Re: music video - pull up with 744


On Wed, 9 Aug 2006 12:30:20 -0400, Courtney Goodin wrote

... When we are talking about playback of audio tracks it really has nothing to do with Time Code generators or the computer's internal real-time clock. The D to A converters usually have a dedicated crystal and their sample clocks determine the accuracy of the playback.

Well if we're talking about most gear, the ADC are the ruling clock, not the DAC. But since they're often on the same card, that's moot. What is critical is that in a pro situation, the ADC's dedicated crystal is usually ignored. Instead the entire audio system is locked either to input, to external word clock, or to external video blackburst. Since other gear and video playback shares the same reference, accuracy is a relative term:

- everybody marches to the same drummer, so once sync is established, things stay in sync for the duration of that pass. (Assuming, of course, that editorial hadn't already destroyed sync... but that's another issue.)

- there is no guarantee that studio's 'accuracy' matches any others, unless you're deriving house sync from GPS or similar standard clocks. In other words, if two dub theaters start playing the same film at exactly the same microsecond, a couple of hours later the two theaters could be seconds or more out of sync with each other. But within a single theater, all the dubbers / VTRs / DAWs etc will stay in sync with each other.

On 2006-08-13 11:46:31 -0400, Ty Ford <tyreeford@xxxxxxxxxxx> said:...

The little I know is that I once recorded two tracks of a concert to my laptop and two more tracks to DAT. It took about 10 seconds of playback for them to noticeably drift out of sync....

Bad luck. I did the calculations based on moderate priced crystals, a few years ago, and came up with 1 frame drift in 40 minutes per device. That's assuming constant equal temperatures, well designed software/circuits, and no physical shocks. Since two devices can drift in opposite directions, the net error could be 1 frame in 20 minutes.

I've had correspondents report noticeable sync errors between miniDV and standalone CD recorder after 6 minutes. And I had a very early DAW (1989 design, analog-only inputs) drift from my house video after about 4 minutes. But ten seconds?

Or by 'noticeably', are you referring to a flange sound rather than a lipsync error? That would happen after ~ 7 ms, which is roughly 1/4 video frame, which is still 1 frame/40 seconds and pretty bad for two crystal-controlled devices.




--
Jay Rose CAS
tutorials and other sound goodies at dplay.com
email is "jay@" plus the dot-com in the previous line.

.

Relevant Pages

  • Re: How play 2 videos synchronized
    ... clock in both (get the clock from the first and set it in ... i've got a clock from first video and set it to ... but ive noticed that sometimes the videos come out of sync. ... I've noticed it adding a custom transform filter before video render ...
    (microsoft.public.win32.programmer.directx.video)
  • Re: music video - pull up with 744
    ... Well if we're talking about most gear, the ADC are the ruling clock, ... things stay in sync for the duration of that pass. ... I did the calculations based on moderate priced crystals, ... and came up with 1 frame drift in 40 minutes per device. ...
    (rec.arts.movies.production.sound)
  • Re: Error Code 13: Guide Info Will Not Download
    ... austin, tx ... Also I had the time sync issue resolved yesterday, ... The internet connection and firewall can be ruled out as suspects. ... internet connectivity and system clock can be ...
    (microsoft.public.windows.mediacenter)
  • Re: WTF with my computer clock?
    ... You may need to build a real LF antenna, ... ethernet port on one end for NTP sync and a 60kHz transmitter to sync a ... clock on the other. ...
    (sci.electronics.repair)
  • Re: Why SR? (was: Source Independency of Light Speed Without an Aether???????)
    ... >>> If the coordinate system is constructed using light ... > Well you could defer to the sync of another frame. ... And if you DID adjsut the tick rate (which is not really what is covered by clock sync) .. ... Would comparing wavelengh of light in each direction and seeing that they are the same, show isotropy of light, without and clock being required ?? ...
    (sci.physics.relativity)