A disaster! The following csd gives discontinuities every k-cycle on both channels Looks like time for significant investigation ==John ff Quoting jpff@cs.bath.ac.uk: > Well you could try the butlp with a-rate cutoff frequency to see if > it still happens. That code is in git now. > ==John ff -d 0dbfs=1 nchnls = 2 ksmps = 1000 instr 1 koct lfo 3,4,0 aoct oscil 3,4 asig vco2 0.5,400,4,0.5 asigk butlp asig,cpsoct(7+koct) asiga butterlp asig,cpsoct(7+aoct) outs asigk,asiga endin i1 0 10 e