On 8/21/06, Timothy Miller <[EMAIL PROTECTED]> wrote:
On 8/20/06, Jon Smirl <[EMAIL PROTECTED]> wrote:
> What bus interface is the card going to have? PCI is hopelessly slow
> for graphics.
No, it's not. Yes, it's bloody slow, but we've already done the
calculations, and it's just barely fast enough. But only if you use
DMA. (DMA makes the transfer a few % faster, but mostly, it just
frees up the rest of the system to do useful work.)
You might want to poll your consumers and see if they are going to buy
a PCI board. I only use PCI versions of cards for test and
development. They are visibly too slow for normal use.
There is also the problem with bus saturation. If I have a PCI
graphics card running at full speed it eats all of the bus bandwidth
from other PCI devices. Internal IDE controllers are often on the same
PCI bus even though they are on the motherboard.
This is why we are disconnecting on discussions about frequency of DMA
use. With AGP around DMA is less relevant.
I remember when PCI was the rage.
Anyhow, I've been writing graphics drivers for PCI cards for a LONG
time, even on systems with severe limitations like Sparc, and have
been able to get excellent performance. If your bus is slow, you just
have to make efficient use of it.
> If you do AGP you also need AGP port support.
>
> PCI-express needs the PCI-express way of mapping VRAM into the CPU/GPU
> address spaces.
What is the "PCI-express way"?
The GPU operates in it's own internal 32b address space. There are
still address translation tables on the graphics card mapping from
internal addresses to system addresses.
My suggestion would be to go PCI express and make the card adapt from
1/4/8 lanes. That way I can plug it into an extra 1 lane socket for
development and an 8/16 lane socket for general use. There is no real
gain going from 8 to 16 lanes for any existing apps.
--
Jon Smirl
[EMAIL PROTECTED]
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
