Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors Home PC Hardware from Made-in-China.com Search: Enter your search terms   Web www.lostcircuits.com Submit search form	.
Prices:

With the original CNPS7000-Cu we had the complaint that the spring was a bit to loose and we needed to increase the tension to warrant good contact with the CPU surface. With the CNPS7000 AlCu, we had no such issues at all in the same setup. All that was needed was to tighten the screws and the cooler was sitting on the processor like it was held by a vise. Most likely, our original problems were an isolated event.

It almost looks like a really bad level of UnrealTournament2003. Keep in mind, whatever you do, do not open the hex screws that are holding the entire assembly together.

Results

Instead of pitting the CNPS7000 against some other coolers, we show how the two different incarnations of the CNPS7000 series stack up against each other. We used the test criteria outlined on the last page and the only caveat is that it was not possible to keep the room temperature absolutely constant, that is, during the testing period, the ambient temperature varied between 19.1 and 20 centigrades as measured inside the open case. The AlCu was tested first and overall had a slight advantage over the pure copper version.

At 2900 rpm (without interposed Fan Mate1), there is hardly any difference.

With the Fan Mate1 installed, the higest rpm drop to about 2630 rpm. No difference here either that could not be accounted for by the 1 degree difference in RT.

With lower rpm, it appears as if the capabilty of aluminum to give off heat better than copper pays off.

Lowering the rpm even further shows the same picture.

At the lowest possible speed, both coolers perform about the same.

Conclusion

Regardless of the testing conditions, both coolers perform about the same. There are subtle differences with the caveat that we can only monitor integer numbers as they are displayed by MBM5. One thing we would expect is that the -AlCu is more sensitive to transient spikes than the -Cu. Our data seem to support this but the resolution of our measurements does not warrant any definite statement. Likewise, even though the overall lower temperature under load, courtesy of better heat dissipation of aluminum nicely fits our expectation, one needs to be very careful not to fall into a wishful thinking trap. There were other factors that could influence the outcome of the tests as well, for example, the air humidity changed during the course of the testing because of a rainshower. Of course, it is possible to repeat everything over and again but by the end of the day, there will always be other variables.

Bottom line is that either there are no differences between the -Cu and the -AlCu version, or if any differences exist, the AlCu bimetal design is better than the all copper version. Add the facts that the pure copper version weighs 50% more, further that copper will oxidize over time, which causes deterioration of heat exchange because of the insulating properties of copper oxide and last not least that copper is softer than aluminum and, hence, be more prone to mechanical damage and ... the winner is... CNPS7000-AlCu.

Quotable: "Quite honestly, after looking at the CNPS7000-Cu, we did not think that there could be much improvement but the CNPS7000-AlCu proved us wrong."

next page:    => More =>