Core i7: 4-Way CrossFire, 3-way SLI, Paradise?

X58 And SLI Get Busy

The use of Intel’s X58 chipset does not automatically imply SLI compatibility. In order for a motherboard to support SLI, it must first be certified by Nvidia—a process that reportedly involves stability, power, compatibility, and layout testing. If a prospective platform passes those four metrics, it is technically eligible for an SLI license.

We’re not sure what Nvidia is charging for a license, but that price will most definitely be passed on through the board vendors. Then again, with prices on X58-based boards expected to hover close to the $300 mark, we have to figure that the relative addition with SLI licensing will be small versus motherboards without SLI support.

Nvidia's preliminary list of certified SLI X58 motherboards

According to Nvidia, the validation process to make sure an X58 platform works properly with SLI is fairly extensive. Motherboards are tested with two-way, 3-way, and quad-SLI setups and a number of different third-party devices. The boards are then run through the top 15 titles for sales, the top 15 used for benchmarking, titles with PhysX support, and a barrage of video performance measurements. The mechanical specs of each candidate are tested. They’re then subjected to a stress test and PCI Express-specification verification.

This information is likely very similar to what AMD does in certifying boards for CrossFireX compatibility as well. It’s only newsworthy now because Nvidia only recently opened the doors to licensing, while we’ve had CrossFire compatibility on Intel chipsets for some time.

Naturally, if SLI support is one of your qualifications for buying a new X58-based board, you’ll want to keep a sharp eye out to make sure the feature is included. It’s worth noting that Intel’s DX58SO motherboard does not support SLI. We asked Intel to comment on the situation and heard back that the two companies had not found mutually-acceptable business terms for the certification.

Getting Your Slots Straight

There are a grand total of six different configurations you may or may not see motherboard vendors use as they add SLI support to their products. Four are considered native, and rely on X58’s twin x16 links of PCI Express 2.0 connectivity. The other two involve Nvidia’s nForce 200 companion chip.

The most basic implementation of X58 involves two x16 slots, each running at x16 signaling rates. Naturally, an SLI-certified board with this setup maxes out with a pair of GeForce GTX 280s. That’s certainly no slouch of a configuration, but most of the motherboards we’ve seen strive for more. As an aside, a board armed with what Nvidia calls a native two-slot setup will also support a pair of Radeon HD 4870 X2s. So this is hardly a constraining arrangement.

A native three-slot configuration is fancier, but it’s still possible using nothing but Intel’s X58 core logic. One of the chipset’s x16 links remains x16 electrically, but the second one is split into a pair of x8 links connected to x16 slots. The catch, according to Nvidia, is that those x8 connections don’t give you the full bandwidth of a x16 slot. However, we’re talking about PCI Express 2.0 here—we haven’t seen any evidence yet to suggest today’s cards need more throughput than that connection can provide (though we do have a PCI Express scaling story in the works, which should shed additional light on this matter). Using the three-slot setup, certified boards are able to run 3-way SLI—a claim we confirm and explore in this story.

We haven’t seen any motherboards with this slot arrangement yet, but Nvidia mentions a native four-slot mode enabled by dividing both of X58’s x16 links in half. This would be the way to go if you wanted to run 3-way SLI with a fourth GPU dedicated to PhysX acceleration. We had enough trouble running 3-way SLI, though. Heat is not this configuration’s friend, and the standard fan control simply doesn’t compensate for the fact that a trio of boards is cranking out fiery temperatures. Throwing a fourth card into the mix sounds like a recipe for frustration.

The fifth possibility involves X58 with Nvidia’s nForce 200 chip. Intel’s Tylersburg (X58’s internal designation) maintains one of its x16 links. The other is hooked up to nForce 200, which in turn creates two more x16 links of PCI Express 2.0. Nvidia says that the use of broadcast and peer-to-peer writes gives this configuration an advantage, even though it ultimately has two cards hooked to the same x16 link as a native arrangement. Purportedly, because the boards attached to the nForce 200 can communicate directly, potential for better performance exists. In our talks with motherboard vendors, though, none seemed particularly enthusiastic about integrating nForce 200.

Finally, we may see workstation or server-class motherboards incorporating two nForce 200 chips—one for each of X58’s links and designed to facilitate maximum throughput between all of the attached cards in a four-way setup. This would be the way to go for a machine loaded down with Tesla boards, Nvidia tells us.

Create a new thread in the US Reviews comments forum about this subject
This thread is closed for comments
Comment from the forums
    Your comment
  • randomizer
    SLI scales so nicely on X58.
  • DFGum
    Yep, i hafta say being able to switch brands of graphics cards on a whim and selling off the old is great. Knowing im going to be getting the preformance these cards are capable of (better price to preformance ratio) is nice also.
  • cangelini
    randomizerSLI scales so nicely on X58.

    Hey you even got a "First" in there Randomizer!
  • randomizer
    cangeliniHey you even got a "First" in there Randomizer!

    And modest old me didn't even mention it. :lol:
  • enewmen
    Still waiting for the 4870 X2s to be used in these bechmarks. I thought THG got a couple for the $4500 exteme system. But still happy to see articles like this so early!
  • cangelini
    enewmenStill waiting for the 4870 X2s to be used in these bechmarks. I thought THG got a couple for the $4500 exteme system. But still happy to see articles like this so early!

