Cooling Efficiency Tests

Using the “weakest” cooling system of the today’s testing participants we managed to overclock our 45 nm quad-core processor to 3.84 GHz (+43.8%). The nominal processor Vcore was increased to 1.31875 V in the mainboard BIOS (+9.9%):

During CPU overclocking we activated in the mainboard BIOS the “Load-Line Calibration” function that lowers the voltage drop on the part of the voltage regulator circuitry before the CPU. The system memory voltage was locked at 1.525V and its frequency was at 1500MHz (8-8-8-18 timings). All other parameters available in the mainboard BIOS and connected with CPU or memory overclocking remained unchanged (set to Auto).

The obtained results are given on the diagram below in order of increasing cooling efficiency (lowering of the peak CPU temperature):

XIGMATEK Cobra turned out the least efficient cooler in our today’s test session. Although, frankly speaking I expected this cooler to be at least as good as Dark Knight if not the best of the three. Looks like small effective surface area of the cooler heatsink prevails over the four heatpipes sitting closely to one another in the cooler base. Even a slightly cheaper Nepartak with one heatpipe less and almost twice the gap between the heatpipes in the base is about 4°C more efficient than Cobra in quiet mode and all 9°C better at maximum fan rotation speed. BY the way, Nepartak turned out 1°C better than a more expensive Dark Knight, although it noise level was also higher this case.

Dark Knight proved to be the best cooler of the three discussed today. It is only 2°C behind XIGMATEK’s flagship Thor’s Hammer solution at maximum fan rotation speed. The latter, even at maximum fan sped, yields to Thermalright IFX-14 3°C and 5°C in both modes respectively. By adding the results of this super-cooler into our today’s tests gives us a better idea of its advantage over the competitors even during moderate CPU overclocking for Thermalright IFX-14.