Is the definition of a "sound" here (each one of the 16 allowed for part of the Song/Pattern Mix mode) limited to a pure single voice/patch? Or can a "sound" consist of a multilayered combination sound too?
This is a little complicated. In this case, a "sound" -- what Yamaha calls a Voice -- can indeed consist of a multilayered combination, but not so easily. A Voice consists of up to 8 elements. Most often, it is a single instrument sound, and all of the elements are related to that same instrument sounds, i.e the sound of a given instrument when played at different velocities, or with alternate attacks, etc. But there is no reason a single Voice could not have its elements used to combine the sounds of different instruments. One of the stock single Voices for example, is the sound of layered piano and strings. However this kind of Voice editing, to come up with your own combinations, is not as straight-forward as simply selecting two stock instrument Voices and splitting or layering them. So the answer is, yes, any of the 16 "sounds" can consist of a multilayered combination, but you might not want to go down that path unless you're seriously into learning the ins and outs of programming the Yamaha on a pretty deep level.
There is an "in between" way to do this. As I mentioned, the Yamaha allows you to create sets of 16 sounds you can switch among without cutoffs, but the way it works is that you can also "link" some of these sounds (it is done by assigning them to the same MIDI channel). So it is actually pretty easy to use multi-layered combinations of sound here at the expense of additional individually selectable sounds within that set. So for example, instead of defining a set of 16 individual Voices for this seamless-switching group, it could just as well be configure as 8 pairs of split/layered Voices. Or 4 pairs of split/layered Voices plus 8 additional single Voices (12 selectable patches in total). Or one set of 3 split/layered Voices, two pairs of 2 split/layered Voices, and 5 individual Voices. Any combination that totals 16 Voices will work. (Though again, in some cases, like the piano+string example I gave, one Voice can have more than one sound, but that's something of an exception unless, again, you want to get into some pretty deep programming.)
Thanks much Scott for your detailed explanation. Since I have decided that I still prefer to get the FA-08 sound palette rather than the MOXF, would you phrase your answer any differently with respect to using the FA to accomplish the seamless transitions or is it identical to the MOXF explanation you just gave me?
