Digitakt 1.11 : Bug reports

One bug I am still experiencing from time to time is that after saving a sample and returning to the sampling screen, all of the controls are gone, just the sample waveform is visible - until you exit this screen and re-enter it again:

Another visual bug, still present since the very early firmware versions, is that the sample waveform in the sampling screen will sometimes display either a high peak or a gap when there is nothing like that in the actual sample, and is thus misleading:

is that not the loop point?

Welcome to The Matrix! :slight_smile:

No, these are not loop points, actually you don’t set loop points in the sampling/recording screen, you set the trim start/end points, and in this very case they are not even adjusted yet (as you can see at the top left and top right views). This is a bogus peak that appears where the sample is totally silent.

Hello, first time poster here, I saw this issue had been experienced by other users back in 2017! Am I crazy or this is still a problem?

Basically I’m sending midi CC messages via MIDI track A to one of my synths (Monologue) and I’ve set the LFO of that MIDI track A to change the value of the CC command I’m targeting.
Although when I mute MIDI track A it still affects my synth, the only way I found to go around that is to turn off the MIDI source channel via the SRC tab but it’s dumb.

Anybody know if this bug got fixed? I have OS 1.11

Originally experienced by this user - Digitakt 1.04 : Bug reports

This is expected behaviour. You extend pattern B when in reference to to pattern you are not at the first bar. It can mess with your mind but take a pen and paper and draw one line that is 4 bars and under that a line that is 4x 1 bar. Then extend one of them to 4 bars and you will see what it going on :slight_smile:

Set the reset parameter to the same amount of steps as your longest pattern and they will sync up. This is what this is for.

2 Likes

Thanks! Now I know the reason to have M.LEN something other than infinite.

I just tested this with a Pattern M.LEN of 64, and it works correctly. It allows me to add pages and record with the sequencer running and everything aligns. Works well for making “straight” beats in real-time.

Best regards,

Gino

2 Likes

I’m not sure that this would be considered a bug:

Muting a track on the Digitakt means only “ignore the trigs”. LFOs will continue to run, and will continue to modify the parameter they are controlling - on both audio or MIDI tracks.

If you want the LFO to stop when there is nothing playing on the track - Either set it to TRG mode and give it a FADE out, or set it to ONE or HLF modes, depending on what you need. Then it will only fire when there is a trig, and mute will keep it from firing.

2 Likes

Makes sense thanks! Would be cool to keep it running and only have it be active when the track is muted though.

Hey guyz, I’ve been battling with a bug for a few months and just sent a ticket to Elektron. Thought I would share to see if anyone would be able to replicate this. I’m too lazy to re-write the whole thing so this is just a copy/paste from the ticket :

I’m running OS 1.11 and I use my Digitakt for live looping, I usually use my Keystep to play different DT tracks. I do all of this as the sequencer is running and I quickly switch between MIDI channels on the Keystep. The issue is that sometimes I will play the full cycle of my melody (64-steps long) but when the sequencer reaches the end and starts over, the first notes won’t playback. Only the second half of the melody is recorded and plays back.

So far the bug was not very consistent. Sometimes it would never work and I would have to play the full melody twice to be sure it would be properly recorded, some other times it would work as expected. However, I think I discovered the origin of the problem yesterday! This happens when the highlighted on the DT has a shorter length than the track I’m recording on. For example if the track 8 is selected on the DT and has a 32-step length, I won’t be able to record a 64-step melody on track 1 properly.

Here is a step-by-step guide to replicate this bug on a blank project (you will need an external controller, I use a Keystep) :

- Create a new project
- Enable the Length per track option on the Scale page
- Set the length of track 1 to 64 steps and the length of track 8 to 32 steps
- Set the M.LEN to INF
- Exit the Scale menu and highlight the track 8
- Now put the metronome on and simply press Rec+Play, record some notes with your MIDI controller on track 1 which is the kick by default
- If everything goes “well” the first 2 bars or so won’t be recorded, only the second half of the pattern will play

If you wait for one full cycle before playing some notes, the pattern might be recorded properly, but that’s kinda random in my case.

Any input on this? Am I right to assume this is a bug in the first place? Cheers<3

Is this related?

Best regards,

Gino

Hey Gino, thanks for your input! I wasn’t aware of that problem, that’s kinda weird but I can imagine how changing the track length could mess up with the metric of it all. However in my case there’s no sudden length change, I set the track lengths before starting the sequencer at all. Also what’s weird is that the DT doesn’t behave the same depending on which track is highlighted, that’s the main thing that makes me think that it’s not “working as designed” but who knows!

Anyway thanks again and have a great sunday. :slight_smile:

1 Like

Same here !
I have the same behaviour here.
my problem is if I use different scales for 2 tracks I 'll end with the first step on one of the tracks to have buggy restarts of the pattern. I can hear delay between tracks that seem to correct themselves a few steps later.
on the pattern I am working at the moment, I have 64 steps with 1/2 dividers on one track, 64 1/4 on a second one and only 16 steps 1/2 on the next one.
I should mention that I tried to set master length parameter to values 64 / 128 / 256.

this problem only exists if I set the different tracks to different lengths/ scales.
once a pattern’s sync is “corrupted”, I can only work with it in “scale per pattern” mode ie same number of steps / divider on every track.
I tried to find a workaround without success.
I submitted a ticket today.

Okay so I found something funny regarding the LFO speed, I’m running 1.11 since a few days because I couldn’t be bothered to update haha.

I’ve got a sequence at 152 bpm with a sample running on track 8, the sample fully covers 32 steps. The LFO is routed to the overdrive and should be accentuating all the 16ths. SPD is set to 32.00 and MULT to BPM 64. That’s all working good and fine.
Now when I introduce another track with a different LFO speed, the LFO of track 8 somehow goes all out of sync. Specifically, I use track 3 which has got the SPD at 51.43 and a non-bpm-synced MULT set to 256.

According to the manual, the non-bpm-synced MULT assumes 120 bpm; I suspect these different bpm settings conflict, so that track 3 somehow influences the MULT on track 8 with a wrong number or something. It’s kinda weird that one audio track influences the behaviour of another.

Is this a known issue? Is it even clear what I’m talking about?

1 Like

It happens whenever you enter per track scale from a pattern multiplier other than 1x. To fix it, just revert back to your original state, and switch to x1, and then enter per track scale.

1 Like

Hi! Thanks for the reply, but I don’t understand what you mean. How do I revert back to original state?

Revert back to original state = Reverse engineer to the state before you initiated per track rate scale.

IE. If you adjusted rates from a pattern rate scale of 1/2x:

  1. Switch all tracks back to 1/2x.
  2. Switch back to per pattern rate scale.
  3. Switch from a global rate scale of 1/2x to a global rate scale of 1x.
  4. Switch to per track rate scale.
  5. Adjust the rates back to the desired rates.

Perhaps there is a more elegant way of doing this, but I haven’t found one.

1 Like

Had this happen tonight while jamming away, hard boot needed. Latest FW.
Anyone else run in to this one?

A few people:
https://www.elektronauts.com/search?q=exception%20error%20%23elektron-gear%3Adigitakt

Best to report it to Elektron.

1 Like