> > > > > > > slow AGP bus
> > > > > >
> > > > > > OGC will have an even slower bus.
> > > > >
> > > > > Depends on how many PCIe channels we design into it.  There will be at
> > > > > least one (about as fast as PCI-32 66MHz).
> > > >
> > > > So OGC is PCIe now?  Last I read it was going to be PCI.
> > >
> > > The engineer who has been working on our PCB for us has expressed
> > > interest in doing a PCIe variant of the board.  A tentative plan has
> > > been proposed, and a PCI-PCIe bridge chip has been selected.
> >
> > So the ASIC will be PCIe and the PCI board uses a PCI-PCIe bridge chip?
> >
> > Doing it the other way around (ASIC talks PCI and the PCIe board uses bridge
> > chip) doesn't makes sense, as the PCIe board would still have the PCI
> > bottleneck.
> 
> If we stick with PCIe 1x, then it makes no difference since 66MHz PCI
> has roughly the same bandwidth (more in potential, less in practice).

Except that in the real world PCI slots are only 33 MHz.  I just checked
the manual for my newest machine (AMD64 PCIe + PCI, I/O out the wazoo)
and the PCI slots are only 33 MHz.
_______________________________________________
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