Start Octa via MIDI

Hello all, I tried searching the forum for this, as I’m sure it has been discussed. I guess I need help with the forum search feature next, as I was only able to “find” a features petition thread looking for this, but I digress.

I want to start the octa sequencer via MIDI. This can be with or without the arranger for my needs.

I found the MIDI implementation section, and I have successfully recorded ALL midi from the octa, and played it back to the unit. Ex: I can record fader moves, etc, and if I hit play at the same time as ProTools, PTs will feed the midi back and the octa responds as expected (except for lack of sync).

I see that A# (34) is supposed to start the unit, but not matter how I send this note, or to which channel, I can’t get the Octa to fire off via MIDI. What am I missing?

Thanks in advance!

Midi sync menu> enable transport receive.

Make sure your daw us sending transport control to that midi out.

Ok, with a fresh brain this morning, I have made some headway.

In my troubleshooting, I made a new system and forgot to enable transport, CC, etc. Fixed that.

I am now able to get the machine to kick off! It’s not operating how I would expect however; “A#” does NOT start the machine as it should, I (90% of the time) have to use “B”. “A” does not stop the sequencer.

Does anyone have some experience triggering from a computer like this? Any tips?

Thanks again,
APN

Thx for the info Drew, I’m using specific moments as a start point for octa in my computer session, so I’d like to send the octa the transport info as if it were coming from a MIDI keyboard

Ah… gotcha… Yes then that makes sense. you can also use a blank pattern and record pattern changes in DAW where you want your actual patterns to start. I do this all the time, just play through pattern switches and it will send out a midi program message that you can play back.

Drew! Thanks for contributing to this thread.

So I can send “B” to start the machine. This works in seq. mode and arrange mode. HOWEVER; if I touch the play button in arrange mode the arrangement plays as expected. If I trigger the arrangement from the computer, the arrangement never gets past row “0” - that is, it starts, but never switches with the arrangement.

If I key in seq. edit mode, this is functional, but requires me to do some extra workflow to get at my end result. I’d much prefer to trigger the arranger.

Anyone try this?

A

Beyond unreliable arranger triggering, I’m also having this odd behavior;

I have a MIDI track set up to record ALL info from the octa.

As I perform the piece, there are a few tracks I trig free as one shots. In my MIDI track on my computer, I see these one shots show up in octave 0 keys C-G#. This track also records my cross fader movements.

On playback, the fader movements are replicated by the octa, but the one shots are not. Same midi track! What’s going on here, any guesses?

You may not have both program/bank send and receive enabled… they both need to be.

If you already did, have you tried… Record onto midi track the OT doing nothing but switching patterns. There should be program changes recorded that will play back into the OT if the OT is set to receive them.

you can also (depending on the flexibility of your hard/software) just send actual MIDI start stop continue messages (Look for System Realtime messages)

Avantronica - will this allow me to trigger the arranger? The deeper I get into this project, the more I realize that triggering the arranger is where it’s at.

I use ProTools11, or Logic. (I’d love to stay in PTs).

This is whatI must figure out;

The Octatrack arranger both sends and receives MIDI song pointer position.

I hope this becomes a useful thread for someone.

So I have enabled MIDI Beat Clock on ProTools and- It works! If I play from the top of the session, the arranger hums along and plays the arrangement.

(also have my send and receive set up for CC and Program change, thanks Drew)

So here’s the big question; is anyone, on any platform, able to send custom MIDI beat clock info?

For instance, I have an arrangement I want to trigger to video. This cue happens at some arbitrary point in the timeline (not the beginning). How may I trigger the arranger from this arbitrary point?

Anyone have a great solution for how to make this happen?

What do you mean by “custom” MIDI beat clock info?

MIDI clock messages do not start or stop anything - they just denote tempo. Other MIDI system realtime messages (Start, Stop, Continue) control transport.

Can you embed or manually trigger a MIDI start message (FAh) or a MIDI note that the Octatrack can respond to somewhere upstream of your OT?

What do you mean by “custom” MIDI beat clock info?

MIDI clock messages do not start or stop anything - they just denote tempo. Other MIDI system realtime messages (Start, Stop, Continue) control transport.

