Page 1 of 3 123 LastLast
Results 1 to 10 of 25

Thread: Theory question - Audio sync

  1. #1

    Default Theory question - Audio sync

    Please, can anyone point me to any information about the level of accuracy required when syncing of Audio (non-verbal) and Video; and if there are any 'well known' rules about allowing either Audio or Video to occur out of sync.

    I may have devised a way to convert the events in a Midifile into keyframe markers in After Effects. This would make it easy to, for example, to display the letter "B" briefly whenever a bass drum gets played.

    Video has a slower sample rate than MIDI, so one can only sync the event to the nearest video frame.

    I seem to recall (decades ago) being advised that Audio changes should precede Video changes, by a few frames. Is there any truth in this? I had assumed it either because human brains prefer it like that, or perhaps it was a rule when using very old equipment.

    My apologies if I have posted in the incorrect forum.
    Thank you.

  2. #2
    Join Date
    Oct 2008
    Location
    ENGLAND!
    Posts
    774

    Default

    This is a very interesting question and there are lots of different ways of seeing it.

    When i have to sync up audio i do try and sync the audio and video up as perfectly as i can but this may sometimes be almost impossible to do accuratly due to the large sample size of video like you said. then comes the question as to where the audfio should be followed byu the video or not. tbh i don't think there is that much of a difference. nothing that is majorly noticeable will result from the audio starting 1/15th of a seccond before you can visually see something making the audio, not in my opionion anyway.

    When doing my multicamera shoots because each camera is independant, i sync these up by the audio and by looking at the soundwaves you can physically see that you cannot perfectly get the audio to all start at exactly the same time, so lets say on one camera, you have a lav on a presenter and another or a different presenter and a 3rd for ambient, (i dunno why i do this, i think it makes it a bit more naturaly becauser you get the rooms accoustics and stuff mixed in). when i line these up in premiere pro, there is no chance of me ever getting all of the audio to be 100% in sync with each other.they will all always be out of time with each other but in the final edit you will not notice.

    recently i was editing another 3 camera shoot and every was actually perfectly in sync (pure luck with pressing the record button on a frame exactly etc. to put it simply, it will never happen again). we had a clapper board in the center, it "clapped" and i synced it all to that and someone in the crew coughed and he was closer to one mic than an other and you can see the cough come up on one mic before the other showing a delay but not one that it great enough to notice. now here's another question, can the huiman brain specifically pinpoint if something is out of synce or not or will it simple thing "ahh, it's just a little bit far away"?

    I've probably ended up giving you more questions but hopefully i've answered the orriginal.

  3. #3
    Join Date
    Feb 2006
    Location
    Surrey, UK
    Posts
    10,842
    Blog Entries
    1

    Default

    Quote Originally Posted by TimAndrews View Post
    I seem to recall (decades ago) being advised that Audio changes should precede Video changes, by a few frames. Is there any truth in this? I had assumed it either because human brains prefer it like that, or perhaps it was a rule when using very old equipment.
    I have a hunch that you may be confusing this with the oft used technique of using "J" cuts between two different scenes whereby the sound of the second scene is introduced whilst the video of the first scene still plays for a few frames. The transition seems smoother as the fresh audio prepares the way for the new setting.

    I don't watch it now, but I seem to recall EastEnders using anything up to a second of audio from the following scene before cutting to the video.
    Tim

  4. #4

    Default

    Thank you for your responses.

    Smifis's practical experience suggests the effect is minimal. And, even if it were noticeable; the maximum difference would never exceed the time for one frame.

    Incidentally, I don't think Humans naturally convert a lagged audio to mean that the source is distant. With speech, for example, humans won't see the person's lips move if they are a long way away. On the other hand, With distant thunderstorms, we are taught as children, that the thunder occurs after the lightening.

    However, Humans are very good at estimating distances when comparing a nearby audio source to a delayed 'echo' from a distant surface, or 'reverberation' within a space.

    My maths may are not precise, but within the 40ms (40 thousandths of a second) it may take between 2 video frames; audio may have travelled around 40 feet.


    May I also thank TimStannard. It will be obvious from my post that I am rather new to Cinematography. I am kicking myself for not reading about L and J cuts before. But I am wiser now!

