Am 14.09.2022 um 19:00 schrieb Dan:
That's a great idea to use specified colours directly supported by ffmpeg
and helps narrow down the issue a lot!

Would you count this as a bug?

It seems to be a bug.

If this is a known legacy quirk, it seems
one of the ugliest and most misleading quirks I have seen in a while, and I could imagine being responsible for wasting thousands of hours of thought based upon
mis-diagnosed colour-profiling assumptions.

I've spoken with two separate long-time video professionals so far, and they were both convinced this was a colour profile thing and that my source images were the problem, and not ffmpeg's behaviour, despite saying I checked both BMPs in a hex editor, and they were byte-identical (other than resolution/filesize).

Are you using Windows per chance?

yes, I did also test on Windows.

To make things even more confusing, one
of the aforementioned pros says he can't spot a difference on his non-Windows system (presuming Mac, could be Linux) between the two output mp4s' colours.

The crazy thing is when you vertically stack the two videos together, then the colors become again identical in the output:

ffmpeg -i out1.mp4 -i out2.mp4 -lavfi vstack -y out.mp4

Sorry, I have no idea for a solution or workaround.

Michael

_______________________________________________
ffmpeg-user mailing list
ffmpeg-user@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".

Reply via email to