Skip to content

Questions and Answers

jebofponderworthy edited this page Jul 18, 2015 · 3 revisions

What are "presets", meaning those installed under the path `/usr/local/share/yoshimi/presets`, and how can them be used?

Presets are loaded and saved via the copy/paste buffer (the blue C and P icons). To be really useful you need to create a new path. One good location is /home/{yourusername}/.config/yoshimi , but some users use folders directly in their home folder, so as to be easily manipulable in file management.

So after you have created your folder, if you develop (say) a phaser sound you especially like, you can copy it to the clipboard, and from there save it to a preset. You can then pull it back whenever you want to.

Why are .XMZ exports by Yoshimi, so much heavier than the corresponding exports by ZynAddSubFX?

We save more of the data than Zyn does. Particularly, if you've loaded a instrument into a part but then disabled it, when doing a parameter save we will still include it along with the rest. Also we save all possible 64 parts if they contain an instrument, even if they are not currently used. Any part with the default 'Simple Sound' will have nothing but the ID tag so won't take up much space.

This discussion was had a long time ago, and it was generally agreed that the increased size of an entire parameter set was relatively unimportant compared with losing a sound that you might have temporarily disabled.

The new colours look very nice, but why would I want to know which engines a patch uses?

The three engines produce sounds in very different ways and this affects their processing needs.

Working in reverse order, PadSynth is a wavetable player and takes quite a long time to generate a new set of tables, so some parameters can't be changed in real time. However, once running it places the least loading on your processor.

SubSynth generates its sound using a bank of 64 filters fed with white noise. Because the number of filters is fixed (although their settings are fully variable) the processor load is fairly predictable although rather more than that of PadSynth.

AddSynth is the 800LB gorilla! It calculates waveshapes on-the-fly, can have multiple independent 'voices' and complex filters and modulators. As well as a high running load, it performs extensive calculations every time you press a key.