My first idea was to break an incoming audio signal up into dynamic ranges, for each cycle that is. So different sections of the same cycle can be routed differently :)
I have really only tried this with bit and sample rate modification.
It does work well though !! Interesting results with sample rate, a lower rate for the lower thresholds, moving up to a higher rate for the higher..
The next, was to replicate a complex wave with a sine wave :)
The idea was to use zero crossings as points to determine the freq of the sine wave. This itself has been not particularly succesful - this may because Plogue does not do good maths :(
the idea is, when the wave crosses zero, the accum will split out the length of samples since last crossing - this gets converted into the freq of the lower oscillator.
The delay line is to make sure the freq's exist for appropriate lengths - this bit is wrong - needs more work !!
The next part of this one, is to get some sort of average amplitude for individual cycles.