I’ve been thinking about this a lot lately, and it seems rather inefficient/inconvenient to use one pulse for every two steps; clearly some kind of tap tempo functionality is required for this to work. Is there any reason why a sync scheme like this would be better than say, having the sequencer advance one step per pulse?
This sync format was copied from the Volcas (not sure but maybe some other device already used that before). The only reason I can think of, and it is not that plausible, is to allow swing to be applied to even steps independently down the chain…
I always assumed it was to prevent the rate of sync pulses from getting too high.
Really? Nanoloop uses more pulses per step for its internal sync (I think 3 or 6?), and MIDI is even higher than that. I don’t see how ~30 year old devices can handle 24/48/96ppqn perfectly fine, but a PO-12 or Volca can’t handle 4ppqn?! Even more confusing, the Volca accepts MIDI in anyway…?
These aren’t internal clock ticks. These are electrical pulses that the designers have to make sure get transmitted correctly in whatever arrangement of interconnecting cables.
Say a device is trying to lock onto a 96ppq sync pulse. That’s almost 400 pulses per second. If any number of factors line up (length of cables, cable shielding, proximity to other signal cables, impedance of sync inputs and outputs, how many devices you’re trying to slave on one master, what voltages the master outputs and what voltage the slaves expect) you can easily drop a pulse here or there. And one pulse dropped at that rate is enough to be a noticeable skew.
MIDI does, in fact, support what is essentially a 24ppq sync pulse. (Midi clock). But MIDI has a designated, robust hardware specification—all inputs are optically isolated, cables are shielded, grounded at one end only to avoid ground loops and noise. Because it has to support much faster signals (31250 bits per second, iirc).
Now, back to the generic analog clock pulse. I’m not saying that 4ppq is too fast for the basic hardware involved. I’m just saying that there’s a limit to what is practical (since there’s not a global specification, it has to be designed to work in any generic configuration), and perhaps some engineer deemed 2ppq to be a good tradeoff of versatility and robustness.
I’m still not convinced. If I can sync two copies of nanoloop over a 20 year old game boy link cable at 12ppqn, and not lose sync even after an hour, surely 4ppqn would be fine?
This article asks that same question:
So strange. I do remember reading that article at some point. Guess I’ll just put it down to some weird unexplainable reason… unless someone from Korg can give an insight.
I’m almost certain that Korg chose this on the Monotribe because initially it only had 8 step sequences, later the V2 firmware it was expanded to 16 steps sequences but sync stayed at 8, why they never updated the sync code is a mystery, then the Volca range came out and they kept the 8 step sync scheme to allow compatibility with the Monotribe, the later Volcas and electribes can be set to 16 step sync though.
Ideally they would have updated the Monotribe to work with 16 step sync and used that going foreward with the Volca range, so my guess is that it was an oversight.
Ah, the plot unfolds!
I was wondering about @darenager’s idea before his post. My Monotribe is midified ,only the sync is double speed /half speed and seems all wrong. So hyper speed 8step seq on MT or half pace midi gear?
I’m convinced. didn’t even consider the monotribe when I made all my reasoning about reliability and all.
I asked Tatsuya Takahashi from Korg about it and as darenager says, the reason was backwards compatibility with the 8-step monotribe sequencer.
yuk, what a mess
One advantage to the 8 pulses per measure/bar is that it makes stopping and restarting slave devices a bit easier in a live jam setting because you do not have to be exactly hitting the restart on the 16th pulse to get a start on the 1. That is probably the only advantage to it though, so maybe that was why the Monotribe was never changed, and hence why it was rolled out that way to the Volcas initially, but of course the clock divide setting could have easily mitigated this, but perhaps they did not think of it at the time?