Re-sampling: how *not* to sample & pitch-up (doubling original speed) continuously, because it's impossible ... duh, me. (Though with quantum computing ... :s)

Elektron Gear Octatrack
#1

Hello!

I’m currently using the Octatrack as part of my set-up as a percussionist. Essentially what I’m trying to do is re-sample live a rhythmic pattern with a continuous, regular pulse, but pitch-shifted up (an octave, 12 steps) so it is playing back at twice the speed. The problem I’m having is that when the sample plays back (whether on the same ‘flex’ track OR on another track listening to the flex recorder-buffer of the recording track) there is always a gap, a lapse in the continuity of the pitch-shifted sound. I’ve tried everything I can think of, to do with the recording & play-back trigger lengths/placement. (‘Pickup’ machines don’t work because the pitch shift leaves the material at the same tempo).

I’m at a point where I’d like to record with this method, but finding it impossible to figure this one out. The best solution is to cover the gap by repeating the re-sampled material to give continuity - that has been fine to work with this approach, and see what it sounds like… but it’s too much of a compromise.

Let me know if I can clarify anything; if I haven’t explained clearly enough, I can say exactly what I’m doing in the tracks, or share audio. Any suggestions would be greatly appreciated.

https://soundcloud.com/hemidemisemi/double-click

Thanks!

#2

I like to record and double the speed too. Maybe I miss something but the gap is normal if you don’t put enough trigs. Looping or looping pipo can do the job. Changing pattern with different settings, trigs to play the recording can help too.

Can you clarify a few things :
What is the length of your record?
How do you start Recording?
How do you listen your playing while recording?
Can we here the normal speed recording befire you double speed?
A precise exemple would be welcome.

Edit : I would rather say Pitch because Pitch Shift doesn’t change the tempo. Of course time stretch has to be off.
Are you sure you keep the same tempo (x2)? Recordings are looping correctly?
Start and Length are ok?

#3

Thank you for replying!

So here’s roughly what I’m doing at present:
I take a ‘flex’ track with 16 beats, and place both recorder and playback triggers on the first beat. (I’m playing at around 90BPM and doubling it up to 180BPM. I was working with individual track lengths - I tried without, but to no avail.). When I pitch it up 12 steps, I hear the doubled-up sample play for the first 8 beats (with silence on the last 8 beats). I want to fill that silence with the next part of what I’ve played live … Obviously if I don’t pitch up the sample I can just put the recording and playback triggers on the same beat, and I’ll hear a delay of a full ‘bar’. I’ve tried off-setting the triggers, and having different flex tracks ‘listen’ to the recorder buffer of another flex track, of differing lengths! I listen back in headphones, from the headphone jack, because I want to hear the doubled-up drumming, and having a monitor speaker feeds back into the microphones. The normal speed recording plays back in full, and continuously - imagine a break-beat style rhythm, half the speed of a drum’n’bass or jungle track …
I can record my playing if it’d help (may take me a while to upload).

One way to try to do it is to take a flex second track, but put the recorder and/or playback triggers on after the first 8 beats. BUT, for some reason, when I do that, I’m getting a delay-like effect of about 2 beats. I think it might actually be paradoxical - the time, proportions, etc. No matter how I shift everything, I can’t get a continuous rendering of my playing feeding back, in doubled in speed, with no weird repetitions or gaps! Infuriating.

What do you mean by “looping pipo”?

#4

from what I can gather … it’s late here … might be missing something , there’s an issue if you want to see doubling up as a compromise

I don’t see how you can accelerate time on a playback head at the same time the recording is happening … you always have to think of a moving window of sampling time … the playback time has to be at least delayed half that time such that the termination of play and record is concurrent, then you repeat, but what do you play for the next time window, you play the last window, so playback needs to be delayed half the time window and you have to play the buffers twice centred around the end of the buffer window

that’s just my thoughts from what I can gather you’ve described

1 Like
#5

Loop Pipo is a ping-pong loop setting, alternating loop with a reverse playing. I like doubling tempo between 75 and 100 bpm. Sorry but I don’t understand exactly what are your problems.

