Who cares about the 50$ extra expense while we're spending 300$ plus on motherboards these days?
I only hope this is true,and that SLi works good,unlike Skulltrail.
Printable View
Who cares about the 50$ extra expense while we're spending 300$ plus on motherboards these days?
I only hope this is true,and that SLi works good,unlike Skulltrail.
Skulltrail SLI = bad?Quote:
unlike Skulltrail
I hope intel will convince nvidia to open up SLI. it looks like the 4870x2 will sell great if 4870 Xfire numbers are anything to go by.
Pardon me, but if you are buying DUAL GPU's, which currently you buy a 200-400 mobo for, are choking on a $30 cost?
I was asked to go back and look at where it is installed on Skulltrail. It's the nV-100 between the 61X slot and the NB.
http://techreport.com/articles.x/14052/3
nVidia 200 bridge chip thusly upgraded to PCI-E 2.0, fits right between the link to the PCH and the X16 slot on X58. Hell, it downgrades Skulltrail to PCI-E 1.1.Quote:
The bit of the diagram we're concerned with is the top left portion, where the north bridge (or MCH) connects to the PCIe x16 slots. In between the MCH and each pair of PCIe x16 graphics slots sits an nForce 100 PCI Express switch chip from Nvidia. Each nForce 100 has 16 lanes routed to each PCIe x16 slot and 16 lanes back to the MCH. This is by no means the most optimal configuration, because Nvidia's switch chip only supports PCIe version 1.1, whereas the Seaburg MCH supports PCIe 2.0, with twice the bandwidth per lane. The more optimal configuration would be to route eight lanes of PCIe 2.0 connectivity directly from the MCH to each of the four PCIe x16 slots. The Seaburg MCH is designed to work this way, in fact, and such a configuration would yield fewer chips, fewer hops between chips, less complex trace routing, lower power consumption, lower costs, and twice the total bandwidth back to the MCH (via 32 lanes of PCIe 2.0 instead of 32 lanes of PCIe 1.1).
Skulltrail is PCI-E 1.1 because it's a server chipset.
the thing that killed Skulltrail for gaming is the fb dimms. a gaming machine doesn't need that.
True but we don't know if there is a variation in the quality of the CF or SLI experience based on the chipset being used. So, again, saying it's "possible" doesn't necessarily mean it's the same. You might be right, don't take that totally the wrong way. But someone might as well say you could build a Mercedes for $10k just because a car can be built for 10k :)Quote:
No but there are decent $150 boards that support Crossfire already
Fair enough the 100 chip downgrades pci-e to 1.0. Not that we have any use for that bandwidth but you are definitely correct.
Ok in conclusion (if there could even be such a thing) CF offers worse driver support but better in the following:
scaling
full support of Intel chipsets without a bridge
dual monitor support while in CF mode (was true of the X2, and I think it's true of normal CF)
Any other downsides?
I've had many dual cpu systems and the thing that's always held them back is the ram.. now I've got a quad core that blows those systems out of the water. the only way I'll go back to a dual cpu setup is if it can use unbuffered ram in all the slots. That extra cost of the dual cpu setups isn't worth it I'd rather use a quad cpu setup. besides in a year a single cpu setup will smoke Skulltrail.
I'm sorry, but this seems to have gotten lost, or am I missing something :
Nvidia got a license for Nehalem-based chipsets? :oQuote:
Industry sources indicate that NVIDIA has acquired a QPI license to manufacture desktop core-logic (chipset) for the upcoming Nehalem-based processors. NVIDIA looks at a quick-fix solution to offer SLI support for the upcoming LGA-1366 motherboards. It's too late for them to prepare a full-fledged chipset since X58 is dressing up to go to office and that's slated for as early as Q4, 2008.
Or is that bs
That is the part that always digs at using CF or SLI. In a year or more, but less than two, you'll have a single gpu card that will put your CF/SLI to shame.