Midi Clock Offset?


Can the midipal be used to offset (backward or forward) a midi clock ?

No, I have not programmed this function; and since this product is no longer in production, it is unlikely to happen.

Just because I’m curious: how would you implement a technical device that underlies our known laws of physics to offset an incoming signal backwards in time?

FLUX algorithm, combined with wormhole energy !

Every time the offset parameter is increased, you send an extra clock pulse. Every time it is decremented, you “swallow” an incoming clock pulse.

Ah, so its more a one time “drop/insert” thing not a real “time shift” thing……

It’s enough to keep the machine downstream running one tick ahead/before in time.

Too bad… this would very useful since so many drum machines have sloppy timing and/or latency issues. I think inner clock systems sync-shift would do but it’s too much money for this reason alone.

Do you think it’s possible to create a simple device that would delay the midi signal ? In my case (korg esx latency) this would be enough.

Accurately delaying the MIDI data is more complicated than simply stripping or inserting clock messages.

I was thinking about delaying the whole electric signal (much like a delay line works for an audio signal). That wouldn’t cope well with the numeric nature of midi ?

Well, analogue delays are much more complicated things to build than having a simple processor decode the signal and do its thing :slight_smile:

> since so many drum machines have sloppy timing and/or latency issues

Can you give some details on what problem with what setup you’re exactly experiencing?

t2k> ok, say an octatrack triggering a drum kick and a korg esx triggering the same drum kick. Esx is slave or master, doesn’t matter. Midi cable is short( 30cm). Layer the two kicks: you get an very noticeable delay between the two. Not the kind of delay thath gives you a nice phasing effect but the kind of delay that allows you to hear 2 distinctive sounds. A problem for any music that needs precision in attacks.

Are the two machines running their own internal sequencer, sync’ed by MIDI?

Or is the octatrack actually triggering the ESX?

Each machine runs its own sequencer. This could be an audio latency issue, I’ll have to try to trigger the esx from the OT.

An hint: I trigger the shruti from the OT and I don’t get any latency issue, but I don’t use the shruti for drum sounds (bass & stuff)

1. You’ll get better results triggering the ESX from the OT’s MIDI tracks. Both the OT and ESX maintain their own internal clock by averaging MIDI clock messages instead of progressing their sequencer based on incoming MIDI clock messages.

2. Don’t do this. Layering kicks, bass, or any other stuff with percussive envelopes or similarly tuned low-frequency sounds simply doesn’t work well over MIDI. Do you layering inside the same box, or sample a layered hit you’re happy with.

thx t2k for your answer.

1/ probably, but the sequencer itself is probably 75% of the interest of the esx so if I can’t use I might as well turn the whole unit off. Well, maybe not as it would still allow me to extend the number of samples over the OT :slight_smile:

2/ the delay is such that even if I would have all of the drums on the esx and all of the synths on the OT this would still be a problem. Some basslines need precision too !

A syncshift mk2 is just $520, I guess I’d better start to save now ^^

Can you post a video showing the delay issue?

In which way the syncshift mk2 would solve your problem?