You record input AB?
You want to record 32 beats and play it x2 on 16 beats?
You can do it but anyway it will be played after recording.
I prefer to talk about steps instead of beats. You can record 64 steps max, so 16 beats. If you want to record longer, you have to use a 1/2x scale.

#6

You’ve grasped what I was struggling to describe, avantronica, yes! Thanks for making the effort … and apologies for any obscurity in my description. It’s not easy, between the doing and the describing.

The way I’ve been trying to get around this recording/playback concurrence is to distribute recording/playback over different tracks. (I refrained from describing this in the outset because I thought no-one would reply if I did!). I wouldn’t mind if the doubled-up sound were from what I had previously played: all I want is for it to be a continuous stream of sound, exactly what I played, but doubled-up (pitched-up an octave). But, from what you’ve said, which I think is what I’ve been experiencing, that is actually impossible? Even if one had to use all of the octa-tracks solely for this purpose?

I had been using 4 flex tracks. 1 + 2 recording, say (16 sequencer steps each, one after the other). 3 + 4 are playing back the recorder-buffers of 1 + 2, respectively (8 sequencer steps - of the same length as the recorder steps - one after the other). Each square would be 2 sequencer steps, and I underlined where the triggers would be placed. (I can upload audio if this schema doesn’t help clarify anything … it sounds like I’m just re-iterating exactly what you outlined).

But yes, because the playback ‘cycle’ is shorter, you necessarily hear previously recorded material twice. Shit. If there’s anything you can think of or suggest that might work around this - even if utilising the whole OT capacity for it - I’ll give it a try! By ‘compromise’ I just meant in terms of the sound. I could easily record my playing and modify the sample subsequently, and merge it with other material. But the live playing aspect would be lost, thereby, as it is when I hear material repeated. (It sounds a lot like jungle/drum’n’bass sample cutting, but I’m working with irregular rhythms, and they just don’t carry over with re-starts/repetitions like that. Another way is adding delays to ‘fill’ the gaps, but with the same shortcoming.) Maybe I need another sampler, sampling the re-sampled sample?! sigh

Still on the verge of not giving up hope it might somehow be possible! If it isn’t, at least I won’t go mad trying.

1 Like
#7

Hmm – I’m not sure if I’m missing something, but this is basically what timestretching tackles. If you pitch up a sample one octave without timestretching, you get twice the speed. turn on the timestretching for your playback track and the re-pitched sample should play at the same speed as you originally recorded it.

#8

This might well be the case, josker … I’ve never worked with time-stretching up to now, so I could have overlooked it entirely. I do want for the speed of the sampled material to be doubled, however. Basically I’m playing on a rudimentary electro-acoustic percussion setup, and re-sampling my playing in real-time to get a sound faster than I could play (well). Could I re-pitch the sample AND use time-stretching. I haven’t used it until now, because I don’t usually work with loops, and it also often mangles the sound … likely because of settings I don’t understand. I’ll give it a try, though. I’ll have to figure it out though (linking time-stretch attributes, figuring out how pitching up affects the length of playback, etc.), so any directions would be welcomed.

#9

samples playing at double speed are going to take half the time to play back as they took to record. (Sorry if that sounds obvious but stating it helps to clarify the situation I hope)

So, how much gap-free backing can your double-speed samples give you, time-wise?

Well, if you halve the time elapsed between start of recording and start of playback that tells you how long the sample playback will last (before it has to stop or loop).

If you record 2 bars and then start playback while you’re playing the third bar, the double speed sample is going to last 1 bar. So if you want fresh material playing back when you start the fourth bar, you need to have sampled your third bar. That sampled third bar is only going to last half of the fourth bar. By the second half of the fourth bar you are going to need some fresh material. Yes, you may have sampled the first half of the fourth bar, and that can last you while you’re playing the third quarter of the fourth bar (I’d call this the third beat, but let’s keep the terms unambiguous). Anyway, you can see that you are going to run out of usable material very quickly.

