Zalman CNPS9500 vs Asetek Microchill

How we tested

We tested these heatsinks on a Gigabyte GA-K8NXP board using an AMD Athlon FX-55 processor in a Coolermaster Centurion case, with standard case cooling. For reference, we tested them against the reference heatpipe cooler that AMD ships in boxed versions of its X2 processors. Incidentally, this has the greatest mounting mechanism we've ever seen, with a simple lever applying all the pressure you need for a snug fit to the socket.

To test, we first let the system idle in Windows until a steady idle temperature was reached. We then loaded up the system, making it play WMV High Definition video whilst copying files, playing MP3s and encoding DivX, before polishing off with a quick game of F.E.A.R. We used Motherboard Monitor to measure the hottest that the heatsinks got under load.

We tested the Zalman and Asetek coolers both at full fan power and with the power turned down using their noise reduction add-ons that are included in the pack.

Conclusions

Performance: So, the results are fairly self-explanatory. The Zalman, once we'd corrected the base fitting problem, performed well, keeping our roasting FX-55 at low levels. It was easily the coolest here. The Microchill performed well, especially at the low noise setting. At that setting, it is silent, but still performing as well as the stock AMD cooler.

In short, the Asetek will keep your chip silent and cool, whilst the Zalman is a bit better for overclocking.

Noise: In terms of noise levels, the AMD was definitely louder than the others, sporting a faster, smaller fan. Zalman is louder than the Microchill with both on full power, but not by a huge amount. With the fan controllers both Zalman and Asetek supply with their coolers, you can have the fans on near-silent - Zalman works well here, because running silent it's about as cool as the Asetek running on full, and a little quieter. With the low noise fan set to minimum, the Microchill was easily the quietest, whilst still keeping our FX-55 well within tolerances.

Value: Then we come back to the old friend, value. Are these coolers good value for money? The Microchill is £34 with the low-noise fan that we have here or £20 with a louder, better cooling fan. It does its job in that respect - it is quieter than both the AMD stock and the Zalman, whilst offering marginally better performance than the former.

I'd suggest, however, that you'd have to pay me money to use that retention mechanism.

The Zalman doesn't have confirmed pricing yet, but expect it to come in at the £30 mark. It's clearly a good cooler, and will be great on full power for those looking to overclock, and on low-power for those looking for good, quiet cooling.

The Microchill has an awful lot of potential - we've seen that it can cool well, and with a more powerful fan could be a supreme, if loud, cooler. That said, the base needs to go back to the drawing board.

All in all, neither of these coolers has really blown us away, as it were, and certainly neither is a substitute for watercooling your high-end chip. Both get marks for being quiet and powerful, but we were just hoping for a little more.