Dear Karim,
How can I build a patch that subdivides a rhythmic voice into a single duration unit if the unit is over/under (x) seconds long?
Hi,
Here’s a blueprint: compute the number of onsets (N), and the greatest common divisor between consecutive pairs of onsets (GCD); compute the integer n for which 2^n (nth power of 2) best approximates the GCD — the integer part of log2[GCD]. Your desired voice can be given by inputing the list of midicents and a time signature of (N 2^n.)
All the best,
António
Dear Hamish,
An example , a patch even a sketch will be helpful.
Best
k