[music-dsp] Modulation matrix implementation trickery?

Russell Borogove kaleja at estarcion.com
Mon Jun 7 01:55:52 EDT 2004


> Sorry,  I'm all out of grand unified theories at the moment, everything
> around here seems to be be ad-hoc pragmatism these days ;-)

I'll take what I can get.

> For control-rate updates some kind of uniform polymorphic source/sink
> interface with adaptors for the various mapping cases you mention could
work
> quite well. Chain adapters as needed. If you normalize all of the
parameter
> ranges the adapters can probably just be function pointers, or function
> pointers with one word of state.

It's VST, so the params are normalized (from one angle, anyway).

> Do you need audio-rate modulation? an alternative is to use audio-rate
> linear (or some other) interpolation to avoid the zipper noise, but still
do
> the modulation updates at a slower rate.. that way your modulation system
> will be simpler.

So the modulation system itself would run at k-rate but some of
the destinations would do a-rate interpolation independent of
that? Hm, that concept is actually enough for me to want to split
the "sink" abstraction out from where I had it, which was an
extension to the wrapper I use with VST params.

Hmm, that might be the brain-shake I need to get this done.




More information about the music-dsp mailing list