Status of xserver/debrix/modular tree?

Felix Kühling fxkuehl at gmx.de
Sat Feb 12 15:57:39 PST 2005


Am Samstag, den 12.02.2005, 16:46 -0500 schrieb Michel Dänzer:
> On Sat, 2005-02-12 at 13:09 +0100, thomas klein wrote:
> > On Saturday 12 February 2005 05:13, Bernardo Innocenti wrote:
> > [...]
> > > Running a full-GL desktop would be very sexy, but there's still too
> > > much popular hardware that don't support (fast enough) 3D
> > > acceleration. Linux distributors wouldn't take this route, at least
> > > not for a few years.
> > 
> > I don't know where a new xserver will go except full ogl output.
> > 
> > If "too much users" (hey, I got old cards making a pretty good job at 
> > ogl) [...]
> 
> [...]
> 
> > cards, but not usable)(I know, it's experimental). Ogl will simply 
> > put the accel stuff in mesa, it is allready the case, it's a standard 
> > well supported by gfx manufacturer.
> 
> I basically agree, but have you tried Xglx on free vs. proprietary
> OpenGL drivers? The former still have some way to go to make it usable
> (but some pressure like the X server requiring it might help them get
> there pretty fast :), and we don't want to require proprietary drivers
> for the eye candy, do we?

I tried Xglx on the Savage driver one or two months ago. Colors were
completely messed in 16bit color depth (may be fixed by now, I haven't
updated it in a while). In 32bpp it looked ok, but was dog slow (though
SAVAGE_DEBUG=fall didn't indicate any software fallbacks). I'd have to
get back to it some time and run oprofile on it to see what is eating
the performance.

Also rendertest_glitz_glx shows "not supported" for most operations when
used with Savage DRI, while the software renderer supports most
operations. Which extensions would a driver have to support in order to
improve this situation?

Thanks,
  Felix

-- 
| Felix Kühling <fxkuehl at gmx.de>                     http://fxk.de.vu |
| PGP Fingerprint: 6A3C 9566 5B30 DDED 73C3  B152 151C 5CC1 D888 E595 |




More information about the xorg mailing list