How many milliseconds is the MIDI Latency?

  • I'm Sorry to ask this question (for sure it has already been discuss previously) but I didn't find an answer yet..

    I was tring to set the DAW (cakewalk) to send MIDI changes to the KPA and I managed to do it, but I see there's a bit of delay (latency?) between the moment the DAW sends the MIDI change and the moment KPA actually changes rig, turns a stomp on/off or.. whatever.


    So I should have to anticipate a bit the moment when the daw sends the midi message but.. how much? how many ms?

    How could I calculate that time?


    Is it a fixed value or it could change time by time?

    or it could change time by time?


    thanks in advance for your help

  • Hey!


    I was facing the same latency issue when I did the first MIDI program change, but after some experimenting I realised it’s just the screen that is lagging. I always program my preset change 1/16th before the beat. It takes some time to get used to not having control over your sound. Try it for yourself and listen if the sound changes at the exact moment.


    Sorry for bad grammar, not a native speaker.

  • Hi Eric, thanks for your reply!

    me too, don't worry about bad grammar,

    I got you!


    well 1/16th seems to be a good compromise, probably an avarage good setting.. but it's obiouvsly related to the tempo of the song/session, but I can't see any relation with the latency time itself.


    You know what ?

    In this last two hours, while I was waiting a reply here, I've found a video on youtube about this and at the end he said that he shifted everyting around 80ms before..


    So I tried to measure by myself this short delay time: I've setted the Daw to change rig every measure, switching back and forth a Clean Sound and a Hi-gain sound, then I've recorded a clip and I've compared to the Click of the Metronome and I've discoverd this thing:

    When the sound changes from clean to Hi-gain, the Hi-gain actually starts aroud 165ms later.

    When it changes from Hi-gain to a Clean the Clean sound arrives 266ms later.

    (of course I used 2 rig where I've tunerd off any delay or reverb, so tails has been recorded)


    here some screenshot to better see the results




  • Generally I use a 16th for anything up to 140bpm and then a bit earlier with 3/32 for higher tempos, but that is live. The latency is in the patch changing more than the midi. Strangely enough, my old Rocktron Prophesy changes at lightning speed.


    One other thing to remember is that the Kemper sound will change well ahead of the name display. That is an intentional priority.

    Karl


    Kemper Rack OS 9.0.5 - Mac OS X 12.6.7

  • you can achieve a faster rig change when you're switching slots within a performance in performance mode, which is the intended "live" mode. audio wise this should happen in a few milliseconds.

  • you can achieve a faster rig change when you're switching slots within a performance in performance mode, which is the intended "live" mode. audio wise this should happen in a few milliseconds.

    I never understood why the same cannot be achieved in browser mode?

    Karl


    Kemper Rack OS 9.0.5 - Mac OS X 12.6.7

  • So you would have shorter change time if it is one of the 5 rigs in a performance. Is it a slower change time if you select a rig in another performance set?

    Karl


    Kemper Rack OS 9.0.5 - Mac OS X 12.6.7

  • you can achieve a faster rig change when you're switching slots within a performance in performance mode, which is the intended "live" mode. audio wise this should happen in a few milliseconds.

    Ah! That’s why I never had issues with the latency. 98% of the time I switch in a performance.


    Sollazzon What change did you do when you measured this?

  • I was switching Rigs in Browser mode.


    Repeated the same process in Preformance mode brings to a Very Faster reaction:

    around 62ms switching from clean to Hi-Gain (it was 167, so less than a half)

    around 47ms switching from Hi-Gain to Clean (it was 266..)



    here they are