b/src/readybase/overlays.hpp
similarity index 100%
rename from Ready/src/readybase/overlays.hpp
rename to src/readybase/overlays.hpp
diff --git a/Ready/src/readybase/scene_items.cpp b/src/readybase/scene_items.cpp
similarity index 100%
rename from Ready/src/readybase/scene_items.cpp
rename to src/readybase/scene_items.cpp
diff --git a/Ready/src/readybase/scene_items.hpp b/src/readybase/scene_items.hpp
similarity index 100%
rename from Ready/src/readybase/scene_items.hpp
rename to src/readybase/scene_items.hpp
diff --git a/Ready/src/readybase/utils.cpp b/src/readybase/utils.cpp
similarity index 100%
rename from Ready/src/readybase/utils.cpp
rename to src/readybase/utils.cpp
diff --git a/Ready/src/readybase/utils.hpp b/src/readybase/utils.hpp
similarity index 100%
rename from Ready/src/readybase/utils.hpp
rename to src/readybase/utils.hpp