[Mplayer-cvslog] CVS: main/libmpcodecs vd_ffmpeg.c,1.85,1.86 vd_libmpeg2.c,1.20,1.21 vf_pp.c,1.26,1.27 mp_image.h,1.22,1.23
D Richard Felker III
dalias at aerifal.cx
Sat Apr 19 00:04:27 CEST 2003
- Previous message: [Mplayer-cvslog] CVS: main/libmpcodecs vd_ffmpeg.c,1.85,1.86 vd_libmpeg2.c,1.20,1.21 vf_pp.c,1.26,1.27 mp_image.h,1.22,1.23
- Next message: [Mplayer-cvslog] CVS: main/libmpcodecs vd_ffmpeg.c,1.85,1.86 vd_libmpeg2.c,1.20,1.21 vf_pp.c,1.26,1.27 mp_image.h,1.22,1.23
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
On Fri, Apr 18, 2003 at 10:34:13PM +0200, Michael Niedermayer wrote:
> Hi
>
> On Friday 18 April 2003 15:49, D Richard Felker III wrote:
> > On Fri, Apr 18, 2003 at 03:19:25PM +0200, Michael Niedermayer CVS wrote:
> > > Update of /cvsroot/mplayer/main/libmpcodecs
> > > In directory mail:/var/tmp.root/cvs-serv23978
> > >
> > > Modified Files:
> > > vd_ffmpeg.c vd_libmpeg2.c vf_pp.c mp_image.h
> > > Log Message:
> > > passing qscale_type around so the pp code can fix the mpeg2 <<1 thing
> >
> > Filters that pass through the qscale table should also be updated to
> > preserve this.
> yes, but we should find a less braindead way to do that instead of putting it
> in every filter
>
> some clone_mpi_attibutes() perhaps, what do u think? any better ideas?
Not really. There are only a very few filters which should pass quant
tables thru... After many filters, the info in the table becomes
worthless (often even worse than no info at all).
Rich
- Previous message: [Mplayer-cvslog] CVS: main/libmpcodecs vd_ffmpeg.c,1.85,1.86 vd_libmpeg2.c,1.20,1.21 vf_pp.c,1.26,1.27 mp_image.h,1.22,1.23
- Next message: [Mplayer-cvslog] CVS: main/libmpcodecs vd_ffmpeg.c,1.85,1.86 vd_libmpeg2.c,1.20,1.21 vf_pp.c,1.26,1.27 mp_image.h,1.22,1.23
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
More information about the MPlayer-cvslog
mailing list