Hi,
I recently built an LXR and am still trying to get my head around it all but ... I'm having a lot of trouble getting the LXR to trigger every time I send it a MIDI note from my octatrack. By that I mean the external triggers don't work, the internal voice doesn't play and the LED does not light up. How much this happens depends on how many triggers I send the LXR but the more I send the more get dropped and the same triggers appear to get dropped over a 2 bar period as long as the transport is running. I've tried resetting all sorts of things but nothing has helped.
Meanwhile I don't have any trouble getting the LXR to trigger if I trigger it with a MIDI keyboard. The only difference I can see in the MIDI messages being sent is that the octatrack sends a noteoff velocity of 0 while the MIDI keyboard send a note off velocity of 64 but this could be a red herring.
(The only reason I'm trying to sequence one xox drum machine with another is the arpeggiator in the octatrack, I actually like sequencing in the LXR)
Comments
the version number is shown on bootup.
note off velocity is completely ignored by the LXR so this should be no problem.
since it is an arpeggio I guess you are trying to trigger it with different notes?
have you tried triggering it from another midi sequencer?
does it work there and only happens with the OT?
I'll try some arpeggio triggering the next time I have my lxr hooked up to ableton and see if it works there.
so far I did most of the midi trigger testing with a keyboard, too.
I like using arpeggiators set really slow to randomly select between multiple percussion voices. But it seems I can send a different arpeggiator through the octatrack and trigger the lxr just fine without any dropouts!? So I guess it must be something the octatrack is doing which is a bummer since I can't see anything different other than the note off velocity.
If I send only 2 midi signals - note on and note off with an off velocity of 64 I never have had any trouble triggering the LXR. If I send note on and note off with an off velocity of 0 I have a lot of trouble triggering the LXR with more missed triggers the more rapidly midi triggers are sent. Sending the same off velocity of 0 to another drum machine results in no missed triggers.
I will have a look in the code, but I think nowmally the velocity of a note off message is completely ignored.
Only the note on messages are used to trigger the voices.
Should still be fixable, but might take a while...
I think you are right!
I was experiancing this same problem with a Midibox MB SEQ V4 with a STM32F4 core. Once changed the MIDI output to a I2C Midi out (which is 5V) it started to work correctly.
- Reliable triggering, when the external sequencer does not send midi clock (start/stop/continue is OK).
- Random trigger skips when external midi clock is received and individual LXR track notes are set (ie anything except 'any').
- Interestingly if every LXR track has its own midi channel and notes are set to 'any' no skips occur even with external midi clock. This may be a temporary workaround, but it eats 7 from 8 midi channels on the octatrack.
At least in my setup I don't think the problem is hardware related (identical results from OT and MPC, I can try more sequencers if necessary).
Hope that helps with debugging. Let me know if you need additional experiments/tests.
Best,
Christoph
I tried playing around with my gear and yes if no clock is being sent the lxr triggers fine ... but if clock is being sent and I set all tracks to any with their own midi channel like you suggested I am still missing quite a few triggers.
I also noticed that whatever is missing from the lxr is also missing from the soft midi thru which explains some problems I was having chaining stuff after the lxr. So even if you are not trying to trigger the lxr over midi but are trying to trigger something else down the line while sending clock you are in for a world of frustration and head scratching.
Luckily I just bought a larger midi thru box...
Yes, I am in the meantime not convinced that all triggers are getting through. But it seems to work better than on a single channel.
BTW I am getting the same results with a RM1x sequencer, looks like a LXR problem.