SLI vs CrossFire: Which Dual-Card Setup is the Best?
SLI vs CrossFire: Which Dual-Card Setup is the Best?
Posted 08/25/2010 at 2:34pm
| by Loyd Case
26
Comments
Print
We look at dual card graphics and PC gaming. When you add a second card, which cards are faster? Which scale better? And which can you actually stand to have in your system?
Graphics cards have gotten faster and added more features. So we have to ask the question: is it really worth adding a second GPU to your system? Will you get enough of a performance boost to justify the extra power draw and added cost? The answer is more complex than a simple yes or no. It all depends on what games you’re running, how much you dial up features like anti-aliasing, whether you’ve dived into the world of stereoscopic 3D and what monitor you’re running.
Perhaps the most important factor in the decision is display resolution. If you’re running a 1680x1050, 22-inch display, a single midrange or high end card will get the job done. Adding a second GPU is overkill. If you’ve got a 30-inch, 2560x1600 display and want to crank up the AA and postprocessing features, then that second GPU can be a big help.
Stereoscopic 3D, like Nvidia’s 3D Vision, demands more performance as well, since you’re effectively doubling the frame rate requirements of a game. Most 3D displays currently available max out at 1920x1080, however, so the performance demands aren’t overly onerous.
Games themselves are evolving and adding more features. You can see that in a title like Just Cause 2. Still a DirectX 10 title (and it requires DX10 or higher), Just Cause 2 adds a host of postprocessing effects that can demand much from your graphics card. Toss in DX11 titles, like Aliens versus Predator or STALKER: Call of Pripyat, and frame rates can plummet as you add features. So that second GPU can indeed contribute to the overall experience.
With these thoughts in mind, let’s take a look at our SLI and CrossFire X candidates.
The Cards
We looked at seven different DirectX 11 capable cards, including four based on Nvidia GPUs and three using AMD graphics chips. We also tossed in the fastest single GPU card from the previous generation, Nvidia’s 285 GTX, in the form of a pair of eVGA GeForce 285 GTX SSC cards. Note that not all cards were identical. With the exception of the GTX 460 768MB, Radeon HD 5870 and GeForce GTX 480, all were run at reference speeds. The GTX 460s, HD 5870s and GTX 480s were factory overclocked by a few percentage points, as were the previous generation 285 GTX cards.
Here’s the lineup; note there were two of each, and we call out the models where appropriate.
- eVGA GeForce GTX 285 SSC
- A pair of Asus 768MB GTX 460
- One Gigabyte GTX 460 1GB card and one reference GTX 460 1GB
- One eVGA 470 GTX card, one Galaxy GTX 470
- A pair of eVGA 480 GTX SC cards
- One Radeon HD 5830 reference card, one XFX Radeon HD 5830
- One Asus Radeon HD 5850 and one XFX Radeon HD 5850
- One HIS Radeon HD 5870 Modern Warfare 2 Edition and one XFX Radeon HD 5870 XXX Edition. These are both 1GB, factory overclocked cards.
While the core and memory speeds of some of these cards may be higher than stock reference cards, it’s been our experience that actual game performance increases only slightly. So we can still make judgment calls on performance, even though all of these aren’t stock cards.
Our test bed is a 3.33GHz Core i7-975 Extreme Edition in an Asus P6X58D Premium motherboard with 6GB of DDR3/1333 and a Corsair TX850W PSU. The OS is 64-bit Windows Ultimate. All games were run at 1920x1200 with 4x AA. In point of fact, we ran at 2560x1600 with and without AA. But our tests were still pretty demanding at 1920x1200, with 4x AA and all eye candy cranked up. So those are the results we’re reporting here.
We installed the latest release drivers -- 258.96 WHQL for Nvidia and Catalyst 10.7 for AMD. We also made sure each card’s dual GPU feature was enabled. AMD will enable CrossFire X by default if it detects two cards, while you’ll have to manually turn on SLI with Nvidia cards.
Make sure you manually enable SLI in the Nvidia control panel
Let's dive right in.
