Cinema Tools: Reverse TK of 29.97 DVCAM transfer of 23.98 HDCAM original

[Nayeli Garci-crespo]

Gary: “As long as there is no video associated to the audio clip on import to the timeline the time of the clip remains unchanged.”

But isn’t that, in Final Cut Pro, determined by the sequence preset, so that in a way, you can’t really have audio that’s unassociated to video? For example, I opened a project that’s set to 23.98, and I imported sound at 48K. When I put that sound in a 23.98 timeline with no video in it, Final Cut slows it down .1%. It will have a green bar across the clip to indicate it….

-Nayeli

[Gary Adcock]

[Nayeli Garci-Crespo] “But isn’t that, in Final Cut Pro, determined by the sequence preset, so that in a way, you can’t really have audio that’s unassociated to video?”

Its called an Aiff file-

if you are bringing in audio that was captured from a camera is different than DSS audio captured separately, BWF files that import from most audio recorders are tagged with TC if you remove that format condition the files import with out modification.

try saving as an Aiff file and then re-importing into fcp.

gary adcock
Studio37
HD & Film Consultation
Post and Production Workflows

Inside look at the IoHD
https://library.creativecow.net/articles/adcock_gary/AJAIOHD.php

[Nayeli Garci-crespo]

I’m actually now more confused than ever!

I did a conform to the 23.98 video and it’s the same, so it should be 23.98. What also confirms this is that I imported the 29.97 video and the original 24fps/48kHz audio into a regular DV NTSC 29.97 project with 29.97 sequence preset, and it behaves exactly the same as with the 23.98 video. The WAV files that come direct from the Sound Devices 744T have pulldown applied to them by Final Cut (green bar), so they are .1% slower than the video, but that same file converted to AIFF with no pullup or pulldown fits the video perfectly since Final Cut does not apply pulldown (no green bar across sound clip). So it’s definitely not Cinema Tools. Could it be the transfer from HDCAM to DVCAM at the lab? I just can’t think why audio with no pulldown will fit the 23.98 video!

A third test I did was to reverse the 29.97 DVCAM transfer to 24fps, and import the audio witha 24fps preset. In that case, I was really surprised to find that Final Cut did not apply pulldown to the original WAV file (as expected) but DID to all the AIF files I imported (or pullup)! The WAV audio in this case was 1% too long (which makes sense in relation to the other tests). The file that worked was the original file converted to AIF, because Final Cut applied a pullup to it (green bar across clip, and I can tell it’s a pullup instead of a pulldown because it runs faster).

*sigh*

-Nayeli

[Steven Gonzales]

There was another post a while ago, that suggested the settings of the project when you bring in sound has an effect on the timing. I don’t know if it’s true, but here’s the link:

https://forums.creativecow.net/readpost/8/1006953

[Oliver Peters]

[Nayeli Garci-Crespo] “Could it be the transfer from HDCAM to DVCAM at the lab? I just can’t think why audio with no pulldown will fit the 23.98 video! “

You should confirm whether the HDCAM was really shot at 23.98 and not a true 24fps. This can be done with F900 cameras. If so, depending on the playback of the HDCAM deck at the lab, it is possible that the HDCAM tapes at 24fps (maybe) were “corrected” to 29.97 (therefore 23.98) in the dub to DVCAM.

– Oliver

Oliver Peters Post Production Services, LLC
Orlando, FL
http://www.oliverpeters.com

[Nayeli Garci-crespo]

Thanks, and it is true!

-Nayeli

[Nayeli Garci-crespo]

“try saving as an Aiff file and then re-importing into fcp”

I did that, and it’s what synchronizes with the video, because Final Cut doesn’t alter the speed of the file.

Do AIFF files have no timecode information by nature?

-Nayeli

[Nayeli Garci-crespo]

“You should confirm whether the HDCAM was really shot at 23.98 and not a true 24fps. This can be done with F900 cameras. If so, depending on the playback of the HDCAM deck at the lab, it is possible that the HDCAM tapes at 24fps (maybe) were “corrected” to 29.97 (therefore 23.98) in the dub to DVCAM”

That’s one of the first things I checked. The F900 is set to 23.98 for sure, at least that is what the camera operator assures me. And they even told me at the lab that if it was set to 24fps, given the equipment they have, they would need to upload it to their eQ to do a conversion to 29.97.

Now here’s a question that may seem stupid. I’m new to video, as far as capturing goes, and I’d never asked myself this:

When a video camera running at 29.97 or 23.9 record an image that when played back lasts .1% longer than the event did in real time? Or is this just something that happens with a telecine of 24fps film? (Or I suppose conversion from true 24fps video as well.)

-Nayeli

[Steven Gonzales]

Video shot and played back at the same frame rate runs the same time as the original.

When film is shot at 24 fps, and transferred to video, it is slowed down by .1% so that it can evenly fit video at 29.97 or 23.98. So the film is played on the telecine machine at 23.976 frames per second, (slowed down by 1 part in 1000) when transferred to video.

When put in an edit system to be synced with dual system (separate) sound, there are two choices:

1) keep the film_telecined_to_video at 29.97 or 23.98, and slow the sound down by the same .1%. A couple ways that sound is slowed down: when recorded, the recorder’s internal clock is used — but on playback a video signal is used as the clock. Another way: the sound is recorded at 48,048 samples per second, but played back out at 48,000. Timecode is included as an addressing mechanism, but sometimes timecode settings affect sample rate. Sample rate and clock are what determines how fast audio goes by.

Or 2) you can speed the film_telecined_to_video back up, by playing at 24 fps, and just bring the sound in at it’s original sample rate with its original clock.

[Nayeli Garci-crespo]

Thanks a bunch Steve! That actually solves my mystery. I’ve got tons to learn about sound and video, thanks for helping me along!

-Nayeli