[FFmpeg-devel] [PATCH 1/4] avcodec/mdec: DC reading for STRv1 is like STRv2

[Michael Niedermayer]

Hi aybe

On Sat, Jan 13, 2024 at 02:28:52AM +0000, aybe aybe wrote:
> Here are the two STR files I have used when writing this patch: https://github.com/aybe/FFmpeg-PSX-STR-tests

ok, i can confirm the version patch fixes these, i will apply it


> Fanatics would probably say that 30 FPS for NTSC is wrong (i.e. should be 29.97)...
> However, as the reversed-engineered docs in jpsxdec mentions, it is sort of impossible to figure out which value to snap to from how frames spans across CD-ROM sectors.
> The sector count per video frame always seem to juggle between two values, e.g. 6 sectors, then 7 sectors, rinse/repeat. i.e. it's never constant.
> 
> Also, as one can see in the various code samples online on writing a PSX program that plays MDEC videos, it is the responsibility of the coder to present the frames on screen.
> i.e. there is not definitive way on how to do so, and even if there was, you can be sure some folks did not play by the rules back then.
> The only way to figure out how a movie was intended to be played is to reverse engineer a game to see what values they did cram in.
> In short, the computed frame rate is 99% good, not 100%. But for mere mortals, they are unlikely to notice it at all.

Well, with the subset of samples i have, i dont feel confident that i could
write and test timestamping/fps.

The docuemnt you linked says
"Data is read from the disc one sector at a time at either 75 sectors per second (single speed) or 150 sectors per second (double speed). The video and audio are spaced out over these sectors so they can be delivered at the appropriate times."

what i would suggest to try is:
avpriv_set_pts_info(st, 64, 1, 150); (or 75 for single speed)

and then set AVPacket->pts to the sector number
you never set fps, leave it to libavformat to figure it out.

can you try that ? (i assume you have many samples to easily check)


> 
> As for the movies in the samples repository, they are corrupt and FFmpeg fails at them, obviously.
> I figured these ones were by checking them in a hex-editor but also by loading them in https://github.com/m35/jpsxdec<https://> which has great logging.
> Not sure why these samples were corrupt in first instance, maybe it was intentional for testing? I can't tell.

Thats very strange, anyone remembers where these samples are from originally ?

thx

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Modern terrorism, a quick summary: Need oil, start war with country that
has oil, kill hundread thousand in war. Let country fall into chaos,
be surprised about raise of fundamantalists. Drop more bombs, kill more
people, be surprised about them taking revenge and drop even more bombs
and strip your own citizens of their rights and freedoms. to be continued
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20240114/bb1aa206/attachment.sig>