|
Testing
The usual testing methods were applied here. Now whilst have thoughtfully provided some thermal paste with the Aero7, I won't be using it. To be honest I look at the white paste and instantly think cheap and nasty (that said, some companies, don't include any paste at all) so like the majority of you will do, I've used something a bit pricier, Artic Silver II, the same as I've used with all my HSF reviews bar one (the Silver Mountain 2Q which came with Shinetsu thermal pads.) The test system is as follows:-
AthlonXP 1900+ (supplied by ) @ XP2000 1.75v 166FSB
Abit AT7 Max
Skyhawk Midtower with 2x Intake, 1x Exhaust, with all fans including the PSU fans set to low.
Windows XP (SP1)
 
As the Aero7 features an adjustable speed fan, I've tested the fan at 3 different speeds, basically low, medium and high. As a comparison, I've used the Coolermaster X-Dream and the Swiftech MCX462 with a 3000 rpm Coolermaster Fan. For idle settings, Windows was booted up and left for 15 minutes and a reading taken. For load settings Folding@Home was left running for at least an hour and readings taken. This was done 3 times and the average temperature taken for each. Room temperature was maintained at 26C and the case side was left off.
Well, I think the table above speaks for itself. The temperature difference compared to the others is something that I wasn't expecting at all. Both the X-Dream and the Swiftech, with fans at full speed (constant in the case of the Swifty) are idling at 45C. The Aero7 quite literally blows them away with a 4 degree difference, dropping to just 41C. The last time I was this surprised with a HSF was when I upgraded my Thermalright SK7 and Delta assembly to the Swiftech, which not only dropped the noise significantly but also the temperature. Now at this point both the X-Dream and the Aero7's fans are at full speed, which is pretty noisy. The Swiftech however remains pretty much silent. When we drop the fan speed to 2000rpm on the Aero7 the noise level drops to below that of the Swiftech setup and the temperatures are on par with one another in regards to load. Noise wise, you simply cannot hear the Aero7 at its lowest speed. At 2700rpm, you can hear the fan for the first time, that's to say that at 2700 rpm it's at the threshold of when you can and can't distinguish it from the other fans and the PSU. At 3750 rpm it's a pleasant hum. It's not the same kind of hum you would expect from a fan of this size, sounding more like a 120mm axial fan, rather than the 70mm it actually is. So is this HSF's performance thanks to the fan design or is it the 'sink itself? The only way to really tell is to swap the fan out for a traditional axial fan. Now this 'sink is 70mm, and I don't have any other 70mm fans.
So in true DIY fashion, I've rigged up an 80mm fan to the 'sink which is being held on by the same clips that used to hold on the fan to my Thermalright HSF. The 80mm fan in question is the same fan I used on the Swiftech, and I've run both at 3000rpm. This puts the output of the 80mm fan at around 28CFM, with the output of the Aero7 fan at just under 18CFM. With that in mind, make what you will of the next set of results.
The 80mm fan covers the entire heatsink with air, including a slight overlap at the edges. However the 'deadzone' found in centre of the fan would seem to hurt performance by 2C when compared to the Aero7 fan. Using the Aero7 radial fan covers more of the heatsink with air, negating the 'deadzone' dead centre as well as the few millimetres in the 4 corners. Bear in mind the CFM difference of the 2 fans as well, the 80mm pushes a lot more air, yet it would seem from my testing that because the air isn't covering the entire heatsink, efficiency is decreased.
An impressive turn out for this new design from Coolermaster.
Next Page - Final Words
Previous Page - A Closer Look
|
Cooler Master Aero 7 Socket-A: Cooler Master kicks out a new AMD cooler, which uses a different kind of fan than most people are accustomed to. The performance of the package was quite surprising.
