P55 Triple-GPU CrossFireX, Fixed?
Does Nvidia’s nForce 200 PCIe bridge solve the CrossFireX performance problem plaguing previous P55 platforms? You bet.

MSI’s Big Bang–Trinergy fell less than 2% behind the venerable X58, a performance difference that can easily be compensated by the fact that LGA 1156-based processors typically overclock a bit better than their LGA 1366-based predecessors. Non-gamers will make non-gaming arguments, such as the future availability of hexacore processors for X58-based platforms, the X58’s additional support for high-bandwidth devices such as professional RAID controllers, or the LGA 1366 interface’s ability to support triple-channel memory that will probably only noticeably benefit (drum roll, please) hexacore processors. But while those are all valid reasons for choosing a more flexible and less overclockable X58 option, anyone who wants the ultimate gaming experience should already know where to use the extra CPU speed that's usually available with Lynnfield-based Core i5 and Core i7 processors.
Asus fans will note that the P7P55 WS Supercomputer preceded MSI's Trinergy and uses the same NF200 connection method to supply four, rather than three, x16-length slots. The identical connection method should yield nearly identical performance results, but Asus' slot layout is more favorable to placing three double-slot cards into a standard ATX case.
EVGA’s P55 Classified 200 gave up another 0.4% to gain an extra PCIe slot. We’re not quite sure when Nvidia or ATI will develop five-way GPU arrays, but EVGA will be on the leading edge when it happens.
We also took a brief look at the performance benefit of applying this same principle to the X58 chipset to allow 48 lanes of graphics power over 32 lanes of connectivity. We have, unfortunately, not yet seen a graphics card configuration and game that require that much bandwidth, leaving the MSI Eclipse Plus at the orphanage once again. Yet good news for current owners of cheaper X58 motherboards was found in the same test, as the x16/x16/x4 slot configuration of Asus’ lower-cost P6T gave up only around 4% in performance compared to its x16/x8/x8 competition.
Latest Motherboards News
- 25/05 – Thunderbolt Will Become Key Motherboard Spec in 2H 2012
- 22/05 – Does Your AMD FX Platform BSOD with Steam? Read This.
- 04/05 – 54 Million Enthusiast and Performance PC Gamers Globally
- 01/05 – Intel's Next Unit of Computing Rivals Raspberry Pi in Size
- 10/04 – Intel P67 Express Chipset Begins Product Discontinuance Cycle
Latest Motherboards reviews
- 30/04 – Six Sub-£160 Z77 Motherboards, Benchmarked And Reviewed
- 23/04 – Video Teaser: Intel's DZ77GA-70K Motherboard
- 17/04 – Overclocking: Asus Rampage IV Extreme Versus EVGA X79 FTW
- 08/04 – Intel’s Z77 Express And Lucidlogix MVP: New Features...
- 27/02 – Six £165-£215 LGA 2011 Motherboards, Reviewed
looks like we need new Crysis... getting over 100fps in games these days seems way too easy...
What effect does increasing the PCI-E clock from 100mhz to 200mhz on the P55 platform have? Would doubling the clock rate make up for the half bandwidth 1.1 interface over a 2.o interface?
What effect does increasing the PCI-E clock from 100mhz to 200mhz on the P55 platform have? Would doubling the clock rate make up for the half bandwidth 1.1 interface over a 2.o interface?
More likely it'd throw the graphics cards completely and not boot. you'll struggle beyond 120MHz, and even then the gains would be minimal.
Although an interesting read, I'm really not sure why motherboard manufacturers are investing so much time in "fixing" the P55's lack of PCI-E lanes.
Right now, the only thing that requires all these lanes is multiple GPU setups, and if you're a gamer who demands that level of GPU power then you obviously have the cash to burn and go top-end with X58 anyway.
It just seems to me that mobo companies are trying to fix a problem for a market that doesn't exist.
Work in progress, release then try fix is the new way for all devices and games alike. Good read was more interested in the small gains by adding a 3rd card, hopefully if, when they get 16,16,16 or 32/32/32 oh ye what a dream, will be better results, but for now it shows how limited the boards are at present.
Then again we have always been paying usualy double for those very small gains in the past.