Performance

We tested the Vapochill Micro on a DFI LanParty NF4 motherboard. SysTool was used to get the CPU temperature. Room temperature was kept at a constant 20°C, verified with an external thermometer near the fan inlet. Also a temperature probe was stuck on the edge of the IHS to get a feeling of the accuracy of the board's temperature measurements. Small changes in room temperature (max. 0.2°C) were manually compensated for.

The cooler was mounted four times, temperatures were observed. Then we remounted (2 times) until we got the best temperature of those four again.

To test the different fans, only the fan was changed, the cooler remained mounted.

When reviewing this article, Morten from Asetek suggested that running without the CPU motherboard bracket might reduce temperature even more. I tried this and saw no temperature difference, however, this might be something to check out if you think your temperatures are too high.


"Low" means the fan controller on its lowest setting, "High" is its highest setting. If you choose not to connect the fan controller, the fan will automatically run at the "High" setting.

Idle means Windows sitting at the desktop for 30 minutes. Load is after 30 minutes of Prime95. (maximum-heat mode).

First of all, it is not possible to run the Vapochill Micro without a fan. At idle, the temperatures were stable in the 50° region, but once the CPU got loaded, temperature rose quickly - I stopped the test at 80°C.
Later during testing I started Prime95 stress testing and turned away to work on another PC. After a while it made *click* and the benchmarking machine shut off. What happened? The temperature probe measuring the heatspreader's temperature had the answer: 120°C !!! - I forgot to connect the fan's power.
This proves two things. 1. You can not run the Vapochill Micro without a fan and 2. AMD CPUs withstand temperatures of more than 120°C.


When the system is idle, all fans and settings are really close to each other. The CPU heat output is very low, probably lower than the boiling temperature of the R134A. Looking at the graphs, I would say the boiling temperature is somewhere around 32°C-34°C.


Under load the results get more interesting. On the low setting, both the Extreme and the high-End version result in high temperatures. The Silent version is doing better here. You do have to consider that the High-End version does not come with a fan controller, so the number to look at here is the high setting, that's the speed it would run at without a fan controller connected.

As you would expect, the Extreme version leads the temperature charts on the high speed setting. the high-end version is close behind. Another 2°C higher is the Silent version.

Now you have to consider those results in relation to the fan's sound level.



Both the Extreme and High-end version are so quiet that you can barely hear them if you put your ear right next to the fan - but this results in seriously degraded cooling performance. While the Low-noise fan is a bit louder, it is still unbelievably quiet on the low-speed setting.
At the high-speed setting the Extreme Performance fan sounds like a vacuum cleaner, reminds me of GeforceFX times. The high-end fan is still a bit loud, but the low-noise fan comes out with a very acceptable sound level.

Value and Conclusion