[sdiy] Modulation waveforms

[Scott Nordlund]

> Great paper, I like the idea of the random slopes. I don't think you
> could get that behavior with a slew limiter after the LFO.
>
> Another waveform that I like to play with is the random walk (a.k.a
> brownian motion); you randomly change directions instead of randomly
> change values.

Random walks are unbounded, so this presents a little bit of a problem.
It should reverse direction at the boundaries rather than overflow or
saturate.  More interesting (not to say useful) would be to put it
through a sine (or other) waveshaper.

I've been working on reverb algorithms recently, and one of the 
important ingredients here is random modulation.  I wanted it to be
"semi periodic", so I delay each clock pulse by a random time
(adjustable as a fraction of the clock period) and then linearly
interpolate between random levels.  It breaks up the monotony
without letting it get too crazy.

If you just, say, pick a random clock frequency at each step, it 
tends to result in short bursts of many clock pulses with not much
in between.  Not necessarily useless but not what I wanted.  With 
the delay I can maintain an average period and also limit the
maximum slope of the output waveform (important for delay
modulation).

I think random/chaotic behavior is very important to self-playing 
types of synth patches, and adjusting them by ear isn't easy.  It's
unintuitive and takes a long time to get a feel for the result, 
but it plays a big role in the musicality of the output.  So it's 
probably worthwhile to develop some interesting modulation schemes
in advance, rather than just ad hoc patching.
 		 	   		  
_________________________________________________________________
The New Busy is not the old busy. Search, chat and e-mail from your inbox.
http://www.windowslive.com/campaign/thenewbusy?ocid=PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_3