SQ audio/video sync

Forums Forums SQ Forums SQ general discussions SQ audio/video sync

This topic contains 12 replies, has 7 voices, and was last updated by Profile photo of SteffenR SteffenR 3 weeks, 1 day ago.

Viewing 13 posts - 1 through 13 (of 13 total)
  • Author
    Posts
  • #101533
    Profile photo of qball
    qball
    Participant

    Hello,

    I mostly do video work and just upgraded to the SQ5.
    So I’m good to go on everything on the console but just one question.
    Because no word clock (doesn’t really matter that much) I can’t lock to the masterclock of the video gear.
    What can I use to make sure my video/audio is synced?

    So I wanna make shure that sample 1 & frame 1 start at the same time and are synced even if I do a 2,5 hour recording. I want it to be frame accurate.
    What are my options (aside from buying the dante/madi expansion card and locking that way)?

    Thanks guy’s!

    #101535
    Profile photo of Hugo
    Hugo
    Participant

    Hi,

    If you need only an L-R, use an external audio/video recorder 🙂

    If you need multitrack, maybe with AES/EBU but how, that is the question :/

    To Follow

    #101537
    Profile photo of RS
    RS
    Participant

    To my knowledge wordclock would account for drift. In addition timecode and genlock (video) would account for frame sync. But it also depends if you go into post-pro or do live streaming.

    In live streaming I worked with blackmagic mixers, that accounted for the video delay automatically, if the audio ran thru the mixer. I haven’t measured the numbers, but it was “close enough for Rock n Roll”.
    In other situations I measured the delay with a app called “CatchinSync” and delayed the audio sent to the streaming comp within the audio console.
    But these methods would never account for drift, that could occur over time…

    So to me it all depends on what you use and what the end-result it used for.

    #101540
    Profile photo of Hugh
    Hugh
    Participant

    I use Atomos Ninja V video recorders. They offer a calibration tool to adjust the origination of the audio from the SQ5 to the camera’s location. The distance between a video camera and the origination of recorded audio is a differential that usually needs alignment.
    Hugh

    #101541
    Profile photo of Europe
    Europe
    Participant

    Your options are limited for accurate w/c sync to timecode without an optional card.

    A MADI Card ( or Dante card with a RME Digiface Dante ) with a Rosendahl Nanosync enables you to do this.

    If you don’t need multitrack then recording on the same recorder as the video is the easy option.

    If you need more than 2 tracks you could embed 4 or 8 channels of audio in the SDI stream using the little BlackMagic Audio to SDI boxes.

    #101542
    Profile photo of SteffenR
    SteffenR
    Participant

    So I wanna make shure that sample 1 & frame 1 start at the same time and are synced even if I do a 2,5 hour recording. I want it to be frame accurate.

    I think you mix up some things….

    You don’t have to do frame by frame synced recordings if you record audio and video on two different recorders…
    define a point in time that catch both recorders and generate a single event that is seen very clear and produces one spike in the audio
    Usually you can clap your hands…
    now you have a common point in time that enables you in post production to align audio and video

    the people on movie sets do this all the time

    the sync of sample rate and video clock is not needed if you use two recorders

    if you are using one video recording device with SDI and 16ch of audio, the device will sync with the other equipment automatically…
    or connects to a video sync clock, but again the SQ needs no video sync since it doesn’t carry video signals

    if you feed a Dante stream, the Dante stream carries a clock signal for sample syncing
    the AES output, MADI or Waves Soundgrid do the same
    If you have many video devices that need syncing it is likely that you are using more analog connections.

    #101550
    Profile photo of Europe
    Europe
    Participant

    Recordings synced with a clap or similar will drift over the time period the OP mentions. So a clap will be correct at that time point but timecode and wordclock will drift apart ( and upset your video edit house ), hence locking both together is the best way to ensure there’s no drift, this needs to be as the OP mentions – T/C frame 1 is synced with W/C sample 1. A&H have a news article on it from a well respected sound supervisor here: https://www.allen-heath.com/allen-heaths-sq-scores-big-tv-pub-quiz/ he uses a Ferrofish A32D to sync the Dante to a Rosendahl Nanosync, and the desk to the Dante clock of the Ferrofish.

