Exporting to ProRes (packaging?) and frame-rate accuracy.

[IoannisSyrogiannis]

Well, in-house (that is less serious), it does cause duration difference of the video in an editor, so it could end up in an increasingly bad synchronisation of different essences.
Forwarding the export to third parties, on the other hand, could lead to the file not being accepted.

The first issue that concerns me here is when one needs to apply some kind of versioning that they created with the help of DCP-o-Matic or want to create on the next step and the non-existant preset for 24-to-24,01 fps (or the other way around) conversion of the video (or other element) cause inaccuracies.

The deriving concern is archiving. Meaning that, on occasion, one may prefer to keep a ProRes export of a DCP, instead of the DCP itself in one’s archive. If one can’t be certain that the export will “play well” with other programs or future programs/encoding systems that are meant to support ProRes, they’ll have to either search for other export means or keep copies of both formats.

I am now wondering if the title of the thread is not quite evident to the subject in hand.
Would someone with the same concerns find their way here, so they can in fact share experience and troubleshooting info?

You, Carsten, really invested time, thank you! But other people’s input might lead to other ways of looking the facts.

[Carsten]

Oh, I don't want to express this issue as being irrelevant - I am sure Carl will find some reason for this in the export parameters to fix it. As I said, when I reimported the export into DCP-o-matic, the number of frames is the same as for the DCP. So my question was: What effect does this interpretation really have when actually working with the exported ProRes?

I'll do a test with my counting-up test file. Will be interesting to step through it in premiere if the numer of frames given there actually differs. Will need to add some audio sync cues first, though. Although sync issues may only occur with a very long file, I guess.

Google shows loads of hits about FFMPEG and VFR, so, I guess it's just a matter of time to get this solved. Maybe just a flag that Carl omitted.

- Carsten

[IoannisSyrogiannis]

So my question was: What effect does this interpretation really have when actually working with the exported ProRes?

If you are exporting in order to use the export with DCP-o-matic, I don't see any effect on it.
I think I wrote it too, that when I imported the exports on DCP-o-matic, the frame count was as intended.
Yet, if there is a way to have the exports to be program blind, it would be great.

[gunnar]

This start to be real problem when the prores need to be converted to 25fps video for other use.
I use protools to speed the audio up from 24->25fps (to avoid pitch changes) and when the audio is not true 24fps then pro tools are unable to do the conversion correctly since it expect the audio to be true 24fps. When it's not the sound won't be in sync in the 25fps version since the calculation will be incorrect if the source sound is not what pro tools think it is.

[Carsten]

Maybe, yes, but the trouble is, we don't know exactly how these files are interpreted and dealt with. I think they are ok as far as the assets go within them - otherwise, reimporting a ProRes export from a DCP wouldn't give the exact number of frames and audio blocks as in the original DCP source (which is the case for my own tests).
Strange thing - the MP4 version inspected in Premiere is shown as 24.00fps, but with 3 frames less (2194 vs 2197 for DCP/Prores) - but, when reimported into DCP-o-matic, comes back with the exact same length/frames/blocks as the original DCP and ProRes export.

Whatever - that variable frame rate flag is certainly wrong (and useless for DCP-o-matic's working scope) , and there's probably a way to get rid of it.
I can see that Handbrake offers constant vs. variable frame rate for exports explicitly. I have no idea how it actually deals with some cross conversions, like CFR to VFR. The result would be a file with CFR, but flagged as VFR?

- Carsten

[IoannisSyrogiannis]

As the OP of this thread, I maybe should write here what is written in this thread about the fact that the issue is (somehow) resolved.
I did two tests with the latest stable today (2.14.17), and I can't find any frame-rate inconsistencies or misinterpretations on the exports.
Not when someone exports a DCP, nor when someone exports another kind of video.
Thank you for your support, Carl.

Next step, sharing processing power on ProRes exporting!