For testing the Coolit Domino was installed into a system comprising of the following components:
Intel Xeon 3070 (Stock & 3.6Ghz, 1.4v)
DFI DK P45 TR2S Plus
2GB Crucial Balistix
Here I would normally outline the procedure of installing the cooler under the spotlight today, but this time around coolit have out done my simple words with an instructional video!
There really isn't much more to installing the Domino into your case, it's ridiculously easy. Coolit certainly have hit the nail on the head in this department. Something they missed out of the video, that is mentioned in the instruction manual, however, is the idea of using a little bit of tape to hold the back plate on. A mindbogglingly simple addition that can solve the only slightly tricky bit of the installation.
I said on the previous page I'd report on how the rubber mounts worked out during installation, and I can report they were quite good once they were in and secure. A little bit of vibration dampening which was a welcome factor and they weren't going to fall out any time soon. However during the installation, pulling the through the holes wasn't so simple. They had a tendency to stretch a rather long way before actually clipping in. I wasn't willing to find out how far they'd stretch before they break, but I'd advise anyone installing one of these to take care.
Another thing I mentioned on the previous page was the LCD. This certainly did do it's job, displaying a few useful statistics about the cooler without going OTT.
The first test conducted was the tried and tested 'strap the cooler to a chip and see how it fairs' method. The Domino was tested using a Xeon 3070 mentioned above, in both stock and overclocked states. The system was left to undisturbed for 30mins after boot and the temperature then recorded for the idle reading. This was followed by a 30 minute Prime95 romp to take the loaded reading. Ambient throughout was 20° (+/- 0.5°). The test was then repeated 3 times using each of the Domino's performance modes.
The next test conducted was a simulated load test. The cooler was set to work finding it's equilibrium temperature when exposed to the heat generated by a certain amount of power. The cooler was tested at points of 50w, 100w, 150w and 200w to see how it would cope under various levels. Once again ambient was 20° (+/- 0.5°). The test was performed using a custom built load tester. This device uses two 200w cartridge heaters to warm up a small block of copper, allowing a flat surface for the heat sink being tested to be mounted on. The majority of the block is then insulated to minimise heat escaping via anything other than the side that the sink is mounted in. The heat dissipated by the twin cartridges is regulated via a variac, and the power between the variac and the cartridges is measured to ensure that the correct amount of power is being drawn. The temperature of the block is taken via a digital thermometer and k-type probe, from a small hole drilled into the copper between the two cartridges.
The noise levels emitted from the cooler are extremely hard to asitane without some very expensive equipment, therefore you will have to rely on my fair ear for judgement in this department.