[MPlayer-dev-eng] [PATCH] automatic fps for mpeg4 raw/in TS

[Michael Niedermayer]

Hi
On Tuesday 11 January 2005 21:27, Nico Sabbi wrote:
> Nico Sabbi wrote:
> > Nico Sabbi wrote:
> >> Nico Sabbi wrote:
> >>> Michael Niedermayer wrote:
> >>>> Hi
> >>>>
> >>>> On Wednesday 05 January 2005 23:12, Nico Sabbi wrote:
> >>>>> Michael Niedermayer wrote:
> >>>>>> Hi
> >>>>>>
> >>>>>> On Wednesday 05 January 2005 22:09, Nico Sabbi wrote:
> >>>>>>> Hi,
> >>>>>>> this patch permits to determine automatically the initial fps of
> >>>>>>> mpeg4
> >>>>>>> video in the VOL header,
> >>>>>>> so there's no need to specify it with -fps.
> >>>>>>>
> >>>>>>> Michael, can you please check it?
> >>>>>>
> >>>>>> did u try it with a few different framerates like a 24000/1001
> >>>>>> .m4v stream
> >>>>>> generated by ffmpeg? i doubt it will work
> >>>>>>
> >>>>>> [...]
> >>>>>
> >>>>> I checked some fancy fps but not the fractional ones; now I see they
> >>>>> aren't recognized
> >>>>> correctly.
> >>>>> If I understand correctly the specs the fixed_timeinc is given by the
> >>>>> smallest
> >>>>> number of bits (up to 16) to represent   timeinc_resolution-1,
> >>>>> correct ?
> >>>>
> >>>> IIRC, yes
> >>>>
> >>>>> What's wrong? :)
> >>>>
> >>>> many mpeg4 files have no fixed_timeinc
> >>>> [...]
> >>>
> >>> This one works correctly with all fps I tried, both with lavc and
> >>> with xvid, in raw files and in TS.
> >>> Is it ok ?
> >>>
> >>>    Nico
> >>
> >> sorry, this doesn't work with b-frames :(
> >> next time ...
> >
> > given that 3 consecutive frames are always enough to determine the
> > smallest delta_timeinc
> > (or I'm mistaken?) this patch should work correctly (it does for me,
> > even seeking with -sb).
> > Michael, what's your opinion?
[...]
> Michael, any comment on this patch or should I consider it rejected?
> In this last case what other method can you advise?

i havnt had time to check it, feel free to commit it if u think its ok

[....]
-- 
Michael

"In any case, just because code is syntactically "valid" GNU C doesn't 
mean gcc can always compile it." -- justification to close a gcc bug