It can also be how the sample ends if you’re looping - so if you adjust the decay etc. to effect a small fade out, as well as the attack for fade in, it might help.
1 Like
Yep, I don’t think it’s possible once sample is saved to zoom in for that detail, so as you say here, do it when recording it in
OK guys, I’ll try all these ideas and come back on this. Thnx!
Longer sample means less precision with looping points.
1 Like
Yes I was talking about fades on the loop points for looping a sample.
that’s true. there’s probably a sweet spot with enough precision, plus enough parts of the sample to choose from, I guess…
I’m pretty sure that’s why we don’t get zoomin outside of the recording process… we’re stuck with the accuracy we have and although the situation is better than with the AR (which can only increment by 1, not 0.01) it’s still not sample accurate… so if we had the zoomin feature, we still probably couldn’t set our loops exactly at the right position. With this in mind, a non-destructive crossfade loop would probably be a better solution… provided the digitakt cpu is not already maxed out.
2 Likes
I see the digitakt and it’s marketed as a drum machine…
But the fades on the loop would be great for creating pads and textures etc…
And Would be a welcome addition
1 Like
Doing the math:
- DT audio is at 48khz, that mean 48000 samples for a second of sound
- 120.00 position mean we can reach accurately at most 12000 samples (is it 120 or 127 ? I can’t remember, not in front of the DT now…)
48000 / 12000 = 0,25
That means if we want to be able to reach any position in a wav with sample-accuracy, the wav must be at most a quarter of a second. So try to keep your wavs shorter than that and with a bit of luck you’ll manage to find zero crossings, loop points, etc…
1 Like
Suppose it depends on whether you want timbral variety in your loop – quarter second doesn’t allow for much of that (obviously more variety means it’ll be harder to loop without a click or obvious loop point, too)
Yes indeed… there again, a non-destructive xfade loop solve the problem. I don’t see that happening with th DT though.
How long is the sample that you‘re playing?
If we‘re talking something like 16-step lengths or similar, it‘s good to set the note lenth just a bit shorter than the loop like 15.5 and then play around with the amp release.
I‘m not quite sure either how exactly your issue looks like, maybe this helps.
Can’t you just resample it?
1 Like
Not sure how that would help – can you elaborate?
Put your clicking sample on a track, sample/record it and then try to find crossing points.
2 Likes
Ah yeah, I see - then you can zoom in, yep it’s a good point. Personally I’m happy with doing it by ear, I can usually find a good loop point.
Thank you for this idea. I’ve been struggling with looping single cycle waveforms and now it’s finally working!
1 Like
Resurrecting this thread - is someone else running into this roadblock? Looping is practically unusable.
Is there anyone from Elektron reading the forums? Would be great to have loop micro xfades in the next firmware update.
I had clicking when sampling FLEX loops. My problem was solved when I split my audio source cable to fill A and B inputs instead of only A.
Flex loops ?
Flex are Octatrack machines, can click too. (Digitakt topic).
I really don’t understand how it would solve a click problem.
New firmware with zero crossing detection for Werp and Slice machines.
I mostly experimented Slice machine, looped well. Never gelled with Digitakt Start / Length values, much prefer to use Slice machine with well prepared slices. Otherwise I use Octatrack, having a decent audio editor, with exact sample values precision (independent from playback Start/Len values). Digitakt audio editing can’t be precise as is.