Core i7 vs. Phenom II X2 vs. X4 scaling performance analysis
Test Configuration
Test Configuration – Intel Hardware
- Intel Core i7 920 reference 2.66 GHz, underclocked to 2.6 GHz and overclocked to 3.2 GHz and 3.6 GHz; Turbo is off.
- Gigabyte GA-MA790X-UD4P (latest BIOS, PCIe 2.0 specification; CrossFire 8x+8x).
- 6 GB OCZ DDR3 PC 1800 Kingston RAM (3×2 GB, tri-channel at PC 1600 speeds; 2×2 GB supplied by Kingston)
- GeForce GTX 480, 1.5 GB reference design and clocks, supplied by Nvidia
- ATI Radeon HD 5870 (1GB, reference clocks, 850/1200 MHz) by Diamond
- ATi Radeon HD 5870 (1GB, underclocked from factory to reference clocks, 850/1200 MHz) supplied by AMD/PowerColor
- Onboard Realtek Audio
- Two identical 250 GB Seagate Barracuda 7200.10 hard drives configured and set up identically from drive image; one partition for Nvidia GeForce drivers and one for ATI Catalyst drivers
- Thermaltake ToughPower 775 W power supply unit supplied by Thermaltake
- Thermaltake Element G Case supplied by Thermaltake
- Noctua NH-U12P SE2 CPU cooler, supplied by Noctua
- Philips DVD SATA writer
- HP LP3065 2560×1600 thirty inch LCD
Test Configuration – AMD Hardware
- Phenom II 550 X2 overclocked from 3.1 to 3.2 GHz, underclocked to 2.6 GHz and overclocked to 3.6 GHz
- Phenom II 955 X4 at 3.2 GHz, underclocked to 2.6 GHz and overclocked to 3.6 GHz
- ECS A-A890 GXM-A, supplied by ECS (latest BIOS, PCIe 2.0 specification; CrossFire 8x+8x)
- 2×2 GB Kingston DDR3, supplied by Kingston (dual-channel at PC 1600 speeds)
- GeForce GTX 480, 1.5 GB reference design and clocks, supplied by Nvidia
- ATI Radeon HD 5870 (1GB, overclocked clocks, 850/1200 MHz) by Diamond
- ATi Radeon HD 5870 (1GB, reference clocks, 850/1200 MHz) supplied by AMD/PowerColor
- Onboard Realtek Audio
- Two identical 250 GB Seagate Barracuda 7200.10 hard drives configured and set up identically from drive image; one partition for Nvidia GeForce drivers and one for ATI Catalyst drivers
- Thermaltake ToughPower 775 W power supply unit supplied by Thermaltake
- Thermaltake Element G Case supplied by Thermaltake
- Noctua NH-U12P SE2 CPU cooler, supplied by Noctua
- Philips DVD SATA writer
- HP LP3065 2560×1600 thirty inch LCD
Test Configuration – Software
- ATi Catalyst 10-6; highest quality mip-mapping set in the driver, Catalyst AI set to “Standard”
- NVIDIA GeForce 257.21 WHQL drivers; High Quality
- Windows 7 64-bit; very latest updates
- DirectX May 2010
- All games are patched to their latest versions.
- vsync is forced off in the control panels.
- 4x or 8xAA enabled as noted in games and “forced” in each vendor’s control panel in UT3 ( Catalyst Control Center sets AA for Radeons in Batman: Arkham Asylum and it is set in-game for the GTX 480); all in-game settings are specified with 16xAF always applied; 16xAF forced in control panel for Crysis.
- All results show average, minimum and maximum frame rates except as noted.
- Highest quality sound (stereo) used in all games.
- Windows 7 64, all DX10 titles were run under DX10 render paths; DX11 titles under DX11 render paths.
The Benchmarks
- Vantage
- F.E.A.R.
- Enemy Territory: Quake Wars
- Unreal Tournament 3
- Call of Duty 4
- X3:Terran Conflict
- World-in-Conflict
- Left4Dead
- Batman: Arkham Asylum
- Call of Juarez
- Lost Planet
- Crysis
- Far Cry 2
- Resident Evil 5
- Serious Sam, Second Encounter HD (2010)
- Just Cause 2
- H.A.W.X.
- BattleForge
- Alien vs. Predator
- Dirt 2
- STALKER, Call of Pripyat
- Heaven 2.1
Flat-out amazing!!! I’ve never seen anything so epic like this. So, there was not any microstuttering in any of the above games, where the “measured” 40 fps appeared to look more like “perceived” 25 fps? I guess microstuttering is not noticeable if the measured fps is above 70-80, since half of this (45 fps) would still appear to be relatively smooth. Anything below 30 fps becomes really noticeable, so were there ever 40-50 fps instances with 2x 5870 CF that felt like 25fps or so? I’ll take your word for it, if you were actually watching 10,000 hours worth of benchmarking, ha (just kidding, don’t shoot my head off)!
Surprised there are not more comments. For a single video card it does look like a dual core is more than enough. Sure there are a few games that take advantage of four cores, but the fact remains they remain in the minority.
Bobert, I’ve been saying the same thing for months now.
Wow, very thorough and detailed article. It’s one thing to test CPU performance using a single video card, but it must take some brawn to do it for three different video configurations.
There’s so much data here to look at in so many ways. I suppose if you would have included an SLI setup we would then be able to determine how CPU speed affects SLI vs. Crossfire. If I’m looking at this data right, though, it seems Crossfire sees benefits from quad cores more than single video cards do.
Far Cry 2 seems to be a good example of this. Also Far Cry 2 shows interesting relationships between CPU and the GTX 480. The single HD 5870 doesn’t really react to CPU speed and cores the same way dual 5870s and the GTX 480 do. That’s pretty interesting.
So well done. If I only had one suggestion is that I would like to see GTA4 benched, mainly because I own it and good, thorough, and updated benchmarks of it are not easy to come by.:)
Concerning the Far Cry 2 numbers, despite being beaten with faster processors, the HD 5870 paired with the 2.6 dual core is actually outperforming the Crossfire and Nvidia setup. That’s what I find a bit interesting here.
AWESOME REVIEW.
This is EXACTLY what review websites ARE NOT putting out.
A non GPU-bottlenecked review showing how i7 really does have a significant gaming lead over Phenom II.
God you’d be surprised how many AMD fanboys still believe (and spread rumors) that Phenom II is plenty for 5870 crossfire. Psh. Plenty on today’s games maybe, but that is due to the PC gaming community being SNARED by the noob console community and their half a decade old setups.
Ok I’m ranting.
Two Thumbs!!
Raidur means it shows how Phenom II bottlenecks 5870 crossfire.
Everyone knows i7 is faster in games.
PS. I’m not Raidur.
PSS. I’m Raidur.
PSSS. Or am I?
Awesome review GJ.
Please let me know if you’re looking for a article writer for your
site. You have some really good articles and I believe I
would be a good asset. If you ever want to take some of the load off, I’d love to
write some material for your blog in exchange for a link back
to mine. Please shoot me an email if interested.
Thank you!