Related Articles

The Noctua NH-U 12, along with the competition, was tested on the following platform:

• motherboard
• Pentium D 840 (3.2 GHz, stock Vcore) processor
• 1 GB Corsair PC2-8000 DDR2 memory (5-4-4-9/2T)
• Reference NVIDIA GeForce 6600 GT video card

And the software we employed was the following:

ASUS PC-Probe
• 2x CPU Burn-in v1.01 without error checking for maximum heat generation (one for each core)

The first thing we did was to measure the ambient room temperature. Our highly-scientific $5 thermometer read 24.5C throughout all testing. We tested a total of five coolers, and the four that will be compared to the NH-U 12 are:

• Stock Intel LGA 775 cooler (air cooler)
• Gigabyte G-Power Cooler Pro (air cooler)
• Gigabyte 3D Galaxy (water cooler)
• Corsair Nautilus 500 (water cooler)

Our testing methodology was to clean both the CPU and cooler mounting surface as well as possible with rubbing alcohol, apply a fresh drop of Arctic Silver 3, install the cooler, and boot the system up with the fan speed on high (in the case of those coolers with adjustable fans). We waited approximately five minutes for Windows to settle down, and for the cooler and processor to come to equilibrium. We then took an idle temperature reading.

After that, we would fire up two instances of CPU Burn-in v1.01 with error checking disabled, and run those while keeping an eye on ASUS' PC-Probe. We would continue monitoring until the CPU temperature would not go any higher, and sometimes start drifting between two numbers in an up-down fashion. This was our load temperature. We would then shut down CPU Burn-in, and allow the cooler to dissipate its heat quickly while the fans were still on high. Again this was done until the temperature appeared not to drop any lower. The fan speed was then set to low, the cooler was allowed to reach equilibrium again, and the process was restarted.

And now, the results. First up are idle temperatures:

The Noctua NH-U 12 performs exceptionally here. The G-Power is a tough competitor to beat, but the NH-U 12 manages to do just that -- and by a sizable 4 degree margin at the low fan speed with one fan. A second fan got us a further 1 degree decrease. High fan speeds show us that the NH-U 12 with a single fan is identical to the G-Power in terms of cooling performance. Both water coolers have a clear advantage here, while the stock Intel cooler just doesn't stack up at all.

Let's see what happens when we crank things up a few notches:

Well, everything got hotter, but the places didn't change at all. As you can see, the Noctua was nudged aside only by our two water coolers, but not by much at all! For an air cooler, this is nothing short of astounding. The monolithic NH-U 12 completely butchered the G-Power, a cooler that has become our overclocking chiller of choice. Obviously, the stock Intel cooler can be melted down and made into something more useful -- like a stylish paperweight, for instance. I have no doubts in my mind that the Noctua NH-U 12 will become a lab favourite from now on. This graph warrants another look, as the NH-U 12 is outdone by higher-end water coolers by a mere 1-2 degrees. Air is making a comeback.

Noise levels are one of those things that not too many people tend to think about when buying a cooler. When they get it home, what you thought to be a whisper at the store turns out to be a roar in your recreation room. All of the coolers tested have some measurable noise levels, and some of the figures may surprise you.

Our testing methodology here was quite simple. For the close-range measurement, we pointed the microphone of our Omega HHSL-1 decibel meter at the fan of the cooler, approximately 17 cm (5 inches) away, and away from the airflow so as not to introduce "false" noise. The far-range test was done in the same fashion, though approximately one meter away.

Remember, lower is better!

To be fair, the Noctua NH-U 12 doesn't actually create any noise itself -- the fans do. Since the fans are consumer-supplied, take these figures lightly, unless you purchase the same Coolink SWiF-1201 fans that we used.

Alright, so it's clear that the NH-U 12 beats every cooler we pitted it against, and by huge margins -- even with both fans. Some of the figures were within a hair-width of each other, especially when it comes to the 3D Galaxy and G-Power at low fan speeds. The results really do speak for themselves -- once you stick a Noctua NH-U 12 with a Coolink fan or two inside your case, you'll never hear them again.

A lot of this peace and quiet can be attributed to the fans, as I already mentioned, but a great deal is thanks to the sleek shape and design of the NH-U 12. Because of the flat aluminum fins it uses, there is practically no air turbulence produced. Turbulence can create a lot of noise and amplify the drone of a fan. Concurrently, with there being little turbulence in the fin design, the air being passed through can absorb heat far better.

Something that was discovered, and quite by accident mind you, was that there might be some clearance issues with oversized motherboard chipset heat sinks. Our ASUS P5WD2-E makes use of a rather tall, passive heat sink design, and we found that the NH-U 12's lowest fin was right up against this heat sink. This is definitely something to include in your decision process for any over-sized cooler purchase.