Sync24 Patch - Bring External Sync24/DinSync/PPQN Clocks Into Audulus
  • I couldn't find a patch like this, so I patched this little module together to easily integrate my Beatstep Pro's standard Sync24 clock out with Audulus via an ES-8 interface. It should work with any standard Din24/Sync24 or Sync48 style 'pulse per quarter note interface'.


    The auto reset toggle determines what happens when the pulse train stops. When it's on (default) anytime incoming pulses stop the very next pulse will also be the first beat. When off, it assumes the pulses have been paused instead, and will maintain the current beat position.

    beat/2 and beat/4 represent 1/8th notes and 1/16th note beats respectively, derived from the same clock as the beat (1/4 note) clock.


    N

    EDIT:

    Updated the patch to remove the knob (now just a toggle), add some blinking lights, and fixed an edge case with auto reset. Edited the text above to reflect this.

    Sync24.audulus
    12K
    Screen Shot 2017-07-18 at 10.58.34 AM.png
    2510 x 1744 - 475K
  • Very nice. I was about to try and sync some patches with Pamela's New workout, so I will give your patch a try :)

    Some feedback, ther knob node has some issues with the man and min parameters. It's generally a good habit to keep the knobs from 0-1 and have some scaling inside the module. There even a module that will quantize the increments so now my copy has a integer increments of ppqn.

    Sync24ppqn.audulus
    27K
  • Good point - I was thinking about ditching the knob anyways and just making it a 24/48 toggle.
  • Even better!
    edit: The patch works like a charm!
  • Good to hear it, you definitely cleaned up the guts :)

    I fixed a few performance issues, and an edge case with the auto reset function that could cause self oscillation (!). I also added reset in and out, as well as some blinky lights and a hz display. I edited the initial post and attached in there.

    I also got the scope out and did some round trip latency measurements (ppqn from beatstep pro -> es-8-> audulus -> es-8 -> scope, and compared them with the beat step pro's own beat output from the same clock) - see the attached scope shot. Looks like regardless of the BPM, the trip through audulus and back out adds ~15ms as compared to the beat coming out of the beststep pro. (in the screenshot, the blue larger pulse is the beat step pro, and the smaller yellow pulse is coming from audulus via the ES-8) Time per division is 10ms so the yellow's leading edge lags the blue's by about 15ms. That's .015 of a second, which is fine for my use case, but if anyone needs lower round trip latency than that, it might be fun to build a PLL with prediction to cancel out the latency.

    EDIT: Note that after increasing es-8 input / output sample rate from 41khz to 96khz, latency was reduced to about 10ms (from about 15ms) - see second scope shot.
    IMG_1469.jpg
    3836 x 2354 - 2M
    NewFile0.png
    320 x 234 - 6K
  • Off topic, but I saw you mention:

    > I was about to try and sync some patches with Pamela's New workout, so I will give your patch a try :)

    I just got a PNW myself and that thing is awesome!
  • Very nice! My friend has a beatstep so I'll make sure to get him on this :)