One ‘solution’ is to play unaccompanied for longer at the start. If you play for 16 bars without backing, you’ll have enough material to back your next 8 bars, during which you can sample enough to last you another 4. And so on.

So that ‘playing for ages before the samples start’ is probably too much of a compromise, especially as you’d still run out of material pretty quickly. You could play unaccompanied for even longer but the OT buffer will max out even with dynamic recording switched on and reserve recorder lengths off.

The next approach that comes to mind is one I think you have already dismissed. That would be chopping, stretching or otherwise manipulating the samples to make them last longer. This would be programmed in advance but, yes, unless you program very carefully based on what you are going to play, the backing will sound like classic break beat manipulation.

If you combine those last two approaches however, you might reach an acceptable trade-off. You could start the piece unaccompanied for a little longer than you originally planned. The samples generated during that period will last you for a couple/few bars, after which you carry on in the typical live sampling style shown in your diagram, except you PLUG THE GAPS by drawing upon the already-sampled material from the beginning (which you set up NOT to be overwritten) Because there’s a good amount of it you can play it back in chunks long enough to function more like phrases than chopped up breaks.

I’d also consider this approach - sampling TWO recordings of you whatever play. Two recordings that DIFFER (in timbre, relative levels of instruments etc.) enough to be used back-to-back, thereby filling the gaps without sounding like a straight loop. Perhaps radically different mic arrangements for the two simultaneous recordings?

Wish I had the time to get stuck into something like this!

Good luck!

3 Likes
#10

Woaw. It’s the beginning of a novel.
I think there are different solutions, I can work on it but if you still don’t answer my questions I may give up.

Would you prefer to double the speed without the 1 octave pitch with timestretch ?
In live, can the audience hear the normal speed recording ?
The 2x speed result must be minimum 4 bars without repeat ?
What is the maximum delay you can put up with ?

#11

Thank you so much for taking the time to reply!

There’s a lot here for me to grasp & digest, but I just wanted to say that straight away. I really appreciate the comments and support. I’m amazed I managed to convey what I’m trying to do.

I’m also going to upload an audio track later on, which will hopefully give a sense of why I’m pursuing making it work so doggedly. (It has been a long build-up of playing percussion, integrating a sound-palette that ‘works’ to my ear, and some familiarity with the terrain of the OT itself). From your reply I have a few things to try out and work with. (I’ve begun reconciling with the prospect of just recording the ‘half-time’ recordings, and pitching them up afterwards … it’s how I might go about ‘composing’ the various parts of a track, or maybe imagine performing with just an electronics setup. But being in the groove is something I just can’t/won’t give up on so easily).

You’re right that pre-preparing material alone would definitely be too much to ask, as it is precisely in the variations and modulations of rhythmic density/patterning, tempo, phrase-length/repetition, dynamics and accentuation that the vitality of this method seem to dwell (for me). However, filling those gaps with some range of material (especially using recordings of similar playing) and using random triggers to add variation and surprise myself is an excellent idea. There could be continuities and abruptly introduced material. And, yes, the ‘phrasal’ aspect is totally essential. And playing unaccompanied to get things going is fine … I suppose even longer recordings will eventually yield longer gaps, however?!

On your last point: do you mean “sampling TWO recordings … that DIFFER enough to be used back-to-back” when sampling into the OT live (on two separate channels, say?)? Or, when pre-preparing material for the gaps? I didn’t quite grasp this, but varying timbre sounds like a nice refinement too, if you could elaborate further … Hopefully if there are further refinements (if I manage to work with some of your other suggestions) I can follow up here, if you’d be willing/interested.

