State of Linux graphics

[Matthias Hopf]

Jon,

in case I didn't mention so far - thanks for the article, it is a great
summary of the current state. I think it should be linked or even
installed at a pretty prominent place on x.org or freedesktop.org.

> > > Xgl doesn't run standalone, it needs either Xegl or Xglx. Xglx isn't
> > 
> > What do you refer to with Xgl? For me it is the superset of Xegl, Xglx,
> > and all other possible Xservers that are using OpenGL as an abstraction
> > layer.
> > So yes, of course, Xgl can run standalone. If it cannot, so cannot Xegl.
> 
> Xgl is the crossplatform code it's can't run anywhere. You need to
> combine Xgl with platform specific drivers.
> 
> Xglx = Xgl running nested on GLX
> Xegl = Xgl running standalone on OpenGL/EGL
> Xwgl = Xgl running on the MS wgl OpenGL extensions
> Wagl = Xgl running on the Apple agl OpenGL extensions

Ok, so you call Xgl* the actual backends. Very well, that's fine for me,
but David will change the code so that there's actually only one binary
(Xgl) and several backends. Just for info.
I think that was where the confusion came from.

> > > So we are looking at 2007. That means two more year's advances in
> > > hardware and things like a NV 6800GT will be $40. In that timeframe
> > > programmable hardware will be mainstream. We also have time to fix
> > 
> > They are mainstream already. There is just a ton of legacy hardware
> 
> There is no proposal to make GPU programmability a requirement; only a
> proposal to make it accessible. I would prefer to see programmability
> exposed via an existing, standardized API instead of designing a new
> one.

Never talked about requirements - it's just that even nowadays even low
end chips like the Intel chips have pixel shaders, so it's a de-facto
requirement for any company trying to compede.

> Scene graph API is another topic that was not addressed in the paper.
> I do agree that it is worthwhile to expore building one.

Personally I think this is one of the major things we will have to do
after we have (or better: during we build) a working modern hardware
abstraction layer.
I'll take a close look at Avalon in the coming weeks.

> I referenced this paper
> http://www.loria.fr/~levy/publications/papers/2005/VTM/vtm.pdf

That seems to be pretty cool. Pacific Graphics is a good conference,
too.

> allow room in our APIs for something like this in the future. Our
> current subpixel, antialiased fonts are superior today but I have to
> wonder if that will be true on GPUs built two years from now.

The only problem I see is that we will have this discussion client vs.
server side fonts all over again...

CU

Matthias

-- 
Matthias Hopf <mhopf at suse.de>       __        __   __
Maxfeldstr. 5 / 90409 Nuernberg    (_   | |  (_   |__         mat at mshopf.de
Phone +49-911-74053-715            __)  |_|  __)  |__  labs   www.mshopf.de