[sdiy] poor man's alternative to a CS-80?

[Antti Huovilainen]

On Thu, 20 Oct 2005, ChristianH wrote:

> It sounds as if it's something you do over the weekend and then have a
> running CS80 clone on monday. Even assuming some is familiar with signal

I did that with Moog Taurus (in a weekend), and that's a 1:1 clone going 
as far as to model some of the actual circuits (code actually deals 
internally with voltages, amperes, ohms & farads).

> processing - I think it's a huge difference between any basic soft synth
> that makes some noise and something Danjel is looking for.

Yes. However, making a very good sounding softsynth is not that hard. If 
you do not need exact clone of CS-80 (meaning control at 70% on your synth 
and real CS-80 produces exactly same result), it is not really 
that complicated.

A short analysis of CS-80:
You have basically 16 lfo -> vco -> vcf -> vca voices which are 
always layered in twos.

The VCO is normal saw/pulse vco except there is a short blip in the saw. 
This can be done digitally simply by layering bandlimited sawtooth and 
bandlimited pulsewave with the pulsewidth depending on frequency such that 
pulse length is always the same in time.

Jurgen Haible did a pretty nice analysis of the VCF. By comparing it and 
normal SVF, the modifications are rather simple to do to a digital SVF.

All the modulations (aftertouch etc) are trivial to do digitally.

Relating to the chorus, I just published a paper on emulating analog delay 
effects: http://dafx05.ssr.upm.es/Proc_DAFx05/P_155.pdf

Sure, it is easier to just buy CS-60, but this IS synth-diy list afterall 
:). Also I bet making an analog CS-80 clone would be a lot more work.

Antti

"No boom today. Boom tomorrow. There's always a boom tomorrow"
   -- Lt. Cmdr. Ivanova