Antec KUHLER H2O 920 CPU Water Cooler Review

Antec has been getting more serious about its cooling solutions lately and earlier this year Antec introduced its KUHLER 920 model. If you don't need extreme water cooling, these types of coolers are definitely the way to go in terms of noise and ease of installation.


Results: Temperatures

For our stock KUHLER H2O 920 testing our i7 920 will be kept at 2.66 GHz with a minimal 1.25v. For overclocked CPU testing we will be running our i7 920 at 3.6GHz with 1.45v. Idle temperatures will be recorded after a fifteen minute period of inactivity. Any fluctuation during the last sixty seconds will reset the timer for an additional five minutes. Load temperatures will be recorded after a fifteen minute period of 100% load. To obtain this load we will be using Prime95 v25.3. Any fluctuation during the last sixty seconds will reset the timer for an additional five minutes.

Please note the following scores are normalized to the ambient temperature. This will show you visually the rise above ambient temperature as well as the overall temperature. Since we are dealing with water cooling it is physically impossible for the temps to be less than ambient. This is why we start our charts at 25C.

The ambient temperature was kept at the same level as in previous tests but the temperature inside the case was measured to be between 27C and 29C during our testing.

Before we begin I want to explain why you see three scores for the KUHLER H2O 920 instead of the usual two (low & high). When set to silent the KUHLER H2O 920 is super quiet but I quickly noticed that it performed rather poorly as you’ll see in a bit. What I did was to slowly adjust the speed of the fans until I could hear a difference and then lower it a bit until it sounded just as quiet as before. What I realized was that with no discernible difference in noise (my sound meter registered a 0.5 dBA increase) I was getting MUCH better performance. Thus we have three scores. They are silent, low, and high.

Now this was a moving target since as the water warmed up the fans would spin faster. So I would constantly adjust the speed of the fans. For reference, I would adjust the temperature threshold until the sound meter on the vChil software reached the border between green and yellow, leaning towards green. Since you can’t adjust fan speed directly this took a bit of practice. The software would register this as 31-33 dBA but as I said I could not hear any difference in sound since it was less than one dBA difference from the silent setting. All in all it made for a fun tweaking project.

Stock Settings

Article Image

Right away we see a large difference in performance from silent to low. Because of this large difference I continued to test this third setting. At this point Antec should update its vChill software to increase the speed of the fans ever so slightly. The silent setting, while quiet, performs only marginally better than the stock cooler, which is close to silent itself!

Turning the dial up to "extreme" we get a load temperature that is lower than some idle temps I’ve seen. This is impressive performance to say the least.


CoreTemp Idle: 37-35-36-35

CoreTemp Load: 62-62-62-61


CoreTemp Idle: 35-33-36-34

CoreTemp Load: 56-54-56-54


CoreTemp Idle: 34-32-34-32

CoreTemp Load: 51-50-50-51

Overclock Settings

Article Image

With the CPU cranked up we see a similar pattern though this time the silent setting performs much better than the stock cooler. The low setting is still better as far as "free" performance goes without an increase in noise. Rounding out the top, the high setting barely edges out the Corsair unit and easily beats the best air cooler around.

But at what cost to our ears does all the performance come? Keep reading.


CoreTemp Idle: 45-44-46-44

CoreTemp Load: 78-76-76-74


CoreTemp Idle: 41-39-41-39

CoreTemp Load: 73-71-71-70


CoreTemp Idle: 40-38-40-37

CoreTemp Load: 71-69-70-67


Article Image

Article Image

The silent setting is the quietest setting but only just barely. As mentioned in the other section, by tweaking the speed of the fans I was able to squeeze out better performance with the slightest increase in noise, so slight that I was unable to notice a difference. As did my test subject. (Thanks honey!)

Finally the high setting cranks out so much noise that you would have to like bloody ears (or own headphones) to tolerate the noise. It is bad. To be fair, it is just as bad as the Corsair H70 but that doesn’t make it any easier to accept. It can be heard two rooms away and if you were in the next room you’d have a hard time falling asleep with all the noise.