Brilliant TimeStretching Tip from Sampleandhold(prev.onDOA)

This is where to post your questions and tips on creating presets and using cords.

Moderators: stu, Ole

Brilliant TimeStretching Tip from Sampleandhold(prev.onDOA)

Postby madmax » Mon Apr 14, 2003 10:23 pm

do this:

sample something, a vocal or something, that would be best i would think.
then of course place you sample to one key on the keyboard.
lets say c1.

(that is typically were i start my samples at for breaks, and it is now a habit)

then what you need to do is go into the mod matrix screen

(i love the instruction manual, what 500 pages and all it really says is " to turn on sampler push the power button, i will let you work the rest out)

once you are in the matrix mod you should go to the first chord. the one that has the velocity then amp then 24%

what you need to do is either disconnect that controller or set the percentage down to zero

or you could simply do what i am going to tell you to do on that patch chord section.

use the source as: VEL<
destination will be: 'SSTART
set your percentage at: -100%

now the magic starts. go to your sequencer and program a line that is a few measures long, say two, and what i want you to do is, oh say, use 32 notes, since you place the sample on c1, hit c1 over and over again, untill those two measures are full with notes. set a loop point just so you can hear it again and again

now you need to select that entire loop and set the velocity to gradually change from 127 (128) to 0 (1). so basically you should have it "fade" out. but since we disconnect the amp, you should get a steady signal.
now play the sequencer. what you should hear is the sample start at the begainning, stuttering, and slowly moving from one part to the next. so if you have a sample of "rock the beat" it will sound like:
rrrrroooooooccccccckkkkkk ttttttthhhhhhhheeeeeee bbbbbeeeeaaaaaaatttttt.

now if you increase the percentage value, what you will get is a much more delayed movement. once you hit 0 it will just stutter the start of the sample. so really you could have a word go on for an hour if you wanted.

now the bit about using the mod wheel, that is something i am sure you can use as well. try subing the VEL< source with the mod wheel and see what happens when you draw in the controller values. it should work just as well.

may have some kind of graining sound, but i think it is what you are looking for...
madmax
 
Posts: 148
Joined: Mon Apr 14, 2003 8:09 pm
Location: DC, USA

Postby sampleandhold » Tue Apr 15, 2003 5:00 am

you can also use the mod wheel for this. all you have to do is invert the percentage to positive. the velocity bit i used because that is how i found it. a complete mistake, but a good one.
Oh and another thought. you might beable to use the lfo's to do this also. set the lfo rate all the way down so it slowly modulates, i wonder if that would work?
and you can make the sample go backwards too. but it may sound silly. i actually haven't tried that yet.

just somemore thoughts on the subject.
"{jU$t-n3Rv0U$-N-+h3-@Ll3y-W@y}"
sampleandhold
 
Posts: 581
Joined: Tue Apr 15, 2003 4:38 am

Postby ezman » Tue Apr 15, 2003 6:21 pm

This is a good tip! I just added LFO1 as a modulation of sstart at +100% and an lfo rate of 18.01 in addition to velocity. The result with the fade out on velocity is that the EMU can't handle it and spits out random samples from the RAM - got some good sounds from this :lol:
ezman
 
Posts: 340
Joined: Mon Apr 14, 2003 8:14 pm
Location: Leeds, UK

Postby madmax » Wed Apr 23, 2003 3:06 pm

Last night I was working on a tune and I tried to do an old school timestretch and used this method (in addition to running the output through a 17ms delay, which gives that metallic sound). It actually didn't really sound like an OS Akai timestretch at all. I tried different sizes of the pieces of sound and using the modwheel vs velocity but it didn't sound the way I wanted it to (I only tried one vocal sample though -and not the cleanest one at that). After getting frustrated I threw a snare in there to see if it would give that timestretched snare sound and it didn't really but I got the dopest fuckin stab!!!

This technique - while I couldn't get the vocal junglist style TS that I wanted - is seriously powerful for getting some really wierd sounds. I forgot to bring a copy of the stab to work today but I'll bring it tomorrow and post it on the board. I'd imagine that if you threw in other random sounds you might get cool results as well. :thumbs:
madmax
 
Posts: 148
Joined: Mon Apr 14, 2003 8:09 pm
Location: DC, USA

Postby sampleandhold » Wed Apr 23, 2003 4:49 pm

perhaps try using a smaller amout of duration. that might work. or you could possibly get a flange on it with out the modulation. i will look into this and see what i can get, and if i get anything interesting i will send you a wave of it.

if in fact the akai timestretch algo is actually just looping the sample over and over for small amounts then this trick should be sufficeint for recreation. i personally only have the old drum and bass choons with the effect on them, and of course i have no idea if that is really an akai. perhaps it is doing something more...

can't wait for that stab man :)
"{jU$t-n3Rv0U$-N-+h3-@Ll3y-W@y}"
sampleandhold
 
Posts: 581
Joined: Tue Apr 15, 2003 4:38 am

Postby madmax » Fri Apr 25, 2003 4:35 pm

Yeah, you know I really want to try that again with a different vocal. I tried 16th notes 32 64 and 128 but I have a feeling it was the particular sample. Hopefully I can get a chance this weekend - gotta go see TS Monk tonight. I keep forgetting to bring that stab to work (I don't have my computer at home connected to the internet) but I'll try and remember to bring it on Monday. I should have a couple of tracks to play then too.
madmax
 
Posts: 148
Joined: Mon Apr 14, 2003 8:09 pm
Location: DC, USA

Postby replay » Tue Dec 02, 2003 12:40 am

The best way is to get an old s950 (I got mine a gew years back for ?60). You cant get the same effect on any other sampler even the s3000. The reason it sounded so grainy was because you were able to timestretch the sample by 400%. This meant the gaps between the samples was very pronounced. To emulate that effect you might want to patch in some gating effect as well what has already been suggested.

Would be interested to hear if this works. No need for me to try it as I have the original machine
replay
 
Posts: 5
Joined: Mon Dec 01, 2003 8:59 pm

Postby ra coon » Thu Dec 04, 2003 6:26 pm

brilliant ! brilliant ! brilliant ! , this is what im talking about , good stuff , I tried out your method . I never would have come up with that.
Keep sharing those really creative ideas. It really helps to have everybody sharing their knowledge, considering how these things are never explained anywhere else.
ra coon
 
Posts: 80
Joined: Thu Nov 27, 2003 6:40 am


Return to EOS: The Lab