pwl : was RE: [sdiy] Taking a Step towards - - --((FUTURE-PREDICTIONS))-- - -

Czech Martin Martin.Czech at micronas.com
Fri Jan 9 12:30:24 CET 2004


btw.:

during vacation I experimented with piecewise linear control
curves PWL. Like the ones you find in SPICE, or in some digital synths.

The result was that only a few segments were needed to build
a pseudo exponential curve for frequency or gain control.
E.g. with 8 interpolation points the lack of smothness which PWL
has was not audible any more. For very short events (1s and shorter)
even less segments did the trick. Only long fades needed more segments,
i.e. a certain rate of interpolation fixed points per second
must be kept.

Perhaps interesting, because basically this enables audio rate
controll curves out of only a few bytes of information
with modest computational load.


m.c.


> -----Original Message-----
> From: owner-synth-diy at dropmix.xs4all.nl
> [mailto:owner-synth-diy at dropmix.xs4all.nl]On Behalf Of Paul Maddox
> Sent: Freitag, 9. Januar 2004 12:16
> To: Synth DIY
> Subject: Re: [sdiy] Taking a Step towards - - 
> --((FUTURE-PREDICTIONS))--
> - -
> 
> 
> Colin,
> 
> > Since the output would ultimately be an analogue voltage, I 
> think it would
> > be entirely reasonable to call it a 'numerically controlled 
> multi-stage
> > analogue transient generator', and see if anyone notices. ;-)
> 
> NCMSAT, nice acronym :-)
> good point though...
> Dunno myabe after I've finished the PolyDAC (tony found a slight bug).
> Though I have to say, my enthusiasm for things analogue is 
> waining quite
> some.
> 
> Paul
> 
> 



More information about the Synth-diy mailing list