[sdiy] 5-scroll chaos circuit

[Ian Fritz]

At 12:25 PM 2/15/2008, David Moylan wrote:
>Ian, can you comment on how the x/y plot patterns translate into patterns 
>heard on a single output in the time or frequency domain?  The pictures 
>always look cool but I'm not sure how to interpret them.

Dave --

Good question.  We have to be careful to avoid merely making pretty scope art!

There are two basic kinds of applications -- audio signals and low 
frequency control signals.

If you think in terms of continuous control signals, then each range of 
signal voltages corresponds to some range of VCO pitches, filter cutoff 
frequencies, amplitudes, etc.  These signals evolve in a correlated but 
irregular manner, always staying within the bounds of the chaotic attractor 
(or scope patterns).  The resulting impression is generally a sound that is 
"always the same, but always changing".

But is also useful -- perhaps more useful -- to use the chaotic signals to 
define timing events.  For example, whenever one signal crosses a threshold 
a S/H could catch the value of one of the others, which in turn is used to 
control some synth parameter.  When you do this it is often just about 
impossible to correlate the scope traces with what you hear, but the 
impression of irregularity within an overall pattern is still clear.

I hope that the example sound clips I have posted might be helpful in 
conveying some of the many possibilities.  But this is still a fairly new 
area, so everyone needs to experiment on their own.

I haven't found chaotic audio signals to be that interesting, but some 
people seem to.  There are some interesting videos on utube that you could 
probably turn up with a quick search.

Thanks for your interest.

   Ian