Audio edited on 29.97 fps timeline, how to move to 24 fps timeline

[Nicole Elmer]

Hi Dave,

We shot on a Canon 7D. The DP used 29.97 fps and the FCP timeline was set to this frame rate. The footage has all been transcoded to Apple ProRes (LT).

I used Philip Bloom’s tutorial from 2009 about conforming 30 fps footage to 24 fps. https://vimeo.com/4920433

It seems to work but I still have the audio issue. I notice that FCP inserts some blank spaces between some of the audio clips on the timeline, but it seems rather random where it does this. Sometimes the audio fits, sometimes it doesn’t.

When copying the 29.97 audio timeline to the 24fps timeline, several frames are dropped in the process, and in longer passages, as much as a minute is gone.

Thanks for any insight into this!

Nicole

Some contents or functionalities here are not available due to your cookie preferences!

This happens because the functionality/content marked as “Vimeo framework” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.

[Shawn Hyper]

As for timeline, audio has its own metric, that’s sampling rate. The video frame is used as a relative reference. That’s to say, as far as the sampling rate is not changed, the audio will keep its absolute position on the timeline, and will keep its absolute length also. Frame rate conversion is for video, not for audio. If video frame rate conversion is done correctly, the audio should be in sync. If not, maybe something wrong happened during the conversion.

[Michael Gissing]

Wow the Phillip Bloom workflow is comvoluted. basically what you need to do is take the 29.97 quicktime and go straight to Compressor. Set Compressor to convert the file to output as 23.98 with the same duration as the original and set the audio to pass through. You end up with exactly the same file conversion as his process without a whole lot of clumsy steps that basically lead to double handling and separates the audio for no good reason.

If you really want a decent result however, stay in 29.97 and finish your picture and sound. Make a final QT using current settings and take that to a facility with a decent hardware frame rate converter like a Teranex, Snell & Wilcox etc and in real time make a 24 frame master with audio passthrough. If you are going to DCP you want 24 fps not 23.98 and a hardware conversion will do the best at frame interpolation, maintain abolute timing and not affect audio at all. Simple, neatest and fastest.

[Buck Kahler]

I’m not positive this will work, but it might:
1. Export your edited audio from your 29.97 timeline at 48kHz
2. Bring that back into your 23.98 timeline

In theory it should sync up, but check it carefully. Let me know if this works.
-Buck

Buck Kahler
Oak Ridge, TN

[Nicole Elmer]

Thanks! I’ll try this workflow as well. This is a micro-budget project, so I hope the facility option won’t cost us that much. We’ll cross that bridge when we get there.

[Michael Gissing]

I strongly recommend you stay at 29.97 until the project in graded, onlined and sound mixed and do the frame rate conversion on the finished master. This also means you have a 29.97 version for broadcast.

I would never assume copy pasting a complex audio timeline from one frame rate to another will work.

[Nicole Elmer]

Thanks, everyone, for the input. It seems like I have lots of options and the best seems to be taking it to a facility that can do this, although the film will probably no longer be “micro-budget” at this point!

Buck: I tried your workflow suggestion and it worked on one clip that was about a minute and didn’t on a different clip. I did various short tests over the weekend and the audio issue is very inconsistent. Sometimes it is fine, other times it does things as strange as one track looping itself. Very very weird.

[Joseph Owens]

[Nicole Elmer] “So, after getting to rough cut on a feature, we recently discovered that our imbecile DP shot our film at 29.97 frames per second, not the 24 fps we thought we had.”

First: OMG. Somebody is getting away with calling a DP an imbecile? The other part is probably not news, though.
Second: Even when the communication is perfect, we don’t always know where a property is going to end up. Example: broadcast documentary, destined for NTSC, all shot 29.97 –so far everybody is doing their job.

Wait a minute, the European distributor wants a PAL version. The domestic distributor wants a theatrical (means 24 fps).

Now we’re in three frame rates, but this time for valid reasons. The only real solution is the consensus — finish it in the source format, then do the conversions. Hey, Blackmagic is selling the all-in-one Teranex now for a couple grand. You don’t need a facility if you are going to pay that much (probably more) for someone else to do it… And if you really have no budget but scads of time, export the master to compressor, use the “Better” conversion, and (a feature) might pop out a couple of days later. (I have always had to split long shows into 3 or 4 acts for this to work)… but it does, and I do this several times a year now, because lately everyone seems to need everything in every format, several versions, several languages.

Uhh.. I know its tempting to try the “Best” setting. Choose that and add a zero to the render time, whatever the “better” was, which is still massive. But the difference between “better” and “best” is much smaller than the difference between “good” and “better”. Not ten times, that is for sure. Budget about an hour/minute for “Better”, on an 8-core MacPro.

As far as audio sync problems across frame rates goes, though…. As stated, audio has a clock time based on a frequency standard, in this case 48 kHz. It will always be 48k. What FCP does, though is arbitrarily assign timecode to a .wav file (or whatever) that matches the default easy setup for your system. This causes even bigger problems than FCP itself has transcoding between frame rates and recalculating trim points based on 2:3 pulldown or sequence framebase changes (which it also does extremely poorly). You will get the easily predictable 1000/1001 difference if you go from say 29.97 (fractional) to 25, or even 24 (whole frames). You do have to reset the FCP Audio/Video preferences, exit, and restart.

jPo

“I always pass on free advice — its never of any use to me” Oscar Wilde.

[Nicole Elmer]

Thanks for the input. We appreciate it.