Watercooled GeForce GTX 280 Showdown

Test Setup

Testing these cards was always going to be a little different to normal graphics testing thanks to the need for not only a full watercooling setup, but also the fact that it had to be fairly easy to swap out the cards between test runs. In the end we settled on a fairly conservative setup as watercooling goes – a single 120mm radiator fitted with two Noctua NH-12 120mm cooling fans in a push-pull configuration and a combined pump/reservoir from an old Gigabyte 3D Galaxy II watercooling kit capable of 400L/hr flow rate, all fitted together with 1/2” Tygon tubing and filled with distilled water.

While not as powerful as a dual or even triple 120mm radiator setup or a loop using a dedicated pump and reservoir, it allows us the flexibility to easily swap out cards, as well as providing a decent flow rate through wide tubing. However, it’s important to remember that more elaborate loops involving not only GPU, but CPU and even North Bridge, South Bridge or PWM waterblocks will see not only coolant temperatures rise accordingly, but the overall coolant pressure of the loop will drop as water travels further through more components.

As always, we did our best to deliver a clean set of benchmarks, with each test repeated three times and an average of those results is what we’re reporting here. In the rare case where performance was inconsistent, we continued repeating the test until we got three results that were consistent.

The tests performed are a mixture of custom in-game timedemos and manually played sections with FRAPS to record the average and minimum frame rates. We strive to not only record real-world performance you will actually see, but also present the results in a manner that is easy to digest.

ATI Test System

Intel Core 2 Extreme QX6850 (operating at 3.00GHz – 9x333MHz); Asus Maximus Formula motherboard (Intel X38 Express with two PCI-Express 2.0 x16 slots); 2x 2GB Corsair XMS2-6400C5 (operating in dual channel at DDR2-800 5-5-5-12-1T); Seagate Barracuda 7200.9 200GB SATA hard drive; Enermax Galaxy DXX 1000W PSU; Windows Vista Home Premium x86-64 (with Service Pack 1); Intel inf 8.3.0 WHQL.

Graphics cards

• Gainward ATI Radeon HD 4870 Golden Sample - operating at 775/4,000MHz using Catalyst 8.10 hotfix 8-551-1-71310
  
• Sapphire ATI Radeon HD 4850 X2 2GB - operating at 625/1,986MHz using Catalyst 8.11 beta
  
• AMD ATI Radeon HD 4870 X2 2GB - operating at 750/3,600MHz using Catalyst 8.10 hotfix 8-551-1-71310
  
• AMD ATI Radeon HD 4870 1GB - operating at 750/3,600MHz using Catalyst 8.10 hotfix 8-551-1-71310
  
• AMD ATI Radeon HD 4870 512MB - operating at 750/3,600MHz using Catalyst 8.10 hotfix 8-551-1-71310
  
• AMD ATI Radeon HD 4850 512MB - operating at 625/1,986MHz using Catalyst 8.10 hotfix 8-551-1-71310

Nvidia Test System

Intel Core 2 Extreme QX6850 (operating at 3.00GHz – 9x333MHz); XFX nForce 780i SLI motherboard (with two PCI-Express 2.0 x16 slots); 2x 2GB Corsair XMS2-6400C5 (operating in dual channel at DDR2-800 5-5-5-12-1T); Seagate Barracuda 7200.9 200GB SATA hard drive; Enermax Galaxy DXX 1000W PSU; Windows Vista Home Premium x86-64 (with Service Pack 1); Nvidia 15.17 nForce drivers.

Graphics cards

• BFG GeForce GTX 280 H₂OC - operating at 680/1,458/2450MHz using Forceware 180.43 beta
  
• MSI GeForce GTX 280 HydroGen OC - operating at 700/1,400/2300MHz using Forceware 180.43 beta
  
• Nvidia GeForce GTX 280 1GB - operating at 602/1,296/2,214MHz using Forceware 180.43 beta
  
• Nvidia GeForce GTX 260+ 896MB - operating at 576/1,242/1,998MHz using Forceware 180.43 beta
  
• Nvidia GeForce 9800 GTX+ 512MB - operating at 738/1,836/2,200MHz using Forceware 180.43 beta

Games Tested

• Crysis, version 1.21 (64-bit) with DirectX 10 and DirectX 9.0
  
• Far Cry 2, with DirectX 10/10.1 and DirectX 9.0
  
• Fallout 3, with DirectX 9.0
  
• Call of Duty 4: Modern Warfare, version 1.6 with DirectX 9.0