Does NVIDIA know something about Bulldozer?

No, they initially wanted "rigorous and extensive" testing involving sending sample boards to them to "ensure" compatibility with SLI, which drives up the cost of the boards. Boards with NF200 chips were automatically given SLI, but those that weren't, had to be individually sent in. Then nVidia would give them a special code to put into the bios that would allow the drivers to allow SLI. That was before the 5 USD per board thing.

because there wasn't enough bandwidth for tri and quad setups. That came to a end with Intel bringing out QPI interconnect on their chipsets where nvidia had no option but to issue it without the nf200. Intel found a way to increase the bandwidth without the need of the chip. Intel proved it they have control over it with the x58 P55 mobos. Restricting the P55 where lucid hydra or the nf200 is need for quad or tri setups but clearly showed with the qpi which is basically the dmi connection with increased lanes. Boards weren't put thru rigorous testing. Anyone could see with the scaling quad and tri setups was out of the question without a duplexer. The memory controllers was still on the chipsets aswell. Pci-e devices make most of their read and writes directly to the system ram.
 
Good riddance to NVIDIA making shitty chipsets for both Intel and AMD, the last one I owned gave me numerous headaches, waste of time.

NVIDIA should have just cut their losses and licensed SLI the same time to AMD when X58 came out. Better late than never I guess.
 
Yep, all Nvidia chipsets for AMD are garbage.

Have the Asus board in sig and on my sons rig and love it. Both boards have been Rock solid and not a bad Oc on my 1075 but still working on it :p

Glad to see we will be able to use SLi on BD
 
because there wasn't enough bandwidth for tri and quad setups. That came to a end with Intel bringing out QPI interconnect on their chipsets where nvidia had no option but to issue it without the nf200. Intel found a way to increase the bandwidth without the need of the chip. Intel proved it they have control over it with the x58 P55 mobos. Restricting the P55 where lucid hydra or the nf200 is need for quad or tri setups but clearly showed with the qpi which is basically the dmi connection with increased lanes. Boards weren't put thru rigorous testing. Anyone could see with the scaling quad and tri setups was out of the question without a duplexer. The memory controllers was still on the chipsets aswell. Pci-e devices make most of their read and writes directly to the system ram.

It was never about bandwidth, it was all about nVidia wanting more money for certifying boards for SLI. Even if it wasn't tri-SLI capable (which x58 was completely capable of from the beginning), they never certified it for even regular SLI.

Starting with x58 and P55, the memory controllers were moved onto the CPU. Only the older LGA775 CPU's had memory controllers that were on the chipset. Why don't you go troll your misinformation somewhere else?
 
Good riddance to NVIDIA making shitty chipsets for both Intel and AMD, the last one I owned gave me numerous headaches, waste of time.

NVIDIA should have just cut their losses and licensed SLI the same time to AMD when X58 came out. Better late than never I guess.

They probably figured that sales were so low (since who would go dual chip on a Phenom setup and use Nvidia) that a board maker might specialty board with NF200 to eat up the outliers. Or they figured people wouldn't see past the re-branded unsupportedness of the 980a SLI, and use those.
 
because there wasn't enough bandwidth for tri and quad setups. That came to a end with Intel bringing out QPI interconnect on their chipsets where nvidia had no option but to issue it without the nf200. Intel found a way to increase the bandwidth without the need of the chip. Intel proved it they have control over it with the x58 P55 mobos. Restricting the P55 where lucid hydra or the nf200 is need for quad or tri setups but clearly showed with the qpi which is basically the dmi connection with increased lanes. Boards weren't put thru rigorous testing. Anyone could see with the scaling quad and tri setups was out of the question without a duplexer. The memory controllers was still on the chipsets aswell. Pci-e devices make most of their read and writes directly to the system ram.
What?
 
LOL, the comments are openly bashing nVidia without even being nasty! :p Not even a single nVidia defendor of note on the entire comments roll, so far!!

Definately hilarious to read. The sad part is, I think even nVidia fans know most of the comments have at least some partial truth to them. Like 3x6970 > 3x580 in performance for a cheaper price according to hardocp.com, etc etc.

The only comeback a loyal nVidia fan could make might be 'Yeah well, I enjoy playing games from 2 years ago and am still loving PhysX on Batman:AA.' even then people have to wonder why you'd still be playing a single player experience from almost ...2? years ago that takes all of 12 hours to complete on each difficulty level.
 
Yep and since Batman everyone uses PhysX :) because that game sold so many copies and all the Nividia customers bought it twice ;)
 
This has been covered probably in every thread about the SLI licensing. AMD isn't paying a dime. The board partners are.

Doesn't really matter who's paying for it as long as Nvidia is getting paid.
 
Doesn't really matter who's paying for it as long as Nvidia is getting paid.

oh but it does ;P maybe we'll have some "non-SLI" boards that can do sli? AKA the DFI Lanparty" for 939cpus :p?

where as if AMD was paying, then all the boards would go up in price.
 
My board certainly is a "non-SLI" board, and I'm using SLI on it.

We will probably not see any non-SLI certified board versions if there is a version that is SLI certified, because the non-SLI certified ones will cannibalize the sales of SLI certified ones. If you look at all current x58 boards being sold, they're all SLI certified.
 
Back
Top