disable fast reciprocal division in SIMD because it seems to make the…

… final results worse

shared/libebm/unzoned/unzoned.h

@@ -182,7 +182,11 @@ INLINE_ALWAYS static void StopClangAnalysis(void) EBM_NOEXCEPT ANALYZER_NORETURN
          StopClangAnalysis();                                                                                          \
    } while((void)0, 0)
 
-#define FAST_DIVISION
+// Disable this for now. Using fast reciprocals seems to change the results quite a bit
+// and make them worse. Also, the reciprocal algorithm is different between Intel and AMD
+// and it makes comparisons difficult, and I don't think this division is critical to
+// making it faster
+//#define FAST_DIVISION
 
 // 16 byte alignment works for *most* SIMD implementation, but it's even better to align with the 64 byte cache!
 #define SIMD_BYTE_ALIGNMENT STATIC_CAST(size_t, 64)