Testing the Theory
While all this may look good on paper, it is useless if there is not a significant performance difference. To show you just how important choosing the right heatsink/fan unit is, I compared the Globalwin CAK38 to a stock heatsink/fan unit that is usually bundled with AMD processors. The CAK38 features all the right components for a high performance heatsink. It is made out of pure copper, features many thin fins for increased surface area and is cooled by a very effective 7,000RPM fan. The AMD stock heatsink on the other hand is a block of aluminium with fewer; thicker fins and is cooled by a measly 4,800RPM fan. Both coolers were tested on an AMD 1.2GHz Athlon Thunderbird processor and the temperatures was measured using a Senfu Thermometer Probe.
To make things fair, I tested the stock AMD heatsink with a 7,000RPM fan as well as its stock fan. To make sure the processors were running at full load, I ran an hour of Quake 3 Arena loops with Prime95 running in the background.
AMD Stock (4,800RPM)56
AMD Stock (7,000RPM)51
As you can see from the test results, choosing the right heatsink is very important. The Globalwin CAK38 was able to outperform the stock AMD heatsink by 10°C when running with stock fans, and by 5°C when both heatsinks had 7,000RPM fans installed. This is a huge difference and it is clear to see why many users do not even consider using stock heatsink/fan units with their processors. This applies even more so to overclockers.
Hopefully armed with the knowledge given to you in this article, you will be able to make a much more educated decision when purchasing your next heatsink/fan combo.