[NO BUG] Connected Octatrack stopping by itself

[mntbighker]

I'm trying to track down why my Octatrack mk2 sequencer is stopping for no apparent reason. Apparently it will stop if it gets a spurious midi A1, and at the moment the Nerdseq seems to be the likely candidate. I' have no midi tracks in use, and the NS is providing midi clock to Octatrack and Analog Rytm. The Analog Rytm is not stopping and I see no extra events in VCV Rack midi monitor. I have tried different midi cables. Nerdseq is the only midi OUT (or through) on the bus. I'm working on getting a midi data stream recorder set up.

[XORadmin]

Yes, it would be great if you could provide the midi data.

However, I can also send you a special version where everything midi in an out is being recorded onto the SD card of the nerdseq. That might help, too.

But which version are you using? Can you provide your project so I can check if there might be any midi data except for the clock is being generated.
Also, did you contact the elektron support as well?

How do you know about A1 data?

[mntbighker]

Yes, it would be great if you could provide the midi data.

However, I can also send you a special version where everything midi in an out is being recorded onto the SD card of the nerdseq. That might help, too.

But which version are you using? Can you provide your project so I can check if there might be any midi data except for the clock is being generated.
Also, did you contact the elektron support as well?

How do you know about A1 data?

Working on it...

That could be useful as well.

I'm on the latest release NS firmware. I'll see about getting the project file gzipped and uploaded. No word back from Elektron yet. I've been through many iterations of project files and firmware versions with the same behavior. There was a point I was sequencing my Disting EX over midi, but not recently.

The A1 bug was reported by several people on Elektronauts.

[mntbighker]

Yes, it would be great if you could provide the midi data.

However, I can also send you a special version where everything midi in an out is being recorded onto the SD card of the nerdseq. That might help, too.

But which version are you using? Can you provide your project so I can check if there might be any midi data except for the clock is being generated.
Also, did you contact the elektron support as well?

How do you know about A1 data?

In fact I have NO tracks assigned to midi at all at present. So clock and reset are about the extent of it. I'm testing capture with Snoize Midi Monitor.

[XORadmin]

Yes, it would be great if you could provide the midi data.

However, I can also send you a special version where everything midi in an out is being recorded onto the SD card of the nerdseq. That might help, too.

But which version are you using? Can you provide your project so I can check if there might be any midi data except for the clock is being generated.
Also, did you contact the elektron support as well?

How do you know about A1 data?

Working on it...

That could be useful as well.

I'm on the latest release NS firmware. I'll see about getting the project file gzipped and uploaded. No word back from Elektron yet. I've been through many iterations of project files and firmware versions with the same behavior. There was a point I was sequencing my Disting EX over midi, but not recently.

The A1 bug was reported by several people on Elektronauts.

Just to be sure. Reported by several people in combination with the NerdSEQ?

[mvdirty]

I too am interested to see whatever midi data you are able to capture, as I have yet to see this behaviour with my NerdSEQ->OT setup (although right now I’ve enabled only very little in terms of MIDI in features of the latter.)

Out of curiosity, can you share your OT firmware version, AUDIO NOTE IN setting, track midi channel mappings, and the midi note mapping setting set for each track?

[mntbighker]

Yes, it would be great if you could provide the midi data.

However, I can also send you a special version where everything midi in an out is being recorded onto the SD card of the nerdseq. That might help, too.

But which version are you using? Can you provide your project so I can check if there might be any midi data except for the clock is being generated.
Also, did you contact the elektron support as well?

How do you know about A1 data?

Working on it...

That could be useful as well.

I'm on the latest release NS firmware. I'll see about getting the project file gzipped and uploaded. No word back from Elektron yet. I've been through many iterations of project files and firmware versions with the same behavior. There was a point I was sequencing my Disting EX over midi, but not recently.

The A1 bug was reported by several people on Elektronauts.

Just to be sure. Reported by several people in combination with the NerdSEQ?

