Historically Unison mode has been monophonic only, and a way to stack many voices of a polyphonic synth in order to get a very fat sound. As far back as the Prophet 5.
It's only with more recent synths that a dynamic voice allocation Unison mode appeared on some models from other companies.
But, as stated by DSI before, the company has never had the philosophy of following/copying what other companies do.
So the Unison mode from DSI is a static one, and there are no plans to add a dynamic voice allocation mode to it for a Poly-Unison on the REV2.
I really don’t lose any sleep over it, but I must say I don’t find much weight in the historical argument for not having a more dynamic voice allocation on these synths.
The fact is that electronics have come a very very far way since the ‘historical’ synths and playing polyphonic unison notes on one of these modern prophets would be absolutely delightful.
Like I said originally, it didn’t stop me buying another of their synths but I think very clearly a massive shame that this option gets ignored by them.
As for the comment directly above mine here (assuming no-one else has posted while writing this), that’s not really the same though. A partial workaround, one that I use often in fact, is to use stacked mode with the same program in each layer, but that only achieves max 2 voice unison polyphonic playing. Still better than nothing of course.