Re: OT: nVidia driver [was: Wish list] -- understanding the GPU market

["Bryan J. Smith <b j smith ieee org>" <thebs413 earthlink net>]

Alan Cox wrote:  
> The lack of an ATI open R300 driver is at least partly certain Linux
> organisations fault for not taking up opportunities rather than ATI's.
> (and more I cannot say in public)

Thanx for that insight.

> Commoditisation means that very soon all the 'must keep secret' IP
> will be only relevant to real time ray tracing. What happens to the
> Radeon 9800 when Intel 9xx/VIA/etc graphics on chip are good enough
> for gamers. Leading-edge video will be for VR nuts.
> (And guess why Nvidia and ATI are in the chipset business nowdays)

Another good insight.  Of course, I still see the GPU market with a good
3-5 years more of Triple-Moore's Law, unlike CPUs.

In reality, people should be interested in vendors like 3DLabs who want
to keep OpenGL pure.

> Hardly. AGP is a published open specification. People like nVidia do some
> really clever tricks with it for performance like the DMA contexts but nothing
> major we don't actually understand AFAIK.

Well, from what I understand, there are many unpublished aspects to
AGP, ones that only Intel licensees have, or have been reverse engineered.
And nVidia is legally bound by those and not to work on a reverse
engineered version.

Same deal with the NIC in their nForce chipset, and why they release
the nvnet driver, but can't officially work on the forcedeth.

> Which would be why folks are currently just starting to sort out PCI express
> ATI radeons right now of course.

Exactomundo.

The GPU market is a catch-22 when it comes to commodity, at least for
awhile.  But it is very nice to see Fedora Core 3 install DRI/GLX out-of-the-box
on a i855GM in a notebook.


--
Bryan J. Smith   mailto:b j smith ieee org