    #101551
    Profile photo of volounteer
    volounteer
    Participant

    @Europe

    Video might be a bit harder, but I have run two audio recorders and found their sync did not vary. That was long ago in analog.
    Now I would sync the start with the sound board and let it run.
    Then edit the sound length by tweaking the end of sound to also match the end of video.
    What is in the middle will be better than lip sync.

    If you do a lot of recordings then get the clock device when editing for sync becomes too much work.

    #101562
    Profile photo of SteffenR
    SteffenR
    Participant

    syncing the sample rate is not the same thing as syncing timecode

    #101563
    Profile photo of SteffenR
    SteffenR
    Participant

    OK
    I was reading the news article carefully…

    I think what refers to is that he is recording a video file with audio in it… and will use this later to get mixed with the multitrack recording…

    Of course you can sync your audio recording with time code to the video gear, but this has to be done in the recorder…
    The SQ drive is not intended for this kind of recordings.
    So if you are using a multi track recorder(DAW), this has to get synced to the video time code, not the desk…

    #101599
    Profile photo of Europe
    Europe
    Participant

    You can stripe / record timecode on any two devices and they will be in sync enough for most shoots, just feed timecode to both devices as an audio track. So you can use the SQ drive to record multitrack for video, but it won’t be sample accurate as the OP asked ( and most shoots don’t require sample accurate sync so it’s not a problem ), but if you do require sample accurate sync over long periods then you need to lock wordclock and timecode to the same refererence ( Rosendahl NanosyncHD is your friend for this ), and you need someway to sync a console to the external wordclock ( Avantis has a wordclock input for example, but the SQ needs an option card to enable this through Dante or MADI ).

    In TV there are lots of potential clock sources ( video sync, timecode, wordclock ) and in a TV studio these are all referenced to one master so they don’t drift between each other, but on shoots it’s rare to have a master clock source and people use devices like Tentacle Syncs to have a common timecode reference, but this doesn’t tie all the sync sources ( video sync, timecode, wordclock ) together, it just provides a ‘good enough’ reference for each department to be running from the same time.

    #101600
    Profile photo of SteffenR
    SteffenR
    Participant

    you only need sample accurate sync if you are using different digital audio devices (or video devices with digital audio I/O) together

    syncing sample rates makes sure that each sample on all connected devices starts at the same point in time
    word clock drifts will lead to clicks and pops if you connect devices digitally together
    but if you record with one device, the drift is irrelevant (ok jitter could be an issue too)

    the same applies to the video part, every picture (frame) should start at the same point (synced)

    syncing to/with a timecode will ensure that all the content is aligned on the timeline to the same positions
    but it doesn’t provide any sample accurate sync, since a recorded timecode can get affected to sample clock drifts as well

    and if you take all your content into a NLE with audio editing the timecode will help you find the right positions in the audio and the video

    A clap at the beginning of a recorded part defines a position on that the audio event should start, synced with the video event…
    it makes corrections related to processing times of audio and video equipment or other sources for latencies easier.

    but all this is irrelevant to a live sound mixing desk like the SQ as a stand alone system.
    It only provides analog in and out in the first place… (ok 1 AES output)
    If you add digital I/O you can sync the desk with the cards… MADI, WSG or Dante

    Conclusion: you need to sync the wordclock when you connect devices with their digital interfaces, if you connect them with analog signals it’s irrelevant.

    #101601
    Profile photo of SteffenR
    SteffenR
    Participant

    oh… I forget

    a word-clock/house-sync/black-burst/genlock is just a rectangle signal with a fixed frequency

    a timecode is a real timestamp written to the recording/stream every now and then
    you can use a RTC as source for that

Viewing 13 posts - 1 through 13 (of 13 total)

You must be logged in to reply to this topic.