yeah, that is the idea. have a look at Thomas Musil's patch that i sent, as he found the most elegant way of dealing with it: two variable delays that [as far as i understand] crossfade against each other while they lengthen or shorten, so that when the buffer of one runs out, it has already crossfaded to the other where it picks up again. smaller vd~ sizes are good for higher input frequencies, but lower input frequencies will need a bigger buffer to keep from "granulating" them.
at least this is how it was explained to me :-)
d.
Pixelcorrection@wmconnect.com wrote:
So would you adjust the pitch using vd~ by raising and lowering the delay time by putting an input signal into the delaywrite~ buffer?
Hallo, derek holzer hat gesagt: // derek holzer wrote:
yeah, that is the idea. have a look at Thomas Musil's patch that i sent, as he found the most elegant way of dealing with it: two variable delays that [as far as i understand] crossfade against each other while they lengthen or shorten, so that when the buffer of one runs out, it has already crossfaded to the other where it picks up again. smaller vd~ sizes are good for higher input frequencies, but lower input frequencies will need a bigger buffer to keep from "granulating" them.
at least this is how it was explained to me :-)
It's correct in the big picture. It is also included in the pd distribution as doc/3.audio.examples/56.delay.pitchshift.pd In the books it is called a "rotating tape head" delay.
In modern commercial apps often granular snythesis is used to change the tempo and pitch of loops independently. The syncgrain~ external works great for file loops, but not for incoming signals. I guess, munger~ might be the way do do it here.
ciao
Frank Barknecht _ ______footils.org__
-
derek holzer -
Frank Barknecht