Really easy time stretching on the Digitakt

Elektron Gear Digitakt
#1

Hey all, I figured out a way to do timestretching natively on the Digitakt so I made a video to document the method. Setting it up is simple:

1: on the SRC page set the sample start point to 60.0 (the middle of the sample)

2: on the LFO page set the LFO destination to sample start, LFO type to saw, the speed to -8.0, and the depth to 60

3: place a trig on each step in the sequencer and parameter lock the LFO to trigger on step 1

That’s it!! Once you set that up there’s a lot of variations and ways to improve the method. The most basic thing is to alter the amplitude envelope. Try adding a little attack time, short decay time, then bring the hold time all the way down so you can hear the grains at work.

Next thing is retriggers. You can increase the density of grains (aka resolution) of the timestretch by adding retriggers to each step. Experiment with different retrig multipliers and lengths. Try parameter locking different envelope, pitch, and sample slot settings per trig and lower the probability a bit (make sure to p-lock trig 1 at 100% if you want to stay synced).

Also it’s possible to play back 2 or 4 (or longer) bar samples by lowering the LFO multiplier accordingly and experimenting with more trig pages, grid dividers, and 1:2 or 1:4 conditional trigs on step 1 (LFO realigning trigger)

TONS of possibilities, this video just scratches the surface. More videos coming soon, like and subscribe if you want to see more and let me know what you think. excited to see what you create with this method :sun_with_face:

Here’s the link: https://youtu.be/vTHeVaUot7I

81 Likes
Timestretch loops on M/S?
Can it time stretch like digitakt does?
Tell me things i def have to try w/ my digitakt before letting it go
Timestretch loops on M/S?
#2

Saw this on Reddit over breakfast. Fantastic job.

1 Like
#3

great work! the retrig rez trick is really good thinking.

2 Likes
#4

lots of discussions on this technique for DT and AR before that, it’s always encouraged to search before creating new topics, but here linked are some of the others for info anyway

4 Likes
#5

Ive searched quite a bit and it seems the only person doing the same thing as this is @garf (what’s up dude??) Most other methods I’ve seen have harsh discontinuity artifacts.

This is the best way I’ve found to just play back samples and keep them in sync with time stretching so I made a separate thread. Cheers :grin:

9 Likes
#6

I don’t really hear how this method improves on others TBH to merit it’s own topic, plenty of audible artifacts here also :man_shrugging:

#7

I’m not claiming this is artifact free - rather that the artifacts you get are the same ones you get with “real” timestretching vs the harsh discontinuity artifacts from the “short looping time window” method I’ve seen most commonly used.

5 Likes
#8

This is awesome! I’m glad you posted this in a new thread. Otherwise I wouldn’t have noticed.

13 Likes
#9

Cool, thanks for the tips!
I used to another method working on long samples

But I will check your asap :wink:

1 Like
#10

On the contrary, I find the step-by-step tutorial posted by the OP to be vastly different from all of the links you mentioned. They’re all about extreme glitchy stuff and don’t go into detail of what is actually happening.

7 Likes
#11

I assume that if you set the amp envelope to have a faster decay, there will be fewer artifacts on those speeds that otherwise sound glitchy. This is also how Ableton’s warp method works in the beat mode.

2 Likes
#12

For whatever reason this does not work for me :man_shrugging:

It just machine guns the sample loop… :expressionless:

:sob:

Edit: got it working when setting the BPM multiplier at 4 instead of the default 16. :man_shrugging:

2 Likes
#13

Heres how to use the method to chain longer clips. https://youtu.be/rCusXxrftK4

2 Likes
#14

THANK YOU! :3lektron:

1 Like
#15

That’s a nice trick! I’d be interested to understand the maths of the -8 and particularly 30 in the LFO.

4 Likes
#16

LFO Depth of 30 is a division of the sample’s length which is always 120. The Speed of 8 is a division of the step numbers.

The length of a sample is always 120 on the Digitakt, and the LFO is bipolar, so you have to set the sample start in the middle at 60. An LFO depth to 30 sets the sample start to 0 (seems like it should be 60 - half of 120 - but that doesn’t work). You can set the SPD to 8, 16, 32… whatever works. OP is using -8 SPD with +30 DEP. I use +16 SPD with -30 DEP and adjust the MULT BPM to double/half the speeds.

8 Likes
#17

This is the bit I didn’t get! Does the LFO oscillate by 200% of the target?

#18

heh, good question! i still don’t know why.

2 Likes
#19

Experienced exactly the same thing. Had a 4 on the floor Housebeat Loop. With the Multiplier at 16 it got played back 4 times as fast. With MULT at 4, tempo was correct.
I did not understand it in Theory, because the OP used the same Setup with Multiplier set at 16 and it worked out.
Does someone know why?

Edit: Because i‘ve loaded a 4 Bar Sample.
If you load a 1 Bar Sample, set MULT at 16. With a 4-Bar set MULT at 4 when using a 4 Bar Sample

5 Likes
#20

That constant flashing of the trig buttons :crazy_face: