editing 30fps to 24fps?

My DP made a mistake in final weekend's shoot and shot with 30 fps instead of 24fps. Now my editor is working on how to transcode them back to 24fps without losing anything or minimum loss

Any recommendations of the best way doing it?

Thanks so much
 
Well you can go about it two ways as far as i know. 1 Convert all your rushes from 30 to 24 or 2 edit the whole thing as 30fps, export uncompressed as 30 then convert to 24. Or i guess you could accept the mistake and just keep it 30 as most wont notice that much difference ;)
 
Your issue here is going to be you'll need to slow down your 30fps footage to get it at 24, or speed up your 24 footage to get it to 30. Either way you're going to have issues with your audio.
 
You're probably making a fuss out of nothing. Premiere has no problem using 30fps footage in a 24fps timeline. It just drops the frames it doesn't need. No transcoding needed. Just drop it in the timeline like any other footage.

And though I don't know, I was kind of under the impression that Premiere isn't the only software that does this. What software are you using? It probably handles different frame-rates just fine.
 
FCP 7

so 15 days footage of 24 fps mixed with 3 day footage of 24 fps is not noticeable? I doubt that.

Sound is recorded separately, not sure how much input that's gonna be
 
Well, stop doubting. Because I use 60fps footage in 24fps timelines, with great frequency. It only uses the frames you need. You can't notice. You're fine.

Again, I'm not a FCP guy, but I have to assume that it's capabilities, in this regard, are the same as Premiere. If this were Premiere, I could tell you, with absolute certainty, that there is no problem. And though I don't know about FCP, my best guess is that it's the same.
 
Well, as I say, you'd have to slow down the 30fps footage to fit into the 24fps timeline. Your sound is going to be your issue - if it was recorded at 30fps, you might find slowing it down that slight amount won't have too much of a noticeable difference. If it's recorded at 24fps, that's when you'll start to encounter issues.

I'm not sure whether FCP does this for you automatically or not, you may have to conform in Cinema Tools before sticking it into your timeline. But hey, why can't you just give it a go and see what happens...?
 
Well, as I say, you'd have to slow down the 30fps footage to fit into the 24fps timeline. Your sound is going to be your issue - if it was recorded at 30fps, you might find slowing it down that slight amount won't have too much of a noticeable difference. If it's recorded at 24fps, that's when you'll start to encounter issues.

No slowing down is needed. Not in Premiere, anyway. Again, I don't know about FCP, but like you said -- just try it out.

Aside from the stuff that is slo-mo, can you tell which of this footage was shot at which frame rates?

https://vimeo.com/39110568
 
Drop frames, blend frames or spend a lot of render time in After Effects with Timewarp with pixelmotion setting.
It will interpolate the 24 frames from the 30 frames.
Technically speaking AE has to distill 4 frames from 5 frames. The first one is always correct, the other 3 need calculation. I believe this can also be done with Twixtor.
Make sure you don't get artifacts like strange edges on hair that moves in the wind.

With dropping frames motion cause look strange as every 5th frame gets tossed away.
With frameblending you may loose sharpness.
 
No matter what you do, you're going to have issues with your audio.

Also, I"m pretty sure I meant speed up your 30fps footage to conform with the 24fps..

Take it into your local post hosue and get them to do it with Arri Relativity. It'll look great and work well.
Dropping frames and blending frames can cause many issues with things looking odd.
 
I feel like I'm taking crazy pills.

Audio has absolutely nothing to do with frame rate. Why are we even talking about that?

Dropped frames will look weird? According to whom?

Which of the following footage was shot at 24/30/60? You've got a 1-in-3 guess, but most people would be lying if they said that they could tell the difference:

https://vimeo.com/46006306
Password: indietalk

Unless FCP is really shitty, and can't do the basic stuff that Premiere can do, there is no problem here. No need to have any footage converted by anyone. Just put it in the damn timeline, you'll be fine.
 
Of course it does. If you record your audio at 30fps, your audio is going to have to be slowed/sped to conform to a 24fps timeline.

That said, I'm relatively sure FCP (and Premiere, and Avid) automatically conforms the frame rate to your sequence, so it should do an automatic conversion for you.

If you want to look at the difference one frame makes, pick up a copy of 'The Cutting Edge: The Magic of Movie Editing' and have a look at the part where James Cameron talks about Terminator 2 was too long, and so to try and fix it he decided to try cutting one single frame out of every 24. Check out the results, you might be surprised the difference once frame makes ;)
 
Of course it does. If you record your audio at 30fps, your audio is going to have to be slowed/sped to conform to a 24fps timeline.

That said, I'm relatively sure FCP (and Premiere, and Avid) automatically conforms the frame rate to your sequence, so it should do an automatic conversion for you.

If you want to look at the difference one frame makes, pick up a copy of 'The Cutting Edge: The Magic of Movie Editing' and have a look at the part where James Cameron talks about Terminator 2 was too long, and so to try and fix it he decided to try cutting one single frame out of every 24. Check out the results, you might be surprised the difference once frame makes ;)

Dude, I'm an editor. I know how important a single frame is. But that's not what we're talking about. We're talking about whether or not an audience can tell the difference between 30fps footage edited in a 24fps timeline, vs. that which is 100% 24fps. And the answer is no. 30fps, edited to 24fps is just fine. No problems here, nothing to worry about.

And no, there's nothing funky going on with the audio. The audio has no frame rate. As it is embedded in the video, it will be "conformed" along with the video, but in doing so, it stays in it's natural state. The audio changes zero when moving between frame rates.

