On Fri, 2005-11-25 at 22:23 -0500, Timothy Miller wrote:
> On 11/25/05, Jack Carroll <[EMAIL PROTECTED]> wrote:
> > On Thu, Nov 24, 2005 at 06:45:05PM -0500, Timothy Miller wrote:
> > >
> > > Although nothing is set in stone, there may be grounds for not
> > > supporting AGP.  Or if we do, it won't be right away.  But there's
> > > still time to debate that issue.
> > >
> >
> >         Hahh??!  That's going to exclude a heck of a lot of installed base
> > -- including my expensive new workstation.
> 
> If I can help it, we'll have AGP.
> 
> But the thing is, anyone with AGP also has PCI, and if it's "fast
> enough" with PCI, how much do we care about AGP?
> 
> Really, those with AGP aren't excluded.  They're just at a
> disadvantage relative to those who have AGP cards.  But there are lots
> of disadvantages to OGA; they're just ones we are willing to tolerate
> because we demand open source drivers.

Just a thought here..

What about making the card a hybrid?

IE: one side has PCI, the other has AGP.
IF nothing else, the AGP can just be plain PCI using AGP slot.

If there is a possibility after release for the AGP speed to be
increased to AGP-1x or something that would of course be a plus.

Detecting whether AGP or PCI is the one being used should be
no problem. Trace length? Maybe a problem, but I guess only for
the higher-speed grades of AGP?

Benefits:
-One more slot to stuff the card in
-Enthusiasts can now have a FPGA card in an AGP slot
 (I guess this could be used to something fun..)  =)
-Extra "Wooo.....oow" value.

-HK

_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to