Re: Pitch-shifting percussion sounds & Latency rant

[SoundFNR@aol.com]

> There should be no delay of the pitch-shifted signal (very important 
>since 
I 
> want to use it for rhythms) 

Unfortunately delay is always present to some extent when pitch shifting.
Pitch shifting works by dividing the sound into chunks, and playing back 
those 
chunks at a different speed.
There's no way you could ever pitch shift up without a delay (imagine, 
you start playback the instant the sound comes into the device, 
...obviously you can't then replay at a higher speed, as you'd
run out of sound straight away)

Its theoretically possible to shift down with an inperceptable
delay, but there would have to be a way of recognising
incoming transients (=the start of the note) 
and using that to retrigger playback. 
( and this would only really work for percussion & plucked sounds)


As far as I know, no existing device works on that principle though.
(however, as the Repeater pitch shift works on recorded material,
rather than input, maybe it can compensate for the delay) 


As to latency in general, I note that it quite often goes
unnoticed, as people often claim that such and such a
device has "no noticeable latency".
Somehow the brain seems to compensate for 
the delay sometimes, I've even noticed that when changing
from a MIDI controller with significant delay, to one without, 
the notes felt like they were early.


andy butler