I've been using the virus for 2 years now and I though I'd share some basic tips for getting the most voices out of your synth. Most of my tracks evolve around the Ti and many are made almost exclusively using the Ti. However, I always keep it all live and realtime, as this is the way I like to work. Hence I learned, out of necessity, how to get a lot of channels out of it.
1. The mono button - sounds simple enough, just use the mono whenever you don't need the polyphonic note overlaps. It saves voices be default. Sometimes it can even pay of to use a long reverb tail to "fake" a long release time, instead of using long release which would eat up a lot of voices.
2. Use Hypersaw instead of unison - even though hypersaw is not a true replacement for unison mode, you can usually get almost identical results by using hypersaw + 2x unison (+ spread) instead of using something like 6x unison. This saves a lot of voices. Also, if all you want is the stereo spread, try using filters in split mode; this gives you the ability to use spread knob to pan the oscillators without using unison mode.
3. Keep the amp release as low as possible - this one is just common sense; don't open it up all the way unless absolutely necessary. Most of the time keeping it at 12 - 3 o'clock is good even for the long pads, and this also saves more voices. Again, a long reverb tail can replace long note release in certain situations.
4. Make sure to turn off everything you don't use - essential. Most parts of the Ti can be disabled when not in use. Some have dedicated switches, but others are just off when they are at their lowest value. For example:
- if you don't use the mod matrix, make sure to turn it off (select none in the source)
- if you don't use delay/reverb, make sure they are at their lowest value (and/or switched off)
Even turning a knob 1% can turn on an effect/oscilator/filter which will use DSP power for nothing.
5. Export/Bounce the FX/oneshots first - even though I rarely find it necessary, sometimes you just have to export something to keep the Ti from stealing notes. So why not start with the easiest parts first? Keep all the leads, basses, sequences, etc.. and just bounce the oneshot parts. The are quick to bounce and can easily be manipulated within a host anyway.
6. Use both DSP Chips - there are 2 chips in the Ti and they each get one synth part; 1st chip gets part 1. 2nd gets part 2, then the 1st gets part 3, and 2nd chip gets part 4....etc. What does this mean? Simple: If you use a thick unison lead and pads on channels 1, 3 and 5 you might run out of voices, even though the 2nd chip still has DSP power to spare. So before bouncing always try to switch one part to the other chip. It can often give you the extra voices you need.
7. Keep in mind what eats the most DSP power - Some things on the Ti use almost no DSP power, while others eat DSP cycles like crazy. For example; Grain and formant complex oscillators are very DSP intense, while classic and wavetable oscillators are not. The analog filter uses more DSP than the classic one... etc.
Also note that the suboscilator is always on, so make sure you use it whenever you need another square/sine oscillator, instead of turning on an extra oscillator source.
8. If you can EQ/Delay/Reverb through you DAW's mixer channel, do it - this can save you some precious DSP cycles, not to mention you can use more advanced effect(s). Just remember to disable the Ti's internal effects.
9. Don't use presets....much - make your own sounds. Many of the presets are amazing, but many of them are also made to be played in monotimbral mode. So they can easily choke the Ti very quickly. And quite a few of them can be remade to use much less DSP for a 95% similar sound.
10. Experiment - the more you learn about the Ti, it's architecture, routing, possibilities and features, the easier it will be to make your own sounds, and make them as optimized as possible.
Hope this helps.
All the best,