All is well. Nothing to worry about. Just edit the movie. :)
 
He said himself the sound is recorded seperately, so of course there's going to be considerations. It is not 'embedded' at all. If the seperate sound was recorded at a different frame rate to what the video is in - it can cause issues. I can tell you a DP friend of mine shot a film last year, where she had the camera set to 24, and the recorder set to 25. The DP and sound guy didn't confirm frame rates, and so they had a lot of issues trying to keep the audio in sync.

Every time I've ever worked with a sound guy, they ask me what frame rate we're recording - I somehow don't think they're doing this just for fun.

To get 24fps out of a 30fps clip, you need a pulldown. Whether or not that pulldown is done in the import in editing, I'm not 100% sure. But, it's not simply a matter of 'there is no difference, just do it' - there's a pulldown that needs to be applied somewhere, and it's better if you know where the pulldown is happening so that you can troubleshoot any issues. Your editing system may cut every 4th frame (or whatever), or it may speed it up by the tiniest bit, or you may need to conform in Cinema Tools. I can't advise 100% on specific editing tasks, but I can tell you that there's going to be some conversion somewhere, whether it's done manually or automatically. You should know so that you can adjust your audio accordingly.

If there was no issue here, then shooting in 30fps or 1000fps technically shouldn't have made any difference.

I'm not saying that you're necessarily going to have to proactively go through and make sure there's a pulldown happening, but test it out on your editor and see what it does. It may or may not work, I don't have FCP in front of me, and have never shot 30fps in my life.
 
Last edited:
The audio guy only needs to know this stuff for syncing purposes. And they won't be syncing to the raw footage, but to the edited movie. So, all is well.

And I never said there is no difference. I'm just saying, quite factually, that nobody can tell the difference. Did you watch the video I posted? Can you tell the difference?

And no, there is no pulldown involved here. Pulldown is when you add frames. This is the opposite. Call it a "pullup". Every fifth frame will be ignored, and nobody will notice.

For all intents and purposes -- there is no difference, just do it. I do it all the damn time.
 
Hmm...
I ain't no professional editor so I can't comment anymore than I already have.

I'm also not really sure why you'd be syncing your audio after your picture lock. How would one sync the audio after the clapperboard and clap have been stripped out of the edit..?
 
Hmm...
I ain't no professional editor so I can't comment anymore than I already have.

I'm also not really sure why you'd be syncing your audio after your picture lock. How would one sync the audio after the clapperboard and clap have been stripped out of the edit..?

That's a valid question. Though, it's a question that would only be asked by someone who doesn't edit. :P

For audio purposes, the slate is a few decades out-of-date. Nobody syncs to the slate anymore. In a thread not long ago, I openly questioned why the heck we even use the slate. I got some good answers, and all of them had to do with visual organization. There's rarely a need for that CLAP.

This is all assuming that we're talking about video (which is what pretty much everyone on this website shoots on). I suppose actual film is different, but for video, there is scratch audio, recorded with video. Syncing external audio is easy as pie.

If you've got an audio guy, you can rest assured he will not be editing audio in FCP or Premiere. You will deliver him a locked edit, and all the audio files, and they will go to town, using their own preferred audio editing software. They won't need the slate.
 
Yeah, I've sat in on audio sessions on Pro Tools stages...

I suppose I've worked quite a lot in actual film or Red/Alexa, rather than video in general so I suppose my questions and viewpoint originate from that point of view - we always slate everything, but then there's no scratch track on film, and the workflow to get a scratch track into Red or Alexa is hardly anything easy ;)
 
This is quite a complex issue. When working with physical frames, say with 35mm film, frame rates are extremely important and mixing frame rates is going to become an audio nightmare unless it is very carefully planned in advance of filming and the production sound mixer and audio post team really know what they are doing, as it involves carefully manipulating the sample frequency of the audio both during filming and in post.

Working purely in digital for capture, editing and output can make life easier, in a sense, because converting between frame rates can be done with interpolation, rather than speeding up or slowing down fixed physical frames. However, there are numerous dangers with this workflow too. How exactly is the software interpolating, is it just interpolating the images or is it also interpolating the linked sound? If it's the latter, you're likely to run into a bunch of hurt. If it's just the image then the only worries are visual artefacts and any slight timing changes, which could easily affect audio sync.

Very recently I had a 1080p short film delivered to me for audio post and it had been rendered at 24fps by mistake, instead of at the original and delivery frame rate of 23.976fps. This caused all kinds of audio problems and was very difficult to diagnose because of course both these frame rates have 24 frames in a second of time-code. Had this problem not been detected, there's a pretty good chance that by the end of a 5 minute short, even with this very tiny difference in frame rates, the audio would have been out of sync with the picture by about 7 frames.

There are several different workflows to how we sync sound to picture, the best is from time-code embedded in the audio files during filming but this requires the recorder to be time-code sync'ed with the camera/s during filming. Our preliminary guide is from the work of the picture editor but this usually need adjustment as the editor does not have the equipment to accurately sync sound and picture. It's a big subject area to get into in a short post, so I won't! But just to say; if you can, always film with camera/recorder time-code locked and always slate, as the more info we have the better. I gave some info on workflow and materials exchange between picture editor and audio post here: http://www.indietalk.com/showthread.php?t=41292

Bottom line is that changing or mixing frame rates is fraught with dangers, especially if handing off the sound to an audio post guy. You might be lucky and get away with it, depending on what your software does and how you use it but as a general rule, avoid it like the plague. As a cautionary tale, I was once involved in a professional project which ended up going bankrupt and never got finished because of frame rate issues. I've heard similar tales from others.

G
 
Back
Top