No, reported as an Octatrack firmware bug. Not related to Nerdseq. I have the midi capture running now, but so far no Octatrack stoppages.

---

I too am interested to see whatever midi data you are able to capture, as I have yet to see this behaviour with my NerdSEQ->OT setup (although right now I’ve enabled only very little in terms of MIDI in features of the latter.)

Out of curiosity, can you share your OT firmware version, AUDIO NOTE IN setting, track midi channel mappings, and the midi note mapping setting set for each track?

OT 1.40A

No channels in Nerdseq are assigned to the midi interface. The only thing I'm using midi for on Nerdseq is sending clock. Are you asking for Octatrack settings or Nerdseq settings?

[mvdirty]

Those threads are a crazy-to-read mix of a) out-of-date and/or b) people talking past eachother.

As of 1.40A, at least, there is no bug. And there likely wasn’t before that. Interesting and perhaps-unexpected MIDI implementation, but not a bug as such.

A1 is by far not the only note with interesting behaviour, by the way. There are a whole range of interesting things happening from C1 all the way through C7. Check out page 137 of the OT manual.

For the rest of what I’m about to write, let’s assume that there are known (whether expected or unexpected) to be MIDI notes sending on channels the OT is set to receive either for audio tracks 1 through 8 or on the auto channel.

You’ll need set the project-level MIDI control scheme to a value other than STANDARD. If you set FOLLOW TM (Follow Trig Mode) then any time Trig Mode is set to TRACKS then you’ll get STANDARD scheme behaviour on the related MIDI channel, whether the one assigned for the current audio track or the auto channel. If you set the project-level scheme to MAP/TRACK then the individual audio tracks will default to STANDARD. Any combination of settings which results in STANDARD mode MIDI note processing will do all sorts of fun things from C1 through C7 on the relevant MIDI channel(s.) If you want this to never be the case, set the project-level MIDI control scheme to MAP/TRACK and then also set the per-audio track MIDI note map selections to appropriate values other than STANDARD. [You technically could set the scheme to FOLLOW TM and then be very careful to avoid setting Trig Mode to TRACKS, but that’s unlikely and risky.]

FWIW, I verified the above behavioural variations just a bit ago, using a test project sending A2 (yes, A2) from NerdSEQ, testing each of the 16 channels. And then I rebooted everything and verified it all again using a different NerdSEQ test project layout just to make sure.

That all out of the way, if it turns out you have truly no midi events flying around then there’s another way I can reproduce the behaviour, and that’s disconnecting the MIDI cable while the sequencer is running. Might there be a MIDI connection/signalling/etc. issue at play?

[mntbighker]

That all out of the way, if it turns out you have truly no midi events flying around then there’s another way I can reproduce the behaviour, and that’s disconnecting the MIDI cable while the sequencer is running. Might there be a MIDI connection/signalling/etc. issue at play?

That's an impressive level of effort on your part. I need to try and dig into this over the weekend if I can. Thanks for helping. Still very weird that the stop behavior just does not seem to happen if Nerdseq is externally clocked at 112, as opposed to 32.

[mvdirty]

That all out of the way, if it turns out you have truly no midi events flying around then there’s another way I can reproduce the behaviour, and that’s disconnecting the MIDI cable while the sequencer is running. Might there be a MIDI connection/signalling/etc. issue at play?

That's an impressive level of effort on your part. I need to try and dig into this over the weekend if I can. Thanks for helping. Still very weird that the stop behavior just does not seem to happen if Nerdseq is externally clocked at 112, as opposed to 32.

No worries, it didn’t take terribly long and I was happy to help, especially as it provided an opportunity to dig back into some OT user manual sections I had more-or-less only skimmed, and to experiment with MIDI control of the OT as that’s something I’d like to do a bit of soon.

As for 112 vs 32, and external clocking: one difference in my scenario is that the NerdSEQ was the master clock. I did, by chance of the NerdSEQ projects I loaded, try a couple different tempos. IIRC 73 and 120. Nothing as low as 32, though.