Testing and Results
This content was originally featured on Amdmb.com and has been converted to PC Perspective’s website. Some color changes and flaws may appear.
Testing System Specification:Testing will be done on my main system. Specification is below
- AMD XP 1700+ Thoroughbred Revision “