Midi LFO bugs

Working on a live setup involving my modular rig, a Digitone, and an Octatrack. Currently have the Digitone triggering my drum modules via a Erica/Hexinverter Mutant brain. Thought it would be nice to make use of the MIDI LFOs in the DN since I have some extra ports on the mutant brain.

Was able to assign the midi CCs on the CC Select Page, and affect it with the corresponding Track’s LFO, but I’m noticing some inconsistent behavior with starting/stopping the LFOs and switching patterns.

If I monitor the outgoing CV on my scope, I can see the LFOs running (I have one on each track), but unless there are trigs on that track, sometimes the LFO won’t start, and/or, if I switch patterns and the new pattern doesn’t have trigs on that track, the LFO from the previous pattern will keep running. Also, if I stop the pattern, the LFOs keep running until I double tap stop.

I have the LFOs in free running mode, not trig mode.

anyone else run into this?

Expected behavior. There’s nothing telling the lfos to stop or change or otherwise. With midi most of the time you need to program everything deliberately. If you want an lfo to stop on the new pattern, best thing to do is add a (lock)trig with lfo depth set to 0 on the first step of that pattern.

But let’s say I have the same Midi CC settings and free running LFO assignments on all my patterns, and the only difference between patterns is the LFO speed and shape.

When changing patterns, why does the MIDI LFO not also change unless there’s a note on that track? Why does it retain the LFO speed/shape from the previous pattern?

Because if you have a note from the previous pattern still playing, you usually don’t want the settings to change. Having the option to put a silent trig with “1st” in order to stop the previous sound / force the new parameters is better than not having the option to let the previous pattern’s notes carry over to the next pattern.

1 Like

Ok, so is it a general thing with elektron boxes that LFOs are tied to notes? On most synthesizers, a “free running” LFO is exactly that.

It’s not tied to notes. It is free running. The notes or lock trigs rather help in sending out the new state of midi cc and lfo settings. If that makes sense.

1 Like

I understand what you’re saying, but I wouldn’t say it makes sense, if you understand my meaning. what would make sense would be that when you change patterns, the free running LFO settings from that pattern are instantly applied.

seems like a bug to me.

It’s not really a bug: A pattern change will not update the currently running LFO’s per se.

When the sequencer is running, LFO state changes need to be send out bij trigs (with or without a note). This applies for both default values of the LFO in the pattern or when the LFO values are set explicitly with a trig.

The exceptions are:

  • When you press play when the sequencer is not running (then the LFO takes its initial values from the current pattern)
  • When you manually change the values by using the encoders (then the values are instantly updated)
  • When you change the values by incoming MIDI (idem)

It does make sense. It’s just not the way you’d like it to work, which is understandable. :slight_smile:

It’s not a bug, it is how it is supposed to work. There are situations where you actually want the LFO’s to continue from previous pattern. On Octatrack there’s even a specific option to turn this behaviour on or off. On the other boxes there is no such option and it works like you described by default. Mainly because it’s fairly easy to add a trigger to the sequence to ‘reset’ the lfo parameters.

yeah, I experimented a bit more, first I tried putting a trigless lock on beat one of each midi track (unless there’s already a trig there), which works for changing the settings. and fits most situations, but then I realized that sometimes I want the LFO to be in time/phase with the beat, but still in free running mode.

In that case I’ll put trigs on Beats 1 and 2, with the 2nd microtimed all the way to the left. Then I use the 1st to p lock the LFO into trigger mode, and the 2nd to set it back to free running. Sometimes I need to adjust the phase of the waveform depending on the effect I want, but it seems to work well enough.

Your explanation makes sense from our current point of view.
But I would expect in this case the CC Value Setting to do the job.
Only visible CC values should be sent out. There should nothing been sent if value fields are filled with X.

Probably it’s not a bug if it’s not conform to expectations.
The worst about this workaround is that every pattern of a project is potentially concerned.
Maybe it’s better to put a silent trig at the last step of the midi track to set the LFO depth to 0. So you you don’t need to do it in the beginning of each pattern.
Edit:
I just tried it. All you have to do is to set LFO depth to zero at the last trig playing a note. I wouldn’t have come to this solution without your informations . So thanks for helping each other !