Testbed and Methods

I took a high-performance Scythe Infinity cooler as an opponent to the Silent Knight. This cooler was tested with its native fan at 1200rpm as well as with two 120mm fans working for exhaust and intake at 1200rpm, too. Besides that, I added the results of Zalman’s "copper hedgehogs" CNPS9500 LED and CNPS9700 LED. You’ll see how they compare with each other, by the way.

What’s curious, the three copper coolers are all very similar when looked upon from the fan side:

It’s normal for the Zalman cooler since the newer model is just an enlarged copy of its predecessor. The Zalman CNPS9700 LED and the ASUS Silent Knight are almost the same size when viewed from above:

But the ASUS cooler is turned around by 90 degrees in the photo.

I don’t include the popular super-cooler Thermaltake Big Typhoon because we have tested it a number of times on different platforms and with different opponents, also those that are going to oppose the Silent Knight in this test. So, I see no sense in testing the Big Typhoon anymore.

The coolers are tested on an open testbed as well as in a system case with the following configuration:

• Mainboard: ASUS P5B Deluxe/WiFi-AP (Intel P965 chipset, LGA775, BIOS 0804)
• Processor: Intel Core 2 Duo E6300 (1866MHz, 266x4MHz FSB, 2x1024KB L2 cache)
• Thermal interface: Arctic Silver 5
• Memory: 2 x 1024MB Corsair CM2X1024-6400C4 PC6400 DDR2 SDRAM (SPD: 800MHz, 4-4-4-12)
• Graphics card: Chaintech GeForce 7950 GX2 1024MB (500/1200MHz)
• Disk subsystem: Seagate Barracuda 7200.10 (3320620AS, SATA-II, 320GB storage capacity, 7200rpm, 16MB cache, NCQ)
• System case: ASUS ASCOT 6AR2-B Black&Silver + an intake 120mm system fan  Noctua NF-S12 (~800rpm, ~8dBA) + two 120mm system fans Sharkoon Luminous Blue LED (~1000rpm, ~21dBA)
• Power supply: MGE Magnum 500 (500W) + 80mm GlacialTech SilentBlade fan (~1700rpm, 19dBA)

There is no AMD platform in today’s tests because it’s pointless nowadays to test coolers on a single-core AMD Athlon 64 due to the decreasing popularity of such CPUs as well as to their lower heat dissipation in comparison with an overclocked Intel Core 2 Duo. And we hadn’t got our AM2 platform ready for this test.

The Chaintech GeForce 7950 GX2 was overclocked from its default frequencies to 580/1580MHz to make the test conditions somewhat harder because the air from the graphics card’s cooler remained in the system case.

The tests were performed in Windows XP Professional Edition Service Pack 2. S&M version 1.8.2b was used to monitor the temperature of the Intel Core 2 Duo E6300 CPU. The CPU was heated up by means of Intel Thermal Analysis Tool for 20 minutes (according to the method we described in the article called Originality or Efficiency? Cooler Master Mars, Eclipse and Hyper TX Cooling Solutions Reviewed). S&M would report a 1.5°C lower temperature than the Intel Thermal Analysis Tool did in Idle mode, but their readings were identical under load.

Considering that Intel Thermal Analysis Tool provides a CPU load untypical of most ordinary applications, I also simulated a Game test mode by running 3DMark06’s Firefly Forest test with 16x anisotropic filtering and without full-screen antialiasing for 19 times. This helps heat up the graphics card and the CPU both together. If FSAA were in use, the GeForce 7950 GX2 would become hotter, but the CPU would bear a smaller load, which is not what I want.

The temperature was read from the sensor integrated into the CPU. The mainboards’ automatic fan speed management was disabled for the time of the tests. The thermal throttling of the Intel Core 2 Duo processor was controlled with RightMark CPU Clock Utility version 2.2 (our processor would begin to skip clock cycles on reaching a temperature of 81.5°C).

I performed at least two cycles of tests in each mode (TAT and Game). I waited for 25-30 minutes for the temperature to stabilize during each test cycle. The maximum temperature in the two test cycles was considered as the final result (if the difference was not bigger than 1°C).

The ambient temperature was monitored by means of an electric thermometer and remained at 24.5-25°C during the tests.