    Go check out the benchmark pages man! Every one with 1, 2, 4 4870s. The 2x and 4x configs are achieved with X2s, too.

    Oh, and latest drivers all around, too. Crazy, I know! =)
  • enewmen

    I found it, just read the article too quickly. - My bad.
    "A single Radeon HD 4870 X2—representing our 2 x Radeon HD 4870 scores—is similarly capable of scaling fan speed on its own. "
    Hope to see driver updates like you said.
  • spyde
    Hi there, my question regarding these benchmarks with the HD card is, "was a 2G card use or a 1G". I am about to buy a new system and was looking to buy 2 x HD4870X2 2G cards, but with these results its looking a bit ify. I hope you can answer my question.
  • Proximon
    That's a nice article. I especially like the way the graphs are done. everything is scaled right, and you get an accurate representation.
  • cangelini
    These are 2GB cards =)
  • spyde
  • Tjik
    How sure can we be that the difference between a nVidia and an AMD setup isn't related to the motherboard design? From the figures I would make the conclusion that the AMD + AMD setup is able to overcome some of the disadvantages of a weaker CPU, and in several cases there's no obvious - at least to me - reason why a Core i7 with the same single or dual set of AMD graphic cards would perform worse. It's easy to blame it on driver issues, but what proof is there to make that a more plausible conclusion? I'm not into some kind of weird conspiracy theory, I'm just technically curious to know why we should assume the X58 platform to be flawless when figures suggest differently.

    The conclusion I draw from this and other tests made is that Core i7 is great, but you need to spend big money on graphic cards to make it a gamers choice, or put it into a game performance per money perspective. As it is now, before future releases of mid range CPUs, or if AMD unexpectedly release some scary monster, I foremost see Core i7 as a solid solution for serious work. In rendering and other CPU dependent tasks it might be a blessing to cut some 40 % of the time to process.

    Another observation is that if the current scenario doesn't change in the near future we could well be back to old school over-clocking culture, when money and availability set the limits. I'm not against but in the last years we've seen more of a yuppie's over-clocker culture, where money and availability haven't been an issue. To be frank, what we have here is two ways of making priorities: one option is an AMD system which gives you a 790FX motherboard + CPU + RAM for the same price as a single Intel Core i7, and if you're not planning to play at resolutions higher than 1600x1200 and probably not buy anything above a possible single X2 Graphic Card it could well be the best offer for the money.

    Options are good and even with AMD well behind it opens up for many different choices. Some never really use but enjoy knowing they got a monster system, others only buy exactly what gives best value for money, some specialize systems for tasks with a cost conscious approach, and some don't have a clue. Every choice is good as long as the user is happy (and spendings doesn't hurt the family economy).

    Oh, a lot of text there. In conclusion I'm more interested in whether the X58 platform is ill suited (or less good) for AMD graphic cards at the moment, and for proof either way.
  • Anonymous
    hi guys,

    nice @ first : )

    but now i got one big question about this review

    on page 12 you got a nice overview about the 3d mark benchmark

    what i don't understand is the CPU score of the i7 and c2q ex based on nvidia and ati graphic boards

    there is such a huge difference of the cpu score just because of changing graphic boards ???

    how can that be?

    i mean the cpu score is based on the cpu right? or does futuremark test
    something else with this cpu score than just the cpu itself?

    i don't get to see the reason, why just changing from nvidia to ati or otherwise could have such huge effect on that cpu score

    but maybe one of you could explain it to me, otherwise i could think there is something wrong with this or maybe all of these benchmarks in this review

    thx in advance : )
  • arkadi
    well finally we getting more and more info about the i7, i guess it will take us few weeks 2 get it all right.I just love it when new staff comes out :)
  • aznguy0028
    "there is such a huge difference of the cpu score just because of changing graphic boards ???"

    it was stated in the article on that page, at the top. the default run with PhysX artificially inflates the scores.
  • z999
    lolz, there are some crazy problems w/ the ATI cards... like turning on AA and gaining 50FPS...
    You should do this benchmark again in a couple of months when the drivers give more accurate results, and in that one..... lose the phenom :P.
  • chaohsiangchen
    Nice job well done! Thanks for the article.

    However, I didn't see specified CPU clocks, so I presume that all three CPUs tested were run at stock speed. Although I have little doubt that Phenom X4 will still lose to both Ci7 965EE and QX9770, it's just my curiocity to see how Phenom X4 at 3.2GHz would perform.
  • zodiacfml
    I thought there will be little difference between core 2 and i7 in games, so it was just the graphics card that is holding i7 back in games. in my mind, i thought a single gtx 280 was held back by the fastest core2, wasting graphics power which was not the case here.

    i7 is really a cool system, much like the intel SSD. intel is on fire. :)
  • bunnyblaster
    I've been an avid Toms Hardware reader for over 8 years now. This is my first post. It is an issue that has been nagging me the last few updates for Tomshardware. The "page scrolling function" on each page is terribly designed. I find it slow, disappears before I select the page and sometimes does not register at all. I have multiple computers and I draw the same conclusion on all of them. It gets rather annoying when I just want to read some implications or conclusions of some reviews and I can't get there easily without going through 27 pages.

    Why not just revert to the system everyone else uses with a simpler scroll-down bar?
  • fahdriyami
    What a surprise, no Flight Simulator X Benchmarks