[MPlayer-dev-eng] Video output for RV280+Mesa3D

[malc]

On Sat, 6 Dec 2008, Reimar D?ffinger wrote:

> On Sat, Dec 06, 2008 at 03:11:26AM +0300, malc wrote:
>> On Sat, 6 Dec 2008, Reimar D?ffinger wrote:
>>> Probably libswscale is broken. It would be really interesting to know
>>> _what_ part of your code makes it faster, though I suspect that swscale
>>> is what causes most of it...
>>
>> I'm pretty sure i know which part makes it THAT much faster, r200 dri
>> driver (i think i915 too) requires the stride to be multiple of 64 to
>> have zero copy texture uploads, something you don't know till you digg
>> into the Mesa sources. Having custom yuyv->yv12 converter helps some
>> too, but not nearly as much.
>
> Actually it was much more stupid, I forgot to set
> GL_UNPACK_CLIENT_STORAGE_APPLE (fixed in SVN).  Still does not make
> a difference with my i945...

You are mistaken. GL_UNPACK_CLIENT_STORAGE_APPLE has no effect
whatsoever unless the stride is sufficiently aligned (again at least
with r200). Anyhow the whole reliance on `-vf format=yuy2' makes sure
that vo_gl will never approach vo_rv280's speed since there's an extra
copying data around going on (decode -> convert -> copy to tex instead
of decode -> covnert to tex) In fact simply adding `-vf format=yuy2'
to `-vo rv280' increases the time taken to show that clip by 3
seconds.

-- 
mailto:av1474 at comtv.ru