    Once again. Thank you for your assistance.

    I shall now go away are prepare a new thread.

  5. #5
    Join Date
    Oct 2008
    Location
    ENGLAND!
    Posts
    774

    Default

    Well the speed of sound at sea level = 340.29 m / s
    so 340.29/1000=0.34029meters in on millisecond.

    one frame being on 25th of a second (PAL) it would be 340.29/25=13.6116m in one frame of video

  6. #6
    Join Date
    Oct 2008
    Location
    ENGLAND!
    Posts
    774

    Default

    Where was i going with that?

    lets say the speed of sound is 340.29m/s.

    Lets say someone claps 34.029 meters away and we can hear them, it will take one tenth of a second for the sound to reach us, and one tenth of a second is 2.5 frames (PAL) so if we zoom into the subject, it will look out of sync because the sound will be one tenth of a second "late" but would we actually notice it? you would notice perhaps half a second but we would need to be 170.145 meters away and can we really see that far in detail? and can a camera zoom in that far and would someone want to zoom in that far?

  7. #7

    Default

    I suspect that humans are better at recognising 'out of sync' when viewing familiar objects, such as a human's face and voice. I sometimes detect the delay when seeing drummers (as in a rock band) when the camera is at the back of the hall. Although readers of this forum probably have increased sensitivity, I am curious over how much delay (out-of-sync interval) is possible before it becomes unacceceptably noticeable.

  8. #8
    Join Date
    Jan 2005
    Location
    Warsop, Nottinghamshire.
    Posts
    4,465
    Blog Entries
    2

    Default

    I often do multi camera shoots in theatres and so on. There is a wild difference between the sound recorded on cameras at the back of the auditorium as to those directly on stage.

    I will initially sync to a noticeable sound - a hit on a hi-hat or similar - and use an on stage camera to sync to feed from the sound desk. But inevitably the vt from the back cameras will be out. And it is noticeable. I can see the drummer hitting drums - the vt lags and has to be moved several frames at least. When we filmed at Liverpool Arena, back cameras were getting on for a second out of time.

  9. #9
    Join Date
    Feb 2006
    Location
    Surrey, UK
    Posts
    10,842
    Blog Entries
    1

    Default

    Quote Originally Posted by Andy Lockwood View Post
    I often do multi camera shoots in theatres and so on. There is a wild difference between the sound recorded on cameras at the back of the auditorium as to those directly on stage.
    Interesting, Andy. Something I've never thought about.

    Would it not be more acceptable/realistic to keep the delay in shots from the back of the arena? After all that's what we would experience in life.

    Maybe as you zoom in, you would need to reduce the lag?
    Tim

  10. #10
    Join Date
    Oct 2008
    Location
    ENGLAND!
    Posts
    774

    Default

    Quote Originally Posted by TimStannard View Post
    Would it not be more acceptable/realistic to keep the delay in shots from the back of the arena?
    a delay could perhaps be seen as strange to the audience if they have a big tv. lets say you're in a theater and you watching from the back of the venue, you wouldn't see an out-of-sync "image". we just see as if it was 5 feet away. i think it's because we can't see things from a distance in much detail so we wouldn't notice a lag. but if we use a camera and zoom our eyes can see more in detail, especially on a big screen and because we wouldn't normally notice a delay in the sound, this may come across as if the audio is out of sync when actually fact it is technically in sync.

Page 1 of 3 123 LastLast

Similar Threads

  1. Audio out of sync
    By Hobby Mum in forum Pinnacle Studio, Edition including Avid Xpress and Liquid
    Replies: 8
    Last Post: 04-07-2008, 09:55 AM
  2. Replies: 0
    Last Post: 02-13-2007, 12:51 PM
  3. audio out of sync
    By odroller in forum General video editing software help and advice
    Replies: 0
    Last Post: 06-09-2006, 01:15 AM
  4. Audio not in Sync
    By eiyosus1 in forum General video editing software help and advice
    Replies: 7
    Last Post: 12-21-2005, 09:35 AM
  5. audio sync
    By andrew in forum General video editing software help and advice
    Replies: 3
    Last Post: 03-26-2005, 03:51 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •