Issues with pattern changes from computers

Elektron Gear
#1

I have been creating a song arranger for the Digi boxes in FLStudio (PC) and Xequence2 (iOS) so I can easily recreate songs for recording to release. I’ve managed to get my automation recordings to work very well, but I’m still having issues with program changes. I’m posting this on Elektronauts, FLStudio, and Audiobus forums in hopes that someone can help me out.

I have a MIDI Out channel on FLStudio hooked up to the keyboard controller so that I can automate pattern changes on all 3 Digis at the same time. If I only use one box at a time, they work fine, but once I introduce all 3, I end up having issues. You can see in the video how the first 2 PC work fine, but afterwards, they’re all an extra bar late, even with me making up for it with the placement of the triggers. I’ve tried USB and MIDI DIN for MIDI transfer, and this happens on both my iPad with Xequence 2 and my PC with FLStudio.

Any ideas?

#2

I’m fairly sure someone recently posted a solution to the “one bar late” problem when driving elektron sequencers. I’ll see if I can find it.

EDIT - this is what I was thinking of. Pattern Change Between Digitone / Syntakt / Digitakt (FIXED) - #22 by Groundman

#3

I can’t know for sure, and I didn’t read the post linked by @bibenu, but here is a few insights on this:

  • pattern change is subject to CH.Len parameter in the scale menu of these device (this parameter is available only when scale is define “per track”). It defines how long they have to wait before changing to a new pattern when they are asked to.
  • pattern change data (program change via midi, midi note via overbridge, or even manual change on the buttons) has to be received before the beggining of a new loop cycle. As your midi items are right on the grid, it is likely that some/all of them receive the program change too late (due to midi latency). A efficient way to handle this is to a small negative delay to the item on the daw timeline to ensure the command is received before. You can juste nudge these items, or better ask your DAW to playback your midi item with a small negative delay. (I don’t know if that is possible in FL studio).
#4

All of those things have already been accounted for. Thanks.

#5

This has already been accounted for. Thanks.

1 Like
#6

Ok my bad then.
Did you solve your issue ?

1 Like
#7

No worries! I appreciate the attempt. I haven’t had time to dive back into it. I’m going to try some other stuff this week, including a MIDI monitor to see what’s happening. It’s weird that I can trigger it fine manually. I wonder if it has to do with the length of the PC message, if that’s even a thing.

#8

Just doing the obvious check here but are you using a midi splitter or are you daisy chaining?

#9

Nope. All USB MIDI.

#10

I would try using physical midi connections. Sounds like something that would be solved with a quadrathru.

#11

I’d got the impression from Elektron customer services that pattern changes via PGRM CH were not subject to Ch.Len, but instead M.Len.

1 Like
#12

I think it’s not the case, it is contradictory with the manual.
From what I understand:

  • M.LEN (Master len) defines at which point a pattern should restart. My guess is that its main purpose is for polyrythmics patterns. For instance if you have track 1 with 31 steps len, and track 2 with 37, the pattern will restart at … 1147. To prevent this M.LEN allows you to say “ok at step M.LEN the pattern should restart from the beginning”
  • CH.LEN on the other hand gives you the number of steps before a pattern change command
#13

My conversation with customer services included them acknowledging the manual was wrong. I’ll dig it out when I have more time.

I caveat all this… the boxes behave differently when M.Len is set to INF and not. I may be muddling my memories of the two.

I’ll get back to this when I’ve had time, later this week…

#14

Ok I am interested by those experiments, I would gladly hear your feedback on this

#15

@Octagonist is correct, it is not affected the same way by external program changes normally as I use it on a regular basic via manual triggering.

I have found that I think, somewhere, the MIDI feed is getting congested with multiple CC automations being sent out, as well. I’ve found that cutting out blank or unmoving automation data during my arrangement helps prevent the issue. So does restarting FLStudio or the Elektron boxes. I can’t figure out where the congestion is, either.

#16

Maybe you should try with another software (not indefinitly just to figure out if the problem might come from FL studio or if it is a a general issue) ?
You may try in reaper for instance, I can guide you through the process (to avoid spending time watching reaper tutorial). I only Have a digitakt so I can’t make the experiment myself.

1 Like
#17

I also tried with Xequence 2 on my iPad, though I need to try again now with it having sort of worked before. I’ll try Ableton next.