Yes, this is currently my most-desired MOD Source.
Don Solaris wrote:A few random number generators. i.e. RNG1, RNG2, RNG3, RNG4, RNG5. Each time you stuck a key, new set of numbers is generated.
It could either be multiple Random Number Generators
(like you have above) or, I'd be fine with a single RAND-VALUE
as long as it delivers a unique value for each and every instance that it is read.
So... If I were to assign RAND-VALUE
as the Source of a MOD, and VCA PAN as the Destination, I'd get a fresh pan for every note played (even if the notes were played at the same time).
This would also be very useful for slight variations of LFO RATEs, FILTER CUTOFF, RESONANCE, etc.
I'm currently using a SLOW LFO 4 set to S/H, with NOTE RESET On. But this is not ideal as it's a poor use of a precious LFO.
I logged onto this forum to request this very thing. I would add that this random mod source should be distinguished from Sample and Hold or Noise as a modulation source. Those mod sources, like LFOs and ENVs, exist in the time domain. That is, they keep on acting over time instead of being a one-shot instance each time.
For example, if you use something like S&H for panning, when you sustain a chord, then you will get moving parts across the stereo field, like a warbling tremolo effect. This is desirable for some circumstances, but not all.
On the other hand, let's call this new modulation source "Spread" (as suggested by Lady Gaia on page 93 post 2765 on the Gearslutz Moog One thread). When applied to pan, "Spread" will randomly place a voice within the stereo field, say 75% left, but then that voice will stay at that position in the stereo field until the next time the voice is triggered. The next voice will be placed at 52% right, the next voice at 67% right, and so on.
At 0% spread, each new triggered modulation is 0.
At 10% spread, each new triggered modulation is between -10 to 10.
At 50% spread, each new triggered modulation is between -50 to 50. And so on.
(I think the coefficient is called a "seed" in programming terms).
As suggested a couple posts up, this Spread could also be applied to each new voice filter cutoff, giving a slight variance with each new trigger. It could also be applied to an LFO's rate, giving slight changes in rate each time a new LFO is triggered.