DirectX 10 Gaming
First, let’s take a look at games and benchmarks that run on DirectX 10. This allows us to see how much we’ve gained since last generation. We’re using the fastest single GPU card from last generation, a (once) speedy eVGA GeForce 285 GTX SSC, factory overclocked to 702MHz (versus the stock 648MHz) core and 2646MHz memory (the stock memory clocks are 2484MHz.) Let’s see how the new kids on the block handle the now geriatric 285 GTX:
First up is the hoary old 3DMark Vantage, running the “extreme” test. The older 285 GTX just wins out over the newer Radeon HD 5830, but falls short of everything else. Note how even two GTX 460 768MB cards, with their relatively low memory bandwidth, still spanks the older card.
Crysis is still a demanding test, even after three years. Just when you begin to believe it’s CPU bound, tossing in a second, new generation card pumps up the frame rate. A pair of GTX 480s are the overall winner, just falling short of the magical 60fps. It’s impressive how the $230 cards – GTX 460s and Radeon HD 5830s – generally keep up with the once mighty 285 GTX.
Next up is a pair of Far Cry 2 benchmarks.
The longer Ranch benchmark was once a demanding GPU test, but it’s become pretty easy by today’s standards. Once again, AMD’s Radeon HD 5830 is the odd duck, but note how even the low cost 460 GTXs are essentially at performance parity with the pricier Radeons in this graphics intensive test.
Far Cry 2’s action benchmark is more indicative of actual gameplay, so the test is CPU limited on higher end cards – in this case, anything that’s not an older card (285 GTX) or memory limited (GTX 460 768MB). Note how tight the grouping is here – that’s because the benchmark throws of ton of physics around, plus numerous AI characters running around trying to kill you. We wouldn’t make any graphics card decisions based on this benchmark, but it’s worth noting that GPUs aren’t the entire ball of was in PC gaming.
Tom Clancy’s HAWX is an action flight sim; we used the DirectX 10 version in our testing. Once again, AMD’s Radeon HD 5830 is the weak sauce, illustrating what an odd duck AMD’s lower end midrange card really is. While the HD 5870 wins out over the two GTX 460 cards, it’s no match for the paired GTX 470s or 480s.
Just Cause 2 is one of the few titles that requires DirectX 10 as a minimum, as well as running only on Windows Vista or Windows 7. The Concrete Jungle test throws a lot of postprocessing effects at the card, and can hammer frame rate – and it’s not particularly CPU bound. Note that we disabled the Nvidia-specific Bokeh and water effects to keep the playing field even.
AMD stays in the game with Just Cause 2, at least with paired Radeon HD 5870s. The Radeon HD 5830 and 5850s don’t fare quite as well. Once again, the 5830 is the weak sister of the bunch.
Next Page: DirectX 11 Benchmarks...
More like this
jmanley001
November 05, 2010 at 7:40pm
(The opinions expressed by the reader are strictly that...opinion. Haters can get the bawleys..)
I don't really care which wins the GTX480 SLI or 5870 Xfire solution. I've owned both cards, and honestly I'm perfectly happy with my 5870. After two GTX 480's died after only 3 months (the original, and then the factory replacement) I'm pretty much fed up with Nvidia. Sure its a beast of a card, and will stomp all over the 5870. But price/performance, ATI's won this one. In fact, I'd nominate the 5870 as hardware of the year. I hardly notice that much of a difference in performance in any of my games, between the two cards. (I run everything at 1920 x 1200) Only thing that seems to lag is Unigine Heaven benchmark, and the lack of Gpu simulated water in Just Cause 2. Other than that, the 5870 kicks @ss. Not to mention, it runs a heck of a lot cooler, even overclocked...within reason of course.
Regardless, it comes down to personal preference. I've always been a big fan of Nvidia, but their latest offerings, left a bad taste in my mouth. Maybe the GTX 580 will change my mind....we'll see...
raguza123
October 30, 2010 at 5:15am
first of all, you must be idiots, all the cards here are similar, and you can run two different Nvidia cards in SLI mode, get out from behind your rig, I've done it, it will run at the slower one's mode, and as far as ATI, yuk, Nvidia rules alway will, has better drivers, they fine tune there drivers as newer games come out, ATI doesn't and ATI has too many driver issues, If ATI was so great how come they are chasing Nvidia not the other way around DUMMIES
robbtheskater
September 24, 2010 at 6:57am
"You guys are all nvidia fanboys", "these results are biased, they didn't even test the 5970" blah blah blah. Do a single one of you internet flaming morons who think you know everything notice that all those cards are single GPU graphics adapters? No, you're right; you didn't. So please, actually contribute in the damned comment section instead of flaming these tech reporters over nothing. They don't do this for their f***ing health. They actually like to see discussions and not bull**** in the comment section.....
