Test Settings
| Test System Configuration | |
|---|---|
| CPU | Intel Core i7-920 (2.66 GHz, 8MB Cache) O/C to 3.80 GHz (19x 200 MHz), 1.416V |
| Motherboard | Asus P6T, X58 Express IOH BIOS 0801 (09/30/2009) |
| RAM | Kingston KHX16000D3ULT1K3/6GX (6GB) DDR3-2000 at DDR3-1600 CAS 8-8-8-24 |
| Graphics | XFX GeForce GTX 285 1GB 670 MHz GPU, GDDR3-2500 |
| Hard Drive | Western Digital Velociraptor WD3000HLFS, 300GB 10,000 RPM, SATA 3 Gb/s, 16MB cache |
| Sound | Integrated HD Audio |
| Network | Integrated Gigabit Networking |
| Power | OCZ-Z1000 1,000W Modular ATX12V v2.2, EPS12V, 80 PLUS Gold |
| Software | |
| OS | Microsoft Windows 7 Ultimate x64 |
| Graphics | Nvidia Forceware 190.62 WHQL |
| Chipset | Intel INF 9.1.1.1014 |
An old C0 stepping Core i7-920 processor produces a great amount of heat when overclocked, which is perfect for testing the ventilation of today’s cases.
Asus’ P6T keeps the processor stable at high temperatures and current loads.
A factory-overclocked XFX GeForce GTX 285 reflects the graphics power that gamers typically need.
Rosewill’s Fort 120 air cooler is large enough to expose any potential mounting space issues, noisy enough at full speed to reveal a case’s noise-isolation capability, and powerful enough to keep our overclocked CPU well below its thermal limits.
The most efficient 1,000W power supply we could find, OCZ’s Z1000 was originally obtained for use in an overkill CrossFire configuration. Though its debut article was cancelled, modular design and an 80 PLUS Gold rating still make this the perfect unit for high-end case reviews.
| Benchmark Configuration | |
|---|---|
| Prime95 v25.8 | 64-bit executable, Small FFTs, 7 threads |
| FurMark 1.6.5 | Windowed Mode, 1280x1024, 8X AA, Stability Test Minimum and maximum temperature |
| RealTemp 3.40 | Highest-core reading at full CPU load (60 minutes) Highest-core reading at 30 minutes idle |
| Galaxy CM-140 SPL Meter | Tested at .25m, corrected to 1m (-12dB), dBA weighting |
In order to create the maximum possible thermal load from our hardware, we dedicated a single CPU thread to FurMark and the remaining seven threads to Prime95 simultaneously, measuring the component temperatures after warming the system for one hour.
Galaxy’s low-cost CM-140 SPL meter takes care of our audio testing needs, though its 32dB minimum rating forces us to check audio levels at a .25m reduced distance. All readings are corrected to an industry-standard 1m distance (-12dB).