Stefan Turner wrote:
amplitude squared? And do you know if it is what mixer faders, for example, (both software and hardware) would generally use? I am interested to know what the 'right' way is.
afaik, real faders normally have quite "non-linear" behaviour. a very non-linear approach (which tries to mimic this behaviour) is also supplied by the iemlib, called [fadtodb] [fadtorms] and the like.
mfg.a.dr IOhannes