Unless someone can prove to me that they have one of these cards with Dual GPU's then i believe I am right. And you are all morons. (the people who acted like morons that is)
And no, Im not a Nvidia fanboy. I have a 5870 and ATI's drivers suck. Not MaximumPC's fault.
Nycromes
September 03, 2010 at 4:13am
The 480 came out on top in almost every benchmark, not much to dispute there. But to say that the GTX 470 spanks the 5870 is stretching it a little bit. The cards are nearly neck and neck in many/most of your benchmarks and the 5870 lost big in one or two, excepting those tests, the differences aren't exactly as drastic as you make them out to be. That being said, the GTX 470 does come out on top in most of the matchups. Personally, I will stick to the fastest single GPU I can afford for my box.
elo231
August 29, 2010 at 8:21am
According to THG's article on the difference between SLI and CrossFire
They found out that in the different CrossFire modes with BFBC2 @ 1920x1200, 4x AA, 8x AF, HQ, DX10
No connectors: 46.9 FPS
1 connector on left: 82.7 FPS
1 connector on right: 82.9 FPS
Both on left and right: 83.0 FPS
Seriously that article is more detailed and shows the benefits as well as drawbacks into using a dual gpu setup.
http://www.tomshardware.com/reviews/amd-crossfire-nvidia-sli-multi-gpu,2678.html
jorera53
August 28, 2010 at 8:17pm
Your satisfaction is our eternal pursuit.
=========== :) http://www.jorer.com =============
We specialized in the exportation of sport shoes and other products(clothing, bag,sunglasses,watches,belts,etc )which have great enjoyed popularty in the world market Many of our goods are on sales ,we can guarantee the crediblity by Pay-pal and delivery time .we would like to make a long termship.
1)Name : The perfect gift.
2)Grade : AAA+
3)Package : in original boxes
4)Color : various
5)MOQ : 1 pc
6)Payment : Pay-pal/Western Union / Credit card/Moneygram
7)Ship-ping : 4-7 days with guarantee of customs clearance, drop shipp-ing is accepted to customers' demands
8)Who-lesale: very low price, you can make a small order first to test quality and service.
http://jorer.com
Online Contact . thank you.
sk8erboi101
August 28, 2010 at 5:12pm
typical nVidia fanboy. the benchmark test are obviously unfair. rounding up higend nvidia cards with "not-so-highend" ati cards
kr4st
August 28, 2010 at 6:39am
Hay Loyd, looks like you hacked off a few AMD gamers with your benchmarks. Ive never used any Crossfire setup but have used SLI for some time. My outdated 3GHz Core 2 Duo, NVIDIA MoBo and 2 8800 GTS SLI are still running all the DirectX 10 games I play on a 22 inch LCD monitor. Thanks Loyd for the info, readers who are satisfied with their gaming rigs need to chill out, and uses these benchmarks as they were intended.
INTEL and NVIDIA are still my choices for upgrade in the future.
How badly would a single GTX 470 spank my SLI 8800GTS on a 22 in monitor?
gwmckay
August 26, 2010 at 10:30am
The folks over at TomsHardware benchmark the 5970 Ares and the 480 still came out on top. The 480 is a beast, but the power draw and heat makes an SLI setup IMO not worth it.
crazitrain02
August 26, 2010 at 12:20pm
I've looked all over Tom's Hardware for a comparison between the 480 and the 5970, and most of the ones I've looked at have the 4870x2 beating out the 480.
