Wow, this was unexpected. I spent the day coding up various ideas for a digital filter effect but most of them were failures. They either used up too much CPU or didn't work at all.
Right at the end of the day it all came together. I reckon this filter is a success and it sounds OK.
I think the filter topology is very similar to a P.I. (Proportional Integral) controller.
The output of the filter is controlled by the difference between the current input and the previous output. i.e. the error term.
I sum the error term and add it into the mix and this is what gives it the resonance. The output "hunts" around the input and overshoots. It looks like resonance, and sounds like resonance, so it probably is.
Strangely enough though, my resonance control appears to be a better cut-off parameter than the supposed cut-off parameter. Hmm ... I will have to look into that.
I should also add a logarithmic curve on the resonance CV so it has better resolution at the low end.
this is so fun to watch
makes me want an oscilloscope
wakka737 1 month ago