Is the LFO run mode set to “FREE”?
It sounds like it is operating the way a free running LFO would.
The MIDI track mutes mute notes in the sequencer, so it makes sense that the LFO would continue to modulate, independent of the mute state.
Sometimes switching the LFO to other modes stops the stream, sometimes it doesn’t. I had thought of that before posting. Sometimes the LFO doesn’t actually start sending data right away either. It’s a little funky.
Switching to another EMPTY pattern (while sequencer is stopped) doesn’t stop it either. That’s extra funky.
Regardless, FUNC+Press on the SRC page, to “X” out the MIDI Channel doesn’t stop the data stream, and intuitively, it probably should. ALL NOTES OFF should stop it as well, or maybe some kind of PANIC key combo is in order.
Here is a screen shot of the MIDI data coming into channel 1, after having switched to an empty pattern.
Yea, so what I do is have a pattern w/ LFO on channel 1, then switch to an empty pattern (stopped), and the LFO stream on channel 1 keeps pumping out data.
It’s not a show stopper, just interesting to me. Some of my older MIDI devices “choke” when too much data is sent to them, and I was having issues w/ runaway LFO data from DT to these synths, so at least I know how to avoid that now!
I’m also wondering if some of this funkiness w/ LFO’s is responsible for some of the MIDI issues ppl doing loopback were/are having.
Also, I can get the stream to stop immediately, by sending a NOTE ON message down the channel path. So, let’s say the LFO is running amok on an empty pattern - I can get the stream to stop by setting up a new track on that channel, making a trig (note on message) and then hitting PLAY.
So in practice, when switching between patterns that are running, it shouldn’t be an issue - as long as the pattern you are switching to also utilizes that MIDI channel, AND has a NOTE ON message someplace. I think.
See how NOTE on killed the stream? LOL
does the empty pattern still contain the same LFO CC# assignment as the previous pattern? Or is it unassigned?
it’s completely blank/initialized! No assignments anywhere on anything!
Be great if somebody else would see if they get this behavior? Hard to tell w/out using a computer though. The flashing MIDI in light on my moog was the giveaway. Could see DT was spitting out CC data in an unexpected fashion.
1 Like
wild!
I think this should definitely be reported as a support ticket at elektron.se to be sure that Elektron knows, and the devs have it on their radar.
While maybe not a showstopper for you, it might exacerbate other MIDI issues.
1 Like
totally, i’ll submit a ticket.
2 Likes
Hmmmmm, doesn’t this have to do with the fact that the lfo is waiting for a trigger when you switch to a trigger mode?
When you say it stops free running when you send a note on message, this could be because you are triggering the lfo in a 1 shot trig mode for instance.
So when you have it on free running mode and switch the mode to trig one shot, and it keeps running, when you sent a note on message, you also sent a lfo.t message and then it will 1 shot and stop.
Can you check if this is what happens? That would actually be quite normal behavior I think.
2 Likes
i had today a ghost sample playing. os 1.06
Summary
(i´m on the newest OS , 1.06 or so )
i just had a gost sample playing. track 8
i was not able to get that sample to NOT play other than to take down the volume for that track. trk 8
i tryed different ways that should have worked to get that playing sound to stop. taking all triggs off etc. stopp the machine and start again etc.
interestingly was the sample SRC set to slot 0, respectivly off.
i never do that
( i have NO long samples in my machine, a few that are 4 bars, nothing longer )
AND interestingly was the sample play set to reverse-play ( set by me much likely)
if its a bug, the relation would/might be:
trk 8 ( see the post in the other thread linked abobve ), no sample selected for the track, reverse playing.
when i set the play direction to forward was the problem solved IIRC
was an old project of mine.
not used it over many months as the machine was off use for many months also
could be related to this thread
For some reference here’s what happens on OT…
If you mute a midi track with a running lfo it definitely will stop the lfo right then, even if it’s free…
If you switch to a new pattern the lfo will keep running until it hits a trig on that specific track to update the lfo settings to that patterns current edit page settings. OT has an option to auto change the lfo settings at the beginning of the next pattern/part…
I know it’s kind of an esoteric potential bug!
Thanks though, but it’s probably not up to me to figure it out. Just to live with. 
Not how it works on DT
Yea I don’t know. I’m switching to a new init pattern w/ sequencer stopped, LFO keeps running on that channel, even though nothing on the new pattern is assigned to that channel. Might just be the protocol. Good to know though!
Not running it in ‘trigger’ mode, in ‘free run’ - so no.
feel free to try and replicate what i’m seeing - or not. don’t matter!
Had a bug yesterday, not encountered before:
Recording in samples, saved named etc and assigned sample slot,
work on project save multiple times,
switch off the unit and come back to it to find the recorded sample is no longer assigned to the track and is no longer in the plus drive.
The other samples i recorded in after this one & before the switch off still exist 
.
Showed no waveform on the sample page.
Unfortunately Elektron support can’t replicate this problem. Can anyone else check?
Not possible you used the same name twice?, I did once and when I switched back on it was gone.
Would it still work in the project if that was the case ? It only went away after restarting.
I usually give em unique names aha.
As i said before this is the same problem for me. I will report it.
Thanks. I even sent a video but they couldn’t see the problem on their machine
I also have this problem. Undo sometimes just doesn’t work. If I clear and undo a few times, there always comes a point at which the Digitakt just acts as if there was nothing to undo and I’m left with an empty pattern. Sometimes it’s after the first clear, sometimes it takes 4 or 5 times.
Can anybody get these to work (all during live recording mode, see current manual page 34)?
To remove the parameter locks of all tracks while LIVE RECORDING mode is active, press and hold
[FUNC] + [CLEAR].
This one just clears the pattern for me and that’s where I ran into the aforementioned undo bug… 
Press
[NO]+ one (or several) of the[TRIG]keys to erase a sequence of recorded trigs on a specific track (or on several tracks) in time with the sequencer, i.e. all steps reached on the pattern playing will be erased until the keys are released.
This one doesn’t delete all trigs, somehow there’s always one left when I try it.