sniggler
August 26, 2010 at 8:09am
but wait until I dial my xfire 5850's up to 1 GHz core / 1150 mem. We'll see who gets spanked then
crazitrain02
August 26, 2010 at 6:35am
If you're going to do a benchmark and tout to the world that you are an nVidia guy, cool, but at least have some respect and do a proper benchmark here. If you're going to use the top of the line nVidia card and then use ATI's 58xx series when there's the 59xx series available then you might as well compare apples to watermelons. If you were to make the headline correct either put a HUGE asterisk next to the title, or rename it to something else.
lou29
August 26, 2010 at 5:40am
Hey MPC can you please benchmark the 5970 I feel cold curled up in a corner Im a ATI guy put the 5970
to the test along with the best from NVIDIA please make us ATI fans feel better.
jonc1028
August 25, 2010 at 7:38pm
the power graph was a little confusing because it was supposed to be a graph of power consumption, but was labeled as FPS instead of Watts. Also there was no distinction between the blue bars and the red bars. From reading the short explanation below the graph i guessed they were PC at idle and PC at 100% load
Neckername
August 25, 2010 at 5:31pm
When I researched multi video card solutuions a while ago for one of my builds, I remember that Crossfire was the only one that could run multi monitors while nVidia SLI only supported one monitor. That has probably changed since then... I hope. I still hear from friends at LANs and on forum posts that there are some issues with multi monitor multi GPU configs. I would rather trade in two cards for a dual proc single pcb solution that eliminates the SLI connector and extra power plugs and power.
carnivorousslushee23
August 25, 2010 at 8:34pm
I'm gonna guess that they wanted to use only single-GPU cards for their benchmarks. Seeing as the 5970 is a dual-GPU card, it doesn't fit the specifications for this article.
crazitrain02
August 26, 2010 at 6:38am
A single card is always better anyways, plus we all know that the 5970 will spank a 480 in both power consumption and heat. nVidia has always been horrible with power and heat.
Nailer669
August 26, 2010 at 3:29am
I don't care how many GPUs are on a card. I only care that it is a single card. Why test the AMD X6 to a quad-core i7 if they follow this mantra? They say that 4 chips don't scale wel in Crossfire. Who cares? We still want to see the perfromance.
It's a conspiracy. They knew they'd get the question. They are saving it up for another article.... more articles less work. But i know nothing.... *wink wink nudge nudge*
americaeh
August 29, 2010 at 5:47pm
its not a conspiracy but i would like this test to include tri and quad crossfire and sli setups even though the performance increase would be minimum.
elo231
August 29, 2010 at 8:27am
If you want a apples to apples test then it will be a single HD5970 vs 2x GTX480 in SLI.
As the HD5970 is a dual GPU card so it's technically already in CrossFire.
B.A.Frayd
August 25, 2010 at 3:23pm
After reading many forum posts for various games, it seems like SLI And Crossfire have too many compatibility problems to be worth the money.
crazitrain02
August 26, 2010 at 6:44am
With ATI you can have two cards that are in the same Series. nVidia they have to be the same identical card with identical firmware or they have problems. I'm running two different 4870s, one Diamond 512MB and the other Sapphire Vapor-X 2GB with no issues at all. Technically you could run a 4830 and a 4890, and the Crossfire would just downclock the better card to play nice.
Talcum X
August 25, 2010 at 3:17pm
I don't konw when I will get the chance to upgrade (will be a complete rebuild)...will have to see what monitor I pick out. Will only be playing UT2K4 / III still anyways, so I guess 1 will still do, even on a 30". Depends on life at that point.
Connect with MaximumPC
Featured Content
We introduce Intel's latest class of tablet killers and review 4 of the first Ultra...
Where have all the memorable videogame enemies gone?
This month's issue
Feature
11 Hardware Hacks
Feature
Overclock Your CPU
How-To
Supercharge Your Smartphone
Build It
A Powerful $830 Rig
Most Commented Articles
Latest Max PC Tweets
- maximumpc: Analysts: 9-Inch Kindle Fire This Summer http://t.co/QDYApwCZ1 day 1 hour ago
- maximumpc: Micron: DRAM Prices Likely as Low as They're Going to Get http://t.co/fFUWuFOm1 day 4 hours ago
- maximumpc: Yar, Mateys! All Of Pirate Bay Made Available As A Single 90MB Zip File http://t.co/j5LHlXEZ1 day 6 hours ago