#12

I’ve just tried out what I suggested

If you have a Flex track which finds its source from somewhere you can do it on one track

Set the RLEN to e.g. 4
Place a record trig on 1 5 9 13
Place a playback trig on 1 3 5 7 9 11 13 15

now when you are recording in, the double speed buffer will play back 2 steps late but an octave higher

if you want it more 'granular just set RLEN to 2 and double record and play density of trigs
or if you like longer chunks repeating just go for Rlen 8 and half the trig density for rec play

it gives the effect I thought you were striving for

5 Likes
#13

Apologies, I missed your previous reply/questions. It’s the culmination of something, hopefully the beginning of something else! Maybe a cluster of serially composed poems to explode the continuum of history.

I usually record the microphones in input AB, and some electronic percussion in input C. That’s just because I’ve been trying to see if I can make this method work. I have a mixer, so they could all come in together, or separately. I guess I’d want to somehow record AND playback 16 steps … it may just have been my foolishness in thinking that would be possible to do. I’ve played with the scales, yes, but it’s an issue of the sampling ‘window’, as avantronica said. I could try doubling the speed and not the pitch (with time-stretch) but I need to figure out how: would that solve the problem, however? I assume I’d still have less material than I’d sampled, and so the playback of the samples will be interspersed with gaps? Maybe not, though. The audience won’t hear the original recording no … not that I have one! Whilst playing, I’ve been listening in headphones, so listening more to the re-sampled material than the original myself. The length of the repeats could be anything, if it yielded a continuous flow. (Though I think that’s the impossible thing to achieve, without some of the creative compromises/embellishments generously suggested by others here). What do you mean by “maximum delay”?

#14

Check Avantronica’s answer before. :wink:
If it suits you.

#15

I’ll try this when I get a moment to, but … if I do what you outlined, will the playback triggers 1 & 3 play back exactly the same material (i.e. what was recorded on recorded trigger 1)? Will it sound like a phrase repeated twice, and then do the same again? I think that’s what I was already getting, if so. (May not be the case. I’m very hands-on, so I’ll have to try it - may not have got what you meant, otherwise).

Also: could the record length just be ‘MAX’, if there are consecutive recorder triggers in the same track? I was doing this before … not sure if that may affect recorder behaviour.

#16

as I said before, you cannot accelerate time … there’s no solution to this if you want continuous playback at twice speed, you have to have that buffer play twice whilst the next buffer is then captured, you can shrink the moving window to diminish the sense of a long delay, but a double speed playback needs to be played twice unless you want gaps

2 Likes
#17

Yes, that’s sinking in now that I’ve exposed what I’m trying to do to scrutiny.
Thank you so much for your clarifying responses, and for tolerating/humouring my process of figuring-out.

I’m going to try what Clancy suggested, to supplement this ‘live’ method. Otherwise, I’ll work on my own playing to pre-record and then work with longer samples in to OT pre-loaded to a slot, make modifications after that, and blend it with other material.

I’ll set up things as you described and share a recording, if you’d care to hear it?

3 Likes
#18

[quote=“HEMIoDEMIoSEMI, post:13, topic:33378”]
What do you mean by “maximum delay”?[/quote]
I meant the time between the start of the recording, and the start of the playing.
What about other questions ? :smile: It would be easier for me with A B C answers.

A. Would you prefer to double the speed without the 1 octave pitch (with timestretch) ?
B. In live, can the audience hear the normal speed playing/recording ?
C.The 2x speed result must be minimum 4 bars without repeat ? (less is possible ?)

#19

D. What is the maximum delay between the start of the recording, and the start of the playing ?
E. Would it be musically possible to offset your recording ? (Rec 1 played on bar 2, Rec2 played on bar 3…Rec 4 played on bar 1)

#20

No answers…
I achieved something with 4 recorders, at least it is interesting for me.
I need my computer to write and draw something easier to understand my frenglish.
I can record 8 bars, play 4 bars speed 2x, R1 is played bar 3 so there is 2 bars delay (4 bars delay at normal speed).