Can you embed or manually trigger a MIDI start message (FAh) or a MIDI note that the Octatrack can respond to somewhere upstream of your OT?[/quote]
OP is probably refering to SPP-messages:
[url=“http://www.midi.org/techspecs/midimessages.php”]
http://www.midi.org/techspecs/midimessages.php

If I´ve understood it correctly, a 14-bit System Common message that you can send and/or receive. If you get the value correct in the message, the OT Arranger should jump to that position. All relates to the start.

“Song Position Pointer.
This is an internal 14 bit register that holds the number of MIDI beats (1 beat= six MIDI clocks) since the start of the song. l is the LSB, m the MSB.”

Hey all, ressurecting a dead thread here, but this is very relevant.

I have an E-RM midiclock+ (http://www.e-rm.de/products/midiclock.php). it is very useful for resyncing several DAWs (e.g. Live) in case of MIDI clock drift. It does so by sending a “song position pointer” reset followed by a “midi start” message.

Tried it with several Live rigs and with Korg Gadget running on an iPad. Works as advertised.

Now, on my Octactack, I have both Transport and Clock receive enabled. It follows clock from the midiclock+ flawlessly. However, it does not AT ALL react to the song position pointer resets sent by the midiclock+. I tried in both Pattern and Arrangement modes, those messages seem to be simply ignored.

Page 42 of the OT manual states that the OT is supposed to send AND receive song position pointer. This thread confirms it…

Any ideas what’s going on?

Strange. Not sure what´s going on there. As far as I remember, it worked both ways. Will try to double check at home tonight.

A quick look into the pdf-manual of the midiclock+ reveals that, as you mentions, it is sending a “song position pointer” reset followed by a “midi start” message. It continues on page 8:

“It is possible that you still notice shifted tracks after pausing. In this case you can get around this by sending a ‘MIDI song position pointer’. The midiclock sends this pointer at the beginning of playback and re-sync, but not on ‘MIDI continue’. That’s the only way to ensure that all slaves are situated at the same song position and start in sync.”

I´m guessing this isn´t the problem for you since you do mention that the OT doesn´t seem to react at all. So the next step would be to try to pin point where the source of the problem is…

[ul]
[li]The midiclock+ manual doesn´t explicitly mention that it continuesly transmits SPP, is this something you can confirm either way? [/li]
[li]If you are sending SPP from your DAW(s), and connects it/them straight to your OT (without the midiclock+). Does it still not follow the SPP?[/li]
[/ul]

Thanks for the reply, @miketheman. I confirmed that the midiclock+ doesn’t continuously transmit SPP (using a MIDI monitor app (MIDI Wrench)). It only transmits SPP reset on pressing “resync”.
Haven’t tried sending SPP from DAWs. That would be interesting to try indeed!

But actually, I did find out a bit more about this since last night. 2 observations:

  1. if I pause, then resync, then continue: the Octatrack does reset to the top of the bar. So it must be receiving SPP after all.

  2. dug into MIDI spec, and found this: http://www.blitter.com/~russtopia/MIDI/~jglatt/tech/midispec/seq.htm. the interesting tidbit:

    “A Song Position Pointer message should not be sent while the devices are in play. This message should only be sent while devices are stopped. Otherwise, a slave might take too long to cue its new start point and miss a MIDI Clock that it should be processing.”

so i suspect the OT is ignoring SPP while in play, as it should per MIDI spec (if this document is to be believed). And the midiclock+ “resync” hack works for computer-based DAWs because they are designed to continuously listen for SPP in order to maintain some resemblance of sync (otherwise their clocks are too unreliable and drift apart). This last bit is only my theory which i’ve yet to confirm.

Thanks for the reply, @miketheman. I confirmed that the midiclock+ doesn’t continuously transmit SPP (using a MIDI monitor app (MIDI Wrench)). It only transmits SPP reset on pressing “resync”.
Haven’t tried sending SPP from DAWs. That would be interesting to try indeed!

But actually, I did find out a bit more about this since last night. 2 observations:

[b]1) if I pause, then resync, then continue: the Octatrack does reset to the top of the bar. So it must be receiving SPP after all.

  1. dug into MIDI spec, and found this: http://www.blitter.com/~russtopia/MIDI/~jglatt/tech/midispec/seq.htm. the interesting tidbit:

    “A Song Position Pointer message should not be sent while the devices are in play. This message should only be sent while devices are stopped. Otherwise, a slave might take too long to cue its new start point and miss a MIDI Clock that it should be processing.”[/b]

so i suspect the OT is ignoring SPP while in play, as it should per MIDI spec (if this document is to be believed). And the midiclock+ “resync” hack works for computer-based DAWs because they are designed to continuously listen for SPP in order to maintain some resemblance of sync (otherwise their clocks are too unreliable and drift apart). This last bit is only my theory which i’ve yet to confirm.[/quote]
Regarding:

  1. I suspected it to be like this.
  2. That´s my understanding as well.

AFAIK or understand the SPP: it is an arbitrary number that the master unit transmits upon transport action (Stop). So the master just basically have an counter that start counting the ticks that has runned since its start (which equals the minutes, seconds and/or whatever), that is when transport control Play actually has set it off to count. If the counter hasn´t counted anything (= 0). It will start from the beginning. Same if it been resetted. As soon as the master unit is set to Stop(/Pause) it transmits this number once to any slaves set to receive this SPP-information. So SPP tells any slave unit where they should start next time. Likewise, the midiclock ticks tells the units which tempo they should run in but has no information of where to start/pause/stop.

My understanding of SPP might be wrong though, so if anyone knows for sure what is correct or not. Feel free to enlight us all.

I would still check if it is the same between OT and DAW, just to be able to pin point and be a bit more sure of where the issue is.

…yeahhhh… So, I did some more testing and monitoring of the signal sent by the midiclock+, using the MIDI Wrench app. And confirmed that the device sends the SPP continuously.

Which means only one thing: it appears that midiclock+ basically mimics the behaviour of computer-based DAWs, in particular Ableton Live. Because sending SPP continuously is the only way software DAWs can have any hope of keeping some sort of sync.

Actually, the resync functionality seems to be tailored specifically to Live “scene” mode, because if you’re in the “timeline” mode (or using any other DAW, for that matter), then hitting “resync” would cause it to go to 0:0:0, which is definitely not a desired outcome in most cases. in Live’s “scene” mode, it resets all the loops to the beginning, and syncs the top of the bar, which makes all Live rigs jump back into sync.

so the resync functionality is designed specifically for Live’s scene mode, and makes no sense with any other DAW.

@miketheman, thanks for bouncing ideas back and forth - things make sense now - this has nothing to do with the Octatrack itself, and everything to do with 1) computer DAWs being sh*t at midi sync, and 2) E-RM not being completely forthcoming in their marketing materials!

live and learn :slight_smile:

(and regardless of any of this, i’m still super happy to be using awesome hardware instruments instead of mouse clicking my way through music production, and using silly hacks for something that should be as simple as MIDI clock ;))

Hey man no problem. Sharing ideas/thoughts are great ways to learn and both (everyone) may get something out of it… :+1:

I had some session myself yesterday with DAW and OT. And I can agree to a certain point of DAWs being a bit troublesome regarding syncing. However, it was still quite a rewarding experience.

I could set the SPP to be transmitted from Reaper. A bit amusing to have everything stopped and then touch the time marker in the DAW and just see the OT jumping back and forth in its Arranger rows to its corresponding SPP. But the real funs comes when you record knob/fader movements and then just having the DAW continuesly controlling certain parameters in the OT, no matter of its own arranger control etc.

I couldn´t set Reaper to respond to the OT MIDI transport control messages (System real-time messages). It seems to only allow either CC#s or computer keyboard buttons to controlling the transport controls. Which is possible to solve of course, but that would be quite an awkward method. Hoping to be proved wrong about this though, since I´m no pro regarding Reaper. My last hope is to make use of an script that would take care of that. But the problem is that I don´t understand the scripting language. So I´ll need to ask someone for help at the Reaper forum.

I will reply to this thread as soon as I know for sure either way.