Noticed a flaw in the way poly chain is handled
Noticed a flaw in the way poly chain is handled
With the way poly chain is now, if you are running a sequence or entering notes in monophonically, you only get four voices. So if you're using a patch with a long release, notes will start robbing the release of the first voice with the fifth note instead of the ninth. It should really be cycling voices between the two units with every note on, not just when there is a combined five notes or more on at the same time. Or at the very least, in poly chain mode it shouldn't consider a note as being off with the note off midi message, but rather the end of the release.
Re: Noticed a flaw in the way poly chain is handled
Good catch,
yes that's the flaw in the chaining as it is. The first module in the chain does not have any idea how many modules are chained, so when it's voices are free (as the sound is in release state) it will pick it's first free voice. What is should do is skip voices and handover the note to the next voice (and even down the chain) as long as that voice is procducing sound. Only when all voices on all chained devices are still running (in release state) it should start stealing voices, preferably the voice running the longest in release state.
As there is no feedback up the chain, this is not possible.
So it is not completely the same as a synth with 8/12/16 etc voices.
Just my two cents,
Albert
yes that's the flaw in the chaining as it is. The first module in the chain does not have any idea how many modules are chained, so when it's voices are free (as the sound is in release state) it will pick it's first free voice. What is should do is skip voices and handover the note to the next voice (and even down the chain) as long as that voice is procducing sound. Only when all voices on all chained devices are still running (in release state) it should start stealing voices, preferably the voice running the longest in release state.
As there is no feedback up the chain, this is not possible.
So it is not completely the same as a synth with 8/12/16 etc voices.
Just my two cents,
Albert