[sdiy] Delayed LFO fade-in time

[Richie Burnett]

> Richie, first, the assertion that LFO delay should only work for pitch
> modulation is purely a matter of taste and/or intention.

Fair enough.

> Also, the way I know how to introduce such delay is to run
> the LFO through a VCA before the destination, and trigger the VCA with
> a slow EG, so the EG attack rate is responsible for the LFO delay
> time.  If you want that to be variable you need to use a VC-EG.

Doesn't the attack rate need to be variable.  If you set the LFO delay to 
zero, then you want it to kick in immediately at the onset of each note, not 
ramp up with any attack time.  But if you set the LFO delay to 1 second, you 
certainly don't want the LFO modulation to kick in instantly at t=1s because 
there will be a discontinuity, so it needs to fade in.

> As
> for polyphonic implementation, checking with my juno MKS50, the delay
> is global, meaning once the LFO has started, if you play a new note it
> will already be affected by the LFO.

Interesting.  I'll fire up my Alpha Juno and examine what the LFO mod does 
on that with different delay settings.

> This leads me to think it is a
> single dedicated VCA/EG combo under the hood running the LFO delay.

Yes.  Or maybe that the software algorithm generating the LFO oscillation is 
actually "stopped" until the delay time expires, then it starts up from 
phase=0.  I'll examine what my alpha does.

Thanks for the tips.

-Richie,