[MPlayer-users] Bicubic scaling in vo_gl - does it ever work?

Reimar Döffinger Reimar.Doeffinger at stud.uni-karlsruhe.de
Mon Jun 12 22:30:25 CEST 2006

On Mon, Jun 12, 2006 at 09:25:58PM +0200, Guillaume POIRIER wrote:
> On 6/12/06, Vladimir Mosgalin <mosgalin at vm10124.spb.edu> wrote:
> >OK, more interesting info. I recompiled mplayer in debug mode and found
> >two things:
> >1) It doesn't hang. It just becomes really slowwwww.. Like 30 seconds
> >   per frame or so.

I get the same thing at university, but those are older cards and I
think they just can't handle it.

> >How come? My system can do over 10k glxgears! Is that scaling shader
> >THAT slow?

If it is above or near what the limit can do, yes. You haven't mentioned
which card you have, so I can't tell you what to expect. But if it's
older than I'd say 2 years, don't expect it to work I'd say.
My Geforce 5700 is older than that, but nVidia started earlier with this

> Anyway, an update has just been committed to svn, so why don't you try it 
> out?

The generated fragment program looks like he is already using it.
And as to -vo gl working with ATI cards, I also have a Radeon 9250 PCI,
and -vo gl does work with it, though under Linux only without yuv->rgb
and under windows only with yuv=5 - but this is as expected, it is too

Reimar Döffinger

P.S. during my efforts to get the ATI card to work properly under Windows,
I couldn't help but wonder what those people talking about ATI drivers getting
better meant. They seem to be as buggy and crappy as ever, only in
addition nowadays they want to install a pretty useless 50 MB "control
panel" in addition... well, at least I have (extremely) basic OpenGL
support via native, non-binary linux drivers now, which after all was
the main point of it.

More information about the MPlayer-users mailing list