Forums > MaxMSP

question about mattijs timestretcher patch

January 19, 2009 | 9:37 pm

hello mattijs, hello list,

i?ve got a maybe simple question about the timestretcher patch:

http://www.cycling74.com/twiki/bin/view/Share/MattijsKneppers

i don?t get the calculation of the playhead as:

[1000/ sample lengnth (ms)]

feeding the result, multiplyed by the speed factor,
as the frequency to the phasor…
is this the standard formula to compute the the playpoint?

thanks for help.
yohannes


January 19, 2009 | 10:03 pm

I took only a quick look, mattijs can probably give more detail, but that equation looks like a standard one used to calculate the necessary playback speed of the phasor~. If you have a 500ms length of sound, you want a phasor~ to run from 0 to 1 within 500ms. The proper calculation for that is to divide 1000(ms per sec) by 500(ms per cycle) to get 2(cycles per second) as the proper frequency at which to run phasor~ for that length of time. I use the same exact thing, and I even notice there’s a bunch of sah~ to make sure the changes are timed sample-accurate. Pretty sweet!


January 20, 2009 | 10:41 pm

raja schrieb:
> I took only a quick look, mattijs can probably give more detail, but that equation looks like a standard one used to calculate the necessary playback speed of the phasor~. If you have a 500ms length of sound, you want a phasor~ to run from 0 to 1 within 500ms. The proper calculation for that is to divide 1000(ms per sec) by 500(ms per cycle) to get 2(cycles per second) as the proper frequency at which to run phasor~ for that length of time. I use the same exact thing, and I even notice there’s a bunch of sah~ to make sure the changes are timed sample-accurate. Pretty sweet!
>
>
thanks raja, sounds logical…


Viewing 3 posts - 1 through 3 (of 3 total)