28 Nov
2004
28 Nov
'04
10:06 p.m.
> > > That's a perfectly valid point. But
for every disgruntled nvidia user
> > > there are 100 who are just happy to be able to play UT2004.
> >
> > Are you sure? I don't know but i've literally seen tons of complaints.
> > It scared me away and i just stick with my old g400.
>
> I am just judging from reports on LKML. There seem to be a LOT of
> people using this driver successfully. The consensus on LKML seems to
> be that the Nvidia driver is of high quality. The Nvidia folks really
> seem to stay on top of kernel development, and the driver usually works
> with the latest kernel.org kernel, unlike some vendors who build one
> module for one kernel version and claim to "support Linux" because it
> works with RH9 or something equally ancient.
>
We're running NVIDIA graphics cards and the NVIDIA
driver on over 100
systems where I work (because no other card/driver combination can do 3D
as well - using IVS Fledermaus). Almost all of these systems are SMP,
most are state of the art. We have little to no problems with the
drivers.
I'm using Nvidia hw, and the 3D performance from the binary is great. However, the
driver trashes my frame buffer every time I start X, so I can't use any of the VTs
once I start X. This bug has been known for a year and a half now and Nvidia still
hasn't fixed it. There are entire websites dedicated to hacking the old drivers with
binary patches to work with kernel 2.6 just to get around this stupid bug. I appreciate
the work Nvidia does to keep the driver up to date, but I tend to think this would have
been fixed a long time ago if it were an open driver. I don't mind binary drivers
_when_they_work_. I quickly find them irritating when there is a bug and there is nothing
I can do about it. I don't like the idea of having similar situations with audio
hardware.
-Reuben (RME Digi96/8 PAD user)