Graphics Settings For Each Of Our Tests
Assassin’s Creed: All settings are set to maximum. Anti-aliasing and anisotropic filtering are applied via the .ini file. The game runs with DirectX 10. DirectX 9 cards switch it into DirectX 9 mode, which means that older cards may be quicker. The Radeon X800 with Shader Model 2.0 cannot start Assassin’s Creed on the test computer.
Call of Duty 4: All settings are set to maximum. The switch for Dual Video Cards is activated on the SLI and CrossFire configurations.
Crysis: The tests were carried out at High and Very High quality. Anti-aliasing was activated in the game, and anisotropic filtering via the graphics driver. All DirectX 9 graphics chips are unable to switch into the Very High mode, which is only interesting with DirectX 10 anyway—there the water reflections and HDR rendering of the sun can really be seen.
Enemy Territory Quake Wars: The quality is set to High, lighting remains at normal, and shadows and soft particles are activated. The older AMD cards with DirectX 9 have no soft particles. This function is missing in the graphics menu, and its absence improves the frame rates. Why the function is missing is unclear, since the GeForce 6 can display them with OpenGL 1.5, wheres the X1000 series from AMD is specified at OpenGL Version 2.0.
Half Life 2: Episode 2: All settings are set to maximum; Reflect all, HDR rendering and Motion blur are activated.
Mass Effect: All settings are set to maximum. Anisotropic filtering is handled by the game, while anti-aliasing is activated via the graphics driver. We tested 4xAA and 8xAA—the older AMD cards do not offer 8xAA and it is only possible to get more than 6xAA using CrossFire. No custom filters are used for anti-aliasing, only default AA is evaluated.
To use anti-aliasing with the Radeon X1000, HD 2000 or HD 3000, MassEffect.exe was renamed to Bioshock.exe. The Nvidia GeForce 7 with the UT3 Engine does not support HDR rendering with simultaneous anti aliasing. The game runs with DirectX 10, while DirectX 9 cards switch it into DirectX 9 mode, which means that older cards may be quicker. The Radeon X800 with Shader Model 2.0 cannot start Mass Effect on the test system.
Microsoft Flight X SP2: All settings are set to Ultra High. The frame limiter is always deactivated, and additional improvements to graphics and illumination are activated. Anti-aliasing and anisotropic filtering are selected via the graphics menu in the game. We had a little trouble with the X1000 series of AMD cards, but the graphics driver was able to force anti-aliasing. With the DirectX 9 cards, the DirectX 10 preview is not available, and as a result older cards have a greater speed advantage.
World in Conflict: All settings are set to Very High. The Very High setting was also used on 128 and 256 MB graphics cards even when the game recommends selecting a lower quality. Anti-aliasing and anisotropic filtering are set via the game, and since there is no 8xAF option, we used 4xAF. The game runs with DirectX 10; older DirectX-9 cards switch it into DirectX 9 mode, which means that older cards may be quicker.
3DMark06: The synthetic DirectX 9 benchmark runs at 1280x1024 pixels in default quality without filters. The CPU score with the Intel X38 chipset lies between 2521 and 2555. With the nforce 780i chipset, 3DMark06 achieved a CPU score of 2372 to 2427.
Latest Graphics Cards News
- 20/05 – Qualcomm Hires Former AMD CTO Demers
- 18/05 – Nvidia Responds to GeForce 600 Series V-Sync Stuttering Issue
- 18/05 – Behold: Here Are The First Unreal Engine 4 Screens
- 17/05 – Nvidia Debuts GK110-based 7.1 Billion Transistor Super GPU
- 17/05 – GE Announces Kepler Graphics Card for Military and Aviation
Latest Graphics Cards reviews
- 18/05 – Best Gaming CPUs For The Money: May 2012
- 10/05 – Best Graphics Cards For The Money: May 2012
- 10/05 – GeForce GTX 670 2 GB Review: Is It Already Time To Forget GTX...
- 09/05 – Video Teaser: Radeon HD 7800 Series Remains A Solid Value
- 08/05 – Benchmarking AMD's 768-Shader Pitcairn: Not For Public...
















Page 5?
Nothin pointing out the bleedin obvious - mind you I guess you are in a rush to post inanely on every other article that Toms produces

The 110,000w figure is completely useless. kw/h means something, amps kinda means something. Watts just means you added all the peak loads of each test to come up with an impressive sounding number... I doubt you actually drew 110kw from the grid in one go - I doubt Toms has enough plug sockets
wow stats!
Is there any point???
mi1ez, surely you know by now they only have 4 page articles on thw.co.uk
Should we all take bets as to how long it'll take the .co.uk site to fix the problem with this article...
(just heading over to the .com site now, I'm assuming it'll be ok there.)
***PAGE 5 WOULD OF BEEN NICE !!****
Where is the ATI Radeon HD 4870x2 ?
The reason why the Radeon HD4870x2 isn’t listed in this test and the reason why it doesn’t appear on the 2008 Q3 graphic card charts at the top (where it should be!) although the HD4870x2 WAS LAUNCHED IN Q3 THW!! is that the THW team have always had their heads shoved firmly up Nvidia’s dark and scary place! Hopefully one day the THW team will pull their collective heads out, wipe the chicken crap’e from their faces, smell the clean air and start publishing completely unbiased reviews, benchmarks and quarterly charts! Anyone noticed the THW teams almost puke worthy lame excuse for not listing the best PCIE Card for Over $350 in September? Could the real reason be that an ATI card would be sitting at the top instead of Nvidia? THW has been avoiding naming the HD4870x2 as king like the plague holding out for Nvidia to launch an HD4870x2 killer. Come on THW really pathetically obvious!
Why can't i see the test results?? May be any page lost???
Just searching for some info with toms' own search function,
brought up this reference. So where is page 5 and all the rest?
Ian.