From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Thu Dec 12 2013 - 04:32:46 CST

On Thu, Dec 12, 2013 at 11:16 AM, Bogdan Costescu <bcostescu_at_gmail.com> wrote:
> On Thu, Dec 12, 2013 at 7:04 AM, Axel Kohlmeyer <akohlmey_at_gmail.com> wrote:
>> while nvidia engineers do on and off things that irritate the hard
>> core open source people, their proprietary drivers are way ahead of
>> the corresponding amd offerings. that is in terms of features,
>> configurability, consistency and performance.
>
> I very much agree. Whenever Linux, OpenGL and performance are
> mentioned in the same phrase, I only recommend nVidia cards - for more
> than 10 years! Things might change in the near future; Valve's Steam
> Machine was initially designed around nVidia's offerings - for the
> same reasons Axel mentions - but is supposed to get an extended design
> including AMD graphic cards. As SteamOS is Linux-based, this might
> finally give AMD the incentive to improve their Linux support...

the IT business thrives on promises, but very few of those actually
become reality. and having a linux base doesn't automatically mean
that there is a lot of trickle down. just look at android. finally,
please keep in mind that amd is providing graphics/cpu hardware for
both the new xbox and the playstation, that will likely tie quite a
bit of the available resources (not to mention the windows market).
linux and particularly opengl on linux is very small potatoes in
comparison. to me, the increased support from amd for open source
drivers is an indication of that.

axel.

p.s.: perhaps i should also mention in this context that after many,
many, (far too) many years finally the OpenGL support for intel
graphics has reached a point, where it is working ok with VMD and no
longer crashing or rendering scenes incorrectly. while the speed is
nowhere near what good nvidia or amd hardware offers, at least you can
see GLSL content correctly.

>
> Cheers,
> Bogdan

-- 
Dr. Axel Kohlmeyer  akohlmey_at_gmail.com  http://goo.gl/1wk0
International Centre for Theoretical Physics, Trieste. Italy.