This is cool... if you have the CPU power to run 100s of voices
I find that a lot of the time when I'm doing unisons it actually can pay off a lot more to create a mixture or layers of different sounds and timbres. Unison is really a sort of blunt instrument... everything tends to end up sounding like another variation on a "super-saw".
When I'm working with my analog instruments I tend to manually make adjustments to the settings while I'm layering... each take is like a single "voice".
The little adjustments made between each component make up a very significant contribution. For example rather than eight identical narrow pulse waves detuned and panned all over the place you might try combining a ramp "wow" sound, a square "dub", a thick pulse (slightly off square) with some detune, a thin pulse and a very thin pulse with various pan settings.
The result is rather than having a thick pea-soup like "super-saw" sound, you get a whole field of sound like you'd expect from many different instruments. Each with its own variation on velocity, timing and so on.
I find 100% of the time when I get a sound I'm proud of I've gone this route and not used unison.
Certain sounds are indeed "unison sounds". Thick "ahhs", pads and of course the well known "super-saw". Other sounds are not.
