extra latency using SPDIF

  • But I didn't see any explanation for why SPDIF is experiencing more latency and I'm wondering if someone from Kemper has said whether or not it's being looked into. I thought the extra latency through SPDIF was the issue at hand?


    Am I missing something?

    I'm feeling like I'm not understanding all of the thread and I'm very interested in using SPDIF instead of the main outs for recording but, I'd like to avoid using it if the latency is so much more when using SPDIF.

    From what I understood, the SPDIF signal is slower after having to make conversions to other sample rates. CK is looking at a bypass on this process for 44.1K use?

  • Am I missing something?

    Yes, you are. :)

    He explained it pretty well ... it's basically all down to the added feature of "sample rate conversion". You can select various sample rates since a while. The internal sample rate conversion applies linear phase (low pass) filtering which introduces latency.

    The goal is now to avoid going through this sample rate conversion process for the base sample rate (44.1kHz) to get rid of the added latency for this particular sample rate only.

  • Code
    1. Audio
    2. added: new effect Acoustic Simulator in category Equalizer
    3. added: new effects Phase Widener and Delay Widener in category Equalizer
    4. added: new effect Auto Swell in category Compressor
    5. added: Low Cut and High Cut filters in Output Section affecting all outputs added: Pitch, Voice Interval, and Key in pitch and pitch delay effects can now be morphed
    6. added: S/PDIF slave capability for classical PROFILER models manufactured since about 2019. New devices show option Auto/Internal in Output Section like Stage models do.
    7. improved: S/PDIF latency decreased (at 44.1 kHz)


    Haven't tested yet but looks promising

  • Yes, new Kemper Beta-Firmware 7.5 brings faster SPDIF Latency!



    I did a test (44,1kHz) with Constant Latency = ON:


    Main Out = 4.9ms (+217 samples)


    SPDIF = 4.33ms (+191 samples) -> old release = 8.3ms (366 samples)



    So, it’s no longer necessary to use the MainOuts to Audio interface for monitoring while recording SPDIF :-) (Maybe just if you want to record SPDIF DI mono/Wet mono and use stereo effects in the Kemper while monitoring…)



    My new Reaper offsets for aligning the recordings/reampings are:


    “Use audio reported latency” = Flagged


    Output manual offset: -16 samples (-> wet reamp signal arrives a bit faster – I don’t know why… ???) ckemper


    Input manual offset: + 191 samples



    All measurements with RME Fireface -> this device do not have any calculation errors reported from the driver to Reaper. So the Reaper offset values can be just the Kemper.



    Since the testing was done fast, I hope there are no errors… maybe someone can check and confirm what I figured out ;-)

  • Output manual offset: -16 samples (-> wet reamp signal arrives a bit faster – I don’t know why… ???

    ...

    All measurements with RME Fireface ...

    well, I'm not CK but I know this behavior and I know the answer.

    Like all Asio Drivers, RME has built in ADC and DAC conversion latency compensation into their drivers by providing the latency values to the DAW for compensation.

    These values are sample accurate for all the analog Ins and Outs - BUT you can only build in values for all Ins and Outs, regardless of each channel's input format.

    Since there is virtually no time consuming conversion happening on the SPDIF the input is faster than expected by the ASIO driver and the DAW.

    The same thing happens with ADAT and other digital I/Os.


    You can prove this by hooking up a cable from SPDIF Out to In and then recording from one track to the next within your DAW.

    Your audio will pretend to be time travelling with each pass.

  • I've just got some spdif cables, so was testing the current situation with this. (thru a Presonus Quantum 2626)


    At 48khz (Kemper as slave), I measured the spdif connection to have about 1.3ms more latency than the analog outs. So to be honest, not really a big problem. I'll probably stick with this, as 48khz is my preferred rate to work at.


    As mentioned by ckemper on page 3 of this thread, at 44.1khz - with Kemper as master - the spdif signal has slightly less latency than the analog outs - I measured around 0.3ms. But only with Kemper as master, if it's the slave, I get the same latency as at 48.


    Only odd thing I found was that at 44.1, if the kemper was set to slave, I got a lot of digital noise in the signal, pops/crackles etc. I double checked all settings - 44.1 definitely also set in audio interface, and cubase project. And I get no digital noise at 48 with kemper as slave. Not really a problem, as don't need to use this setting, but odd...

  • tbh i don´t care if its 3-4ms or 6ms with SPDIF because its superior sound quality.

    besides that you´d have to correct it later on in either ways.


    which brings me to my question:


    how can i manage to synchronize my Bass-guitar DI-track with the

    later on re-amped kemper track?

    (Cubase 5; ASIO FireFace 800; kemper with SPDIF and constant latency on;

    latest Kemper Firmeware)


    i don´t have any clue how i professionally should do that.


    thanks guys :)

  • Just nudge the reamped track backwards visually where there is a strong transient. I doubt most guitarists or bass players play so accurately that they are less than a few milliseconds off anyway.

  • If you get phase problem, just flip the phase on one of the tracks.