24p to 25p with no change in pitch or duration - how?

[bizz & buzz]

I have a main-stream video-clip (which I assume was shot as film at 24 fps) that was converted to 25p (PAL) but duration and sound pitch didn't change, and are exactly the same as the album version. How it was done?

[GodofaGap]

You can do a field shift, blend conversion, or simply have duplicate frames.

[wonkey_monkey]

I have a main-stream video-clip (which I assume was shot as film at 24 fps)

Your assumption could be faulty. Or it could have been standards-converted with something better than what us Doom9ers can currently get our hands on - hard to say without sample.

David

[zee944]

Or it could have been standards-converted with something better than what us Doom9ers can currently get our hands on
No such thing. Unless somebody invented the perfect motion estimation software, which is virtually impossible. It's field blended or has duplicated frames, something like this. A faulty conversion.

[Mounir]

How was it done.. i 'd say it depend on the length. If it's a short video clip like 4-5min which is standard for v.clips you don't have to bother to change the sound but that's just my opinion.

[Didée]

If it's a short video clip like 4-5min which is standard for v.clips you don't have to bother to change the sound but that's just my opinion.
Did you note the talk is not about 23.976<>24, but about 24<>25? With a 4-minute clip, a speedup to PAL/25 is running 10 seconds shorter than NTSC/FILM/24. It's completely out-of-question whether that is acceptable or not. It is definetly unacceptable.

[wonkey_monkey]

No such thing. Unless somebody invented the perfect motion estimation software, which is virtually impossible. It's field blended or has duplicated frames, something like this. A faulty conversion.

Eh? I'm not claiming it's perfect, I'm just saying there are tools out there that do a better job than the kind of scripts that get posted here, and accordingly costs thousands of pounds/dollars/euros.

And then you conclude it must be field blended or repeated? Have you seen it?

I don't get it :stupid:

David

Edit: I may have made the unwarranted assumption that the clip in question doesn't look as if it was a conversion, whereas B&B was only talking about the audio remaining seemingly unaltered...

[zee944]

And then you conclude it must be field blended or repeated? Have you seen it?

Exactly. I don't have to see it, it's just the law of physics. How do you make 25 pictures from 24? You don't have to write a script for that. Just think by human logic.

[bizz & buzz]

It seems that my initial assumption that music videos all over the world are shot as film was wrong.

The clip in question was produced in the UK which is a PAL land. The possibility it was actually shot at 25 fps resolves the issue of not having any conversion evidence (same length as the album track, same pitch, and no visual evidence either).

Samples:
Video (m2v) (http://www.mediafire.com/file/otdil4mhhjt/sample.demuxed.m2v)
Audio (wav) (http://www.mediafire.com/file/2jwzm2mw2nw/sample Ta0 48K 16bit 2ch.wav)

This conclusion was rather easy to get to, after I examined the rest of the videos on the source DVD (a MixMash.com DVD).
Some of them are blended, some have interlacing artifacts, and some are purely progressive like the example above.

Conclusion: Music videos are not always shot at 23.976 fps. When facing difficulties trying to determine their "true" frame-rate, the country they were produced in seems to be a good starting point to explore.

What do you think?