Sync with MPC Live

[Rastapoef]

I'm having trouble syncing the NDLR. 

When MPC is master, the NDLR seems to keep its own clock but running at the incoming pace, and adjust it constantly to get in sync, like a DJ with turntables, nudging, a little faster, a little slower, etc., you can clearly see it happening watching the tempo indicator. Sometimes it just dropped to 3/4 speed when starting, totally mismatching. 

The other way around, which I don't want because then the MPC can't select (and store) the patches for my synths, doesn't sync nicely either. First it was receiving double midi or something, through the interconnected USB cable which I thought I was just using for power. It was running 2 times too fast. Anyway, at the right tempo, it's not tight at all. 

Uhm...   
The unit should be able to be a tight midi slave, right?

---

Hmm I've been reading up on the MPC Live midi clock history. On the interweb. That's not hopeful. 
I'm running a recent enough firmware for the known issues to be fixed.... 
That wobbly clock is probably courtesy of MPC Live. 

There goes my plan for this little setup 

[Darryl]

It would be easier to troubleshoot if you could tell us what sounds like its out of sync? You hear a delay from the time you press start on the clock source to when The NDLR starts playing? Or does someone get a beat off tempo?

The NDLR doesn't "regenerate" the clock, it uses the external clock ticks as they arrive to drive the parts, just as if the NDLR were generating them internally. So if the external clock isn't perfect, The NDLR won't be either, but it will be in time with the clock source (for better or worse). The trade-off to this architecture is, the NDLR practically "hangs" when there is no incoming ticks. MIDI is always the priority and the heartbeat of The NDLR so it won't move forward without them.

When using external clock, the clock display on NDLR is a real time BPM meter. It accurately reflects what it is receiving. The BPM calculation happens very frequently and there is no "smoothing". Its what you see is what you get. We've compared the BPM display to several other measuring tools and its actually more accurate and faster than most other hardware devices. For instance, if you apply external clock to a Beatstep Pro, there's a delay before it shows the clock, and probably some smoothing going on so its not jumping around a lot. We thought about doing this too, but we liked having the real time measurement and the decimal place.

On the other hand, if there's a tick with a severe jitter and The NDLR BPM meter happens to sample it, it will show it. However, you will never hear that as its a fraction of the 24 MIDI clock ticks that make 1 BPM.

Do you hear something that makes it sound like its out of sync?

[Rastapoef]

Well, I think indeed what I see when it's slave, is the totally unreliable clock of the MPC Live.

I will have to try and slave the MPC to the NDLR clock. And have the MPC pass on midi note data from midi input to midi output... hopefully that is configurable.
But still, MPC as slave didn't get me a solid sync either.
It's just a bit off, the MPC as slave, a bit behind, just too much to not notice it.

[Rastapoef]

Ok. Today I tested again. 

Having the MPC as slave is nonsense; there's noticeable latency which I can't correct. And I don't know how stable it is.

The other way round sounds much more tight. Even though the clock is visually jumping around a bit. Mostly in the decimals. Sometimes when I stop/start the bpm will show a fraction of the actual tempo shortly, like when playing in 110, show 38.3 or something like that. And then 
quickly jump to the right spot. 

These machines take some getting used to each other, but it starts to become workable if you know the quirks. Still have weird motifs if I stop / start the NDLR being slave without it receiving stop / start from master. At least, I think that's the trigger. Further investigation needed. But I didn't want to, because I was actually noodling for the first time and it was very enjoyable!

Ended up with a little noodle... I guess it's workable this way!
https://photos.app.goo.gl/SX4DDAa2ds8G8MfP9

Now I'll have to go and deepdive the manual.

[Darryl]

That was a cool little jam! We've done quite a bit of refining of the behavior of the parts when using external clock, I believe the manual is up to date for this topic. Check the section "External MIDI Clock & Stop/Start/Continue Messages" starting on page 33.

Let us know if anything still seems wonky after your experiments.