I decided to do a thorough benchmark/review of this cooler that everyone seems so interested in.
I began by testing a cooler we all know thus we can all relate => the stock Intel cooler, so may observe how much better this aftermarket solution is.
This review being overclockers oriented I overclocked the daylights out of my high vid, high voltage, high heat generating Q6600 which is one of the most popular quad core chips around here.
Test System:
CPU
Q6600 G0
Motherboard
Asus Striker II Extreme
Memory
4xKingston HyperX PC3-11000
Graphics Card
GTX280
Hard Drive
Samsung SpinPoint F1
Sound Card
SupremeFX II
Power Supply
Asus U-75HA 750W
OS
Windows 7 RC 64bit
Let’s see what we can do with the stock HSF.
Benchmarking the Stock Intel Cooler
Test rig:
Signature Rig (stock Intel Cooler + Stock Thermal Grease)
Open Chassis (mobo exposed, no side panels)
No Extra Fans (only fans present were the one from the cooler and the one from the PSU)
Normal room temperature, still air (windows and doors closed => no drafts)
Method of Testing:
I used Intel Burn Test to generate my load temp. All my overclocks passed the test so we must assume that they are stable. I could have used lower voltages but I didn’t have the patience to tweak around BIOS, but I think they are pretty good. The Intel Burn Test is a synthetic test, especially designed to heat up the CPU, therefore no real life app will generate such high temps, nor other stability test.
Idle temp was measured whilst surfing the net and using MS Word after I started the rig.
The variations you see in those temps represent the difference between the cores, as one may run hotter than another.
This is what I came up with:
Frequency Voltage Idle Temps Load Temps
2.4 1.120V 30C 44C
3.0 1.232V 36-38C 58-62C
3.2 1.376V 38-40C 78-80C
3.4 1.440V 40-42C 87-96C
Update 2:
I have also modified the fan clips than came with my Mugen to fit the Ultra Kaze.
It was a matter of bending the clips, really easy stuff, no need to buy special 38mm clips, or use zip ties and other silly workarounds.
This is before the bending, notice how the clips are attached to the front of the fan.
This is after, notice how they attach to the rear part of the fan.
Some pictures from the benchmarks themselves.
@2.4
@3.0
@3.2
@3.4
I installed the Mugen 2 with an Ultra Kaze 3000, and used the fan that came with the cooler as an exhaust/pull fan.
Here are the results:
Benchmarking the Scythe Mugen 2
Frequency Voltage Idle Temps Load Temps
2.4 1.120V 25C 39C
3.0 1.232V 27C 50C
3.2 1.376V 30C 60C
3.4 1.440V 32C 70C
As you may notice the temps are around 20 degrees lower than the stock HSF, thanks to this cooler right now I'm at 3.6 1.536v 33idle 78load, if I had used the stock cooler, right now I would have smelled some Kentucky Fried Kentsfield.
Conclusion:
I warmly recommend any overclocker to get this cooler, the base is perfectly flat (did the razor blade test) and perfectly finished, even the thermal paste that came with it is good.
It was easy to mount and it feels really well secured to the motherboard.
It may not beat a lapped TRUE, but It will dominate any other air cooler out there and it will do it at a bargain price, and it will fit on any platform.
Coupled with an Ultra Kaze 3000 it is truly unstoppable, it really moves a lot of air, and it’s not as loud as I feared; now I can barely notice it (I haven't gone deaf).
Pics:
Hope this was useful
