Thoughts on working with true 24 fps?

El Director

El Director

Veteran
With the new Magic Lantern release for Canon DSLRs, it's possible to shoot at a true 24 fps. Yes, I realize that it's "basically" the same as 23.98, but I wonder if there might be timecode advantages to shooting at 24 instead. Any thoughts?
 
El Director said:
With the new Magic Lantern release for Canon DSLRs, it's possible to shoot at a true 24 fps. Yes, I realize that it's "basically" the same as 23.98, but I wonder if there might be timecode advantages to shooting at 24 instead. Any thoughts?
Click to expand...

Some one would have to argue with me, and have factual examples that show any difference in the 'short run'. Sure if after a 2 hour movie, which yields a 173 frame 'difference'... someone complains...

But that would only be a real complaint if one were matching up 24 (true) fps, with 23.976 fps...

Otherwise, all slating, sound sync is done on a 'short' shot by shot basis, so there is no real reason to use 24 vs 23.976 other than what is more convenient at the time... (I do think that 23.976 maps into NTSC color 3:2 pulldown 'correctly' with no frames left over...). Whereas 24 fps would have to be 'slowed' by 0.1%

I personally think this would only be a real problem for people who want to fit the program to specific broadcast requirements. The typical theater presentation runs the film to end credits, and having a 7 second 'difference' is negligible...

But I'm sure there's some pixel peeper out there who can tell the difference between a 180 deg shutter on 23.976 fps... vs 24 (true) fps...

The other time frame rate may be a real issue is to capture high speed events... like in the 2000 fps zone... but then one is not shooting for hours at that rate anyway... the electric bill to power the lights would be prohibitive...


That said... if one were shooting in a 'mix media' environment that included 24 (true) fps devices, such as Film Film... then shooting the Canon at 24 (true) fps would be a 'correct' choice, because then frame for frame the Canon's would 'sync' to the film (or be a constant phase 'off'...).... well modulo the time circuit differences... I don't know if the Canons allow for time code 'jamming'... which would be required if Canons and other 24 (true) fps devices were used on the 'same shot'... and absolute sync were required during the whole shot. (For example, Canons in 'crash camera' usage... many of them for some big FX scene...).
 
Last edited:
+1, and:

23.976 is most useful if you're going to output to interlaced video, which is 59.97 in the US (NTSC). 24p works best when going out to film or to computers.
 
+1 on PopcornFlix comment. Even if the difference between the two is minute or virtually undetectable, your acquisition should align with your intended distribution. So 29.97 for NTSC distribution and 24 for film (and PAL too I guess) would be the sensible way to go in my opinion.
 
I'll do some tests and see. With my workflow, I have to do an offline edit in Vegas and conform it in AE later. On the last film, shot at 23.976, I had some major drifting issues (I do ADR for the whole film) that I had to go back and fix by hand. I'm wondering if a straight up 24p would be easier for me to work with. Though in DVD Architect, I can only force 23.976 and not 24 when I lay it down... Sorry, just thinking "out loud".
 
El Director said:
I'll do some tests and see. With my workflow, I have to do an offline edit in Vegas and conform it in AE later. On the last film, shot at 23.976, I had some major drifting issues (I do ADR for the whole film) that I had to go back and fix by hand. I'm wondering if a straight up 24p would be easier for me to work with. Though in DVD Architect, I can only force 23.976 and not 24 when I lay it down... Sorry, just thinking "out loud".
Click to expand...

Can you 'explain' the 'major drifting issues'... If you are doing ADR, sampling at 48K, 24 bits, you should not see 'drift' due to any selected frame rate... what you may see is just 'drift' due to the timing circuits of the devices involved, since you probably don't have a 'house sync'.

However, for ADR, I was under the impression that the talent had the visual media as part of the play back setup, and so they would time their voices to the visuals. I think I've also heard about setups where the talent has the audio reference track playing in headphones.

In any case, the ADR audio clips should be short, and one would not expect 'drift'.

A few years back I wanted to ADR some dialog that got botched. The 'thing' I learned was 1) I needed to transcribe exactly what the actors said on screen, 2) have some 'reference' as mentioned above, to get the dialog as close to the on screen lip movements (for CU shots...).

As it was I decided to work with the botched audio...

A thought does occur to me... if your editing software, Vegas, doesn't have sub frame audio 'slip' capability, you may have a problem there (While I can't state absolutely.. I'm pretty sure AE doesn't have any 'sub frame' audio capability...). Final Cut Express had a rather awkward 1/100 increment adjustment 'editing' of audio to allow for finer adjustments of where the audio occurred in a frame. CS5 Premiere allows for selecting 'audio' timing which is 48K... plenty of time resolution to slide audio around anywhere in the frame.

That issue would be a problem no matter what your frame rate was for the popular frame rates. Visuals are sampled so much lower than audio...
 
Gotta agree with j1clark. Obviously I don't know exactly what your workflow looks like, but I don't see any reason for frame rate to be causing drift in your ADR tracks.
 
My ADR recording method is solid. The sync has nothing to do with the audio, it's the video, hence why I think 24p might be the way for me to go. The ADR is recorded and synced on the spot to the offline edit. Everything matches perfect. It's not until I brought the online in that stuff was out of sync.
 
That would suggest that there is a discrepancy between your offline and online sources. Perhaps your offline proxies are tagged as 24p rather than 23.98p?
 
brynsank said:
That would suggest that there is a discrepancy between your offline and online sources. Perhaps your offline proxies are tagged as 24p rather than 23.98p?
Click to expand...
That's what I'm thinking, and I seem to remember that there was a reason I had to do that, just don't remember why. Anyway, I'll do some tests to DVD and see how it all goes. Thanks
 
You must log in or register to reply here.
Back
Top