Re: [Dri-devel] I'm willing to help

[F. <j_r...@ya...>]

On Thu, Jul 25, 2002 at 02:34:15AM -0700, Ed Yu wrote:
> Hi, Jose:
>   I have TNT2, TNT2 Ultra, S3 Vision968, S3 Vision864,
> and ATI Xpert (ATI Rage Pro Turbo) cards and a laptop
> with Intel i810 (I need to double check). I'm
> interested in both driver development or other coding.
> I want DRI to succeed (so eventually, hopefully,
> developers can make more games on FreeBSD or Linux) so
> I'll do whatever needs to be done. What other card do
> I need to get.

Ok, let's go one by one. Regarding nVidia cards see below. I'm not 
familiar with the S3 Vision chipsets but from a quick search on google 
it seems that they don't have 3D capabilities. The ATI Rage 128 chipset 
(included in ATI Xpert) as the Intel i810 are both well supported. 

This is roughly the areas which have been under development lately (by
no particular order):
 - Keith Whitell added TCL (transform-clipping-n-lighting) to Radeon
   (after restructured Mesa internals for that purpose), and is now
   working on the Radeon 8500 driver sponsored by the Wheather Channel
 - *a lot* of people have working in Radeon contributing inumerous
   bugfixes and improvements
 - Leif Delgass and I have been working on the ATI Mach64 driver
 - Erik Anholt and Alan Hourinhane have been working on BSD.
 - Max Lingua added initial support for the S3 Virge
 - Several people (including myself) are interested in supporting S3 Savage 
   chipsets (this has been delayed due to lack of time and chip
   documentation to some extent)

As mentioned on the DRI status page, multihead needs some work in all
drivers, but very few has been done on this that I'm aware of.

>   BTW, just out of curiosity, I understand that the
> nvidia code is closed source but that's probably the
> one everyone wants to use for games, is there anything
> anyone can do?

I'm not quite sure whether the main reason for no open-source nVidia
drivers is the non-disclosure of hardware information from nVidia, or
the availability of binary drivers. Probably both.

The Utah-GLX project has limited support for some nVidia chips. I
recently bought a dusty Riva TNT2 Vanta (to temporarily replace the 
Savage 4 on my brother's computer while I work on it) and I was planning 
to eventually port the Utah driver to DRI. Nothing too serious - it
would be pratically impossible to match the performance of the
closed-source drivers without any documentation. Nevertheless it seems
an interesting endeavour (e.g., for BSD which has no alternatives), and
someone has to give the first step.

I hope I shed some light on all this, and that you have a better
perspective on the DRI development. I believe it's not my role to tell 
you or anyone what he should work on - it's something that you should 
find out by yourself to avoid latter frustration or deception. As you 
make your way I'm sure that it will be clearer.

Regards,

José Fonseca