granular time-stretch pitch-shift
Dear all, i've searched archives and googled the net for days for that..there are patches around (even the granular example in the max folder) but none fully satisfying for me since i need clean playback not granular effects right now..my attempts:
1- i've built a patch upon sakonda grain: according to my tastes to avoid metallic and comb effects, 2 overlapping grains are enough (i know i could use up to 8 or + grains , each with a proper phase,but i prefer some "flam" artifacts rather than metallic) and i divide the sample length with the desired number of slices (i.e 4,8,16,32 etc) to get grain length
problems:
-it seems to me that grains sometimes do not keep the right phase distance so i get that undesired chorus, phaser effect
-the first grain/slice of the loop is always cancelled because the play position it's locked with sah 0. to the windowing phasor with 0.5 phase offset..
2- i've built a patch based on wolek's granular toolkit objects grain.phase and phasor.shift, using only 2 overlapping grains
the result sounds aesthetically pleasant to me
problems:
-i would like to do all without externals if possible at this point, then i could use externals eventually for efficiency
actually this is becoming an hot issue for me!
any help would be greatly appreciated
best
michele
mic wrote:
> problems: -it seems to me that grains sometimes do not keep the right
> phase distance so i get that undesired chorus, phaser effect -the
> first grain/slice of the loop is always cancelled because the play
> position it's locked with sah 0. to the windowing phasor with 0.5
> phase offset..
You could try to differ the length and starting point of the grains by
analysing zero crossings and then syncronize the next grain to the old
one, a bit like a dj is syncing the beats of two discs... That would
also get rid of the clear pitch determined by the grainsize...
And if you're insane, you could adjust the grainsize to subharmonics of
a detected pitch of your signal...
Just ideas...
Stefan
--
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com
try this:
Hi, thanks for the answer
well i know very well the patch you posted but as i said i don't like the artifact it produces (a sort of phase-comb effetc) when the pitch is changed, i mean sakonda patch is great but does not suit my aesthetic needs here..i prefer using less grains for example two..what i want now is a loop player that implements some granular technique for time stretching without affecting pitch and change pitch without changing tempo..
reaktor has an excellent one (it's an object that does it, how they call it a macro?..)
it would be great if msp had a groove~ like object that implement it internally..i know, the granular toolkit has grain.groove but it's too heavy on cpu..
i think radial granular-engine does it well..
is it entirely max/msp written rigth?
here is a simplified grain patch
> You could try to differ the length and starting point of the grains by
> analysing zero crossings and then syncronize the next grain to the old
> one,
ah good one
> And if you're insane, you could adjust the grainsize to subharmonics of
> a detected pitch of your signal...
this sounds like science-fiction to me!!
thanks for the reply
i hope the gurus out there could help us..
;)
and here is a super simple but effective example patch using a single grain..or windowed chops as you like ;)
Max won't open it.
On 2/8/07 3:18 PM, "mic" wrote:
> and here is a super simple but effective example patch using a single
> grain..or windowed chops as you like ;)
>
Cheers
Gary Lee Nelson
Oberlin College
www.timara.oberlin.edu/GaryLeeNelson
Quote: Gary Lee Nelson wrote on Thu, 08 February 2007 21:41
----------------------------------------------------
> Max won't open it.
>
oh..here it is, when pasting it you have to inizialize loadbang inside windowfade patcher
On Feb 8, 2007, at 3:55 PM, mic wrote:
> -a sequence re-order featute (i.e like in modsquad) and loop
> section selection since it's difficult to keep in sync the slices
> sequencer with the windowing pahsor
Based on the patch you attached, it looks like you just need to
restructure that [PosCalc] subpatch. The speed parameter is driving
that [phasor~] inside, which will always produce a linear run through
the sound. I would suggest creating a [buffer~] that contains the
pattern that you would like to implement for resuffling. Then use
the [phasor~] to drive a [cycle~] that is reading that buffer. I
believe that is how modsquad does it.
As far as [grain.phase~] goes, you don't have to worry about skips.
The external will always finish a grain before it starts using new
input variables. The only time this gets troublesome is when you
have rather long grains. But then that isn't granular processing, is
it?
Hope that helps.
------
Nathan Wolek, PhD
nathan@lowkeydigitalstudio.com
http://www.lowkeydigitalstudio.com
On 9 Feb 2007, at 17:03, Mattijs Kneppers wrote:
> Ableton Live uses this:
> http://www.zplane.de/showPage.php?SPRACHE=UK&PAGE=products11
Interesting - I didn't realise they'd outsourced it, but it makes sense.
-- N.
here we are, here is a simple transposition patch inspired by the transposer abstraction, i like how it sounds..
problem:
-looped once the first slice is cancelled, i can understand why but i cannot see a solution..outside it's sunny maybe i need a walk! :)
thanks to everyone for the patience
On Feb 10, 2007, at 12:17 PM, Mattijs Kneppers wrote:
> Nonetheless the zplane stuff is realtime. Hmm.. where did I put my
> Curtis Roads - Microsound?
For info on pitch synchronous granular synthesis, I would recommend
this source:
DePoli, G. & Piccialli, A. (1991). Pitch-synchronous granular
synthesis. In G.
DePoli, A. Piccialli, & C. Roads (Eds.), Representations of musical
signals
(pp. 187-219). Cambridge, MA: MIT Press.
-------------------
Nathan Wolek, PhD --- nwolek@stetson.edu
Assistant Professor of Music Technology
Stetson University - DeLand, FL
http://www.nathanwolek.com
Thanks!
Quote: nathan wolek wrote on Mon, 12 February 2007 02:57
----------------------------------------------------
> On Feb 10, 2007, at 12:17 PM, Mattijs Kneppers wrote:
> > Nonetheless the zplane stuff is realtime. Hmm.. where did I put my
> > Curtis Roads - Microsound?
>
> For info on pitch synchronous granular synthesis, I would recommend
> this source:
>
> DePoli, G. & Piccialli, A. (1991). Pitch-synchronous granular
> synthesis. In G.
> DePoli, A. Piccialli, & C. Roads (Eds.), Representations of musical
> signals
> (pp. 187-219). Cambridge, MA: MIT Press.
>
> -------------------
> Nathan Wolek, PhD --- nwolek@stetson.edu
> Assistant Professor of Music Technology
> Stetson University - DeLand, FL
> http://www.nathanwolek.com
>
>
>
----------------------------------------------------
On Feb 10, 2007, at 10:58 AM, Paul Hertz wrote:
> Six of one versus half a dozen of the other, or is one solution
> possibly more versatile than another?
Hello, Paul! Part of the reason for the different externals is so
that the user can choose based on the application. It really depends
on the situation.
-------------------
Nathan Wolek, PhD --- nwolek@stetson.edu
Assistant Professor of Music Technology
Stetson University - DeLand, FL
http://www.nathanwolek.com