On 3/2/06, Rahul Sundaram <sundaram@xxxxxxxxxxxxxxxxx
> wrote:
You're referring to the GLX_EXT_texture_from_pixmap extension. Of course there will be sharing but ultimately they are two competing methods of improving X. AIGLX, AFAIK just appeared a few weeks ago. Xgl has been in the works for a few years, and there's alot of thought behind it. There's an architecture (much like SELinux...) and while I'm still not certain of this, hopefully more than eye candy.
AIGLX appears to me to be all eye candy. It's been put together last minute as an answer to Xgl. I wasn't impressed by the demos of AIGLX. I was blown away by the demos for Xgl. I like how Miguel de Icaza characterized Xgl as "a lot of work" and that's what's needed. It would be cool if Red Hat got behind Xgl to accelerate development on Xgl, just like Novell should've stayed with SELinux and accelerated it's development.
But it is somewhat black and white to pit XGL versus AIGLX, nevertheless I think it will prove true in the end.
Benji
Benjy Grogan wrote:
>Isn't Red Hat forking the accelerated GL community by working on AIGLX
>instead of Xgl? It seems like Xgl is to AIGLX as SELinux is to AppArmor.
>
>
The analogy here seems to be off the mark since AIGLX and SELinux has
been developed in the open right from the start unlike AppArmor and XGL
and unlike security frameworks such as SELinux, there are different
approaches you can experiment with in a budding development of
accelerated interfaces without any fundamental forks. You might want to
read the FAQ at http://fedoraproject.org/wiki/RenderingProject/aiglx on
doing incremental updates in comparison to whole sale rewrites of a X
server.
The media likes to paint black and white situations of X vs Y while
there is also a code sharing between XGL and AIGLX going on.
You're referring to the GLX_EXT_texture_from_pixmap extension. Of course there will be sharing but ultimately they are two competing methods of improving X. AIGLX, AFAIK just appeared a few weeks ago. Xgl has been in the works for a few years, and there's alot of thought behind it. There's an architecture (much like SELinux...) and while I'm still not certain of this, hopefully more than eye candy.
AIGLX appears to me to be all eye candy. It's been put together last minute as an answer to Xgl. I wasn't impressed by the demos of AIGLX. I was blown away by the demos for Xgl. I like how Miguel de Icaza characterized Xgl as "a lot of work" and that's what's needed. It would be cool if Red Hat got behind Xgl to accelerate development on Xgl, just like Novell should've stayed with SELinux and accelerated it's development.
But it is somewhat black and white to pit XGL versus AIGLX, nevertheless I think it will prove true in the end.
-- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
