|Best CPU Cooler Performance Intel/AMD Q1-2010|
|Reviews - Featured Reviews: Cooling|
|Written by Olin Coles|
|Saturday, 13 March 2010|
Page 22 of 24
Enthusiast Fan Comparison
Over the past few months there have been several requests by our readers for Benchmark Reviews to begin testing these coolers by using many different enthusiast level fans. While the idea seems interesting enough, the actual undertaking would be unreasonably time consuming. As it is now, each cooler already takes about two hours to install, prepare, and test; and most coolers receive five or more tests in stock and overclocked conditions. The test process for an individual cooler could take several days depending on ambient temperature and humidity.
In this section, we've taken the Yate Loon D12SH-12 that's been used for the past year of high-output testing and compared it to the Scythe Ultra Kaze, which is a 38mm thick enthusiast fan. I've heard a great number of enthusiasts claiming phenomenal results with the Ultra-Kaze, but it's not like me to believe hearsay. In this section, I've taken the opportunity to directly compare several 'high volume' test fans. Here's how it was done:
Lavalys EVEREST Ultimate Edition is used to create full loads on each core utilizing the system stability test (Stress CPU and Stress FPU options), and also measure individual CPU core temperatures. After a minimum of sixty minutes at full load, temperatures sustain a plateau and the ending ambient room temperature and individual CPU core levels are recorded. The fans are quickly replaced while the system is still under load, annd given thirty additional minutes with EVEREST loading the CPU cores before final temperature readings are again taken and recorded.
It becomes evident that there's a point of diminishing returns for every CPU cooler, and when 'over-fanned' the results collide closely together. Although the 2.66 GHz Intel Core i7-920 processor we've used for this test does a decent job of separating the crowd when these coolers keep their stock fan, once a high-output fan is attached there's really not much separating them all. So it's a good bet that most of these top-performing products could cool an overclocked system extremely well, but these days only the very best will do. Much like the Thermal Interface Material testing we've conducted, all of the high-performance products are beginning to perform at nearly the same levels. Eventually, I expect to see the same technology used in all cooling products with the difference being the application. This is where experience comes in handy, and we've shared some of this with you in our Best Thermal Paste Application Methods article. Remember, less is more when it comes to thermal paste, and soon CPU coolers may offer the same paradigm.
Taking a small collection of CPU coolers, Benchmark Reviews tested the performance of four different fans.
As you can see from the chart above, the Yate Loon 120x120x25mm model D12SH-12 fan performs the best compared to other enthusiast cooling solutions. Rated at 88 CFM by the manufacturer, I suspect that these are conservative figures. Operating on 0.30A, the Yate Loon D12SH-12 certainly moves a lot more air than suspected for a $7 fan. The real trick is in the operating voltage, which ranges from 6.5-13.8V, allowing enthusiasts to move the hot wire to either the 5V+ or 12V+ power lead. So until I'm given a good reason to think otherwise, Benchmark Reviews will continue to test high-output overclocked results with what I consider to be the best value in fans.