camera audio & DAT audio out of sync, drifts

[Bouncing Account needs new email address]

DF and NDF have nothing whatever to do with sync or timing.

It is simply a numbering system.

The same number of frames are recorded in either TC mode.

I’m surprised that there is any drift at all.

I have no problems with sync-slippage when using digital audio sources and digital video tapes. None.

How are you capturing the DAT tapes?

[Joanie Spina]

I am not capturing the DAT tapes. The sound man was doing that and transferring to a CD after mixing. They are given to me as audio files from Pro tools or on a CD with .aif files

[Bouncing Account needs new email address]

Well, without knowing anything about your setup or his setup. I’d suspect that the problem could be coming in at some point in his transfer process.

Your original question was about DAT audio.

Now there’s an extra “layer” of transfer and sampling conversion in going to the CD.

If the audio tech probably recorded the DAT at 48 kHz sampling, but a CD is actually 44 kHz sampling.

That alone should not be a problem, but he MUST use the proper conversion to burn a CD from the DAT files (I have no way of knowing what he’s doing from here.

OTOH, you didn’t tell us what kind of cameras and format you are shooting.
Or how you are capturing the video.

DETAILS are ESSENTIAL for better response to your problem.

[Bouncing Account needs new email address]

My 44 kHz CD reply was based on him burning an “AUDIO CD” (one that will play on any standard CD player).

OR, is he burning a “DATA” CD with the raw DAT files recorded on it, that you then drag onto your HD and import to FCP?

How are you getting the CD audio into FCP?

[Joanie Spina]

The aif files are on an audio CD, 44.1
I shot on standard dv on a sony pd150, 48 khz samling rate, 16 bit

[Michael Gissing]

Dat to aif, burned to a CDRom doesn’t require a sample rate change from 48khz to 44.1khz. So if that is happening, it is unnecessary.

There is a slight drift between DAT 48khz and 29.97 frames rates. Tell your sound man to sample rate convert the original files from 48khz to 47.952khz to correct this drift. It is caused by the difference between 29.97 and 30 frames per second. On short grabs, you won’t notice. Anything longer than a minute and you will have drift.

You might be able to use quicktime conversion to change the aifs to 47.952. I haven’t tried this as I do all audio processing on other systems, but it should correct the drift.

[Bouncing Account needs new email address]

[Michael G] “Dat to aif, burned to a CDRom doesn’t require a sample rate change from 48khz to 44.1khz. So if that is happening, it is unnecessary.”

Yes, I added that to another post.

Her new reply, states (as I first suspected, that the DAT audio is being converted to an audio CD.

Since that’s the case, if the audio guy is correctly making the transfer, there should be NO drift.

—————
The DAT is not locked to the camera… its “locked” to TIME (real-world), as is the camcorder.

Therefore:
32 minutes, 22 seconds of VIDEO is the same length as 32 minutes, 22 seconds of AUDIO… IF there is not some sort of transfer error.
—————
There’s got to be something the audio guy is doing wrong in the transfer to the CD.

[Joanie Spina]

I sent the audio people your thoughts and requested they try adjusting audio as you suggested. Thank you very much. I’ll let you know the outcome.

[Joanie Spina]

I tried exporting the CD audio track at 47.952 and reimporting, but it still drifts, It was 44.1 to start with

[Bouncing Account needs new email address]

Be careful.

I’m not so sure that’s the specific problem.

If they DO that, they WILL be changing the PLAYBACK length of the DAT audio.
It just MIGHT work, BTW, but only IF they have been doing something otherwise WRONG in the previous transfers.

Consider this:

The camera is rolling, pointed at a clock with a sweep second-hand and a ticking pendulum.

There is a DAT deck recording the audio of this independently of the video.

The camera and the DAT deck both record for exactly 58 minutes and 32 seconds.

What you WANT is the DAT files to play at exactly the same speed it was recorded (58 minutes and 32 seconds)… not “adjusted” to some different time (frame-rate).

Likewise, you want the video file to play for exactly 58 minutes and 32 seconds.

If the audio gets transfered at some OTHER rate (as I assume it has been in the PAST during the transfer to CD) the length of the audio will be wrong compared to “real-time” (the clock, in this case) and the sync will get farther off as it plays against the video.