[Dri-devel] Re: Number Nine Revolution IV - DRI?

[Mike A. H. <mh...@re...>]

On Tue, 4 Mar 2003, Carlos Noguera wrote:

>>Is there any reason why you have to have support for that card?  AFAIK,
>>there is no documentation available for that card, and the company that
>>made the chip no longer exists (at least not in its previous form).  Not
>>only that, there are MUCH better cards available for PCI and AGP that
>>perform many orders of magnitude better.  I seem to remember that this
>>card can't even play the original GLQuake at a decent frame rate.
>
>I don't *need* DRI/OpenGL support for the card, but it would be incredibly
>nice to have it. Right now it is in my current setup, driving my SGI 1600SW
>flat panel and is the only AGP card that will drive that monitor without
>using SGI's Multilink Adapter (expensive, $450+ on eBay). It's an excellent
>display, and has Xfree support, so I am not really about to part with it.
>The 1600SW is hands down the best flat panel at an affordable price that I
>have seen.
>
>That being said, I thought this might be a nice challenge to a developer,
>help someone make some money, get me OpenGL (not looking for a gaming card
>anyway...), AND give back to the community all at once.

Certainly an admirable offering.  ;o)

Not to discourage you, but just speaking of the likely reality
that the cost of a brand new card, including that $450 Ebay gizmo
on top of it, would more than likely be several thousand dollars
cheaper than paying a developer to write drivers for an ancient
video card from a company that doesn't exist anymore, and do so
without hardware specifications.

In contrast, there are several people working on ATI Mach64 3D 
support right now, whom do have the specifications for the 
hardware, as well as prior source code to use as a basis 
(Utah-GLX), and with help/support from other developers whom have 
worked on the hardware before, including tips from ATI.  This 
project AFAIK is coming along nicely, and will hopefully be in a 
future X release.

To attempt to do the same thing, for a card that barely nobody 
uses anymore, and without docs, or access to people familiar with 
the hardware, is extremely an unlikely thing to happen from 
anyone volunteer or otherwise.  ;o)  Even if you had $10000 to 
put up for driver development, I doubt anything would come out of 
it in the end.

Again.. not to discourage your effort.  Just trying to provide 
what I believe to be a realistic expectation right now all things 
considered.



-- 
Mike A. Harris     ftp://people.redhat.com/mharris
OS Systems Engineer - XFree86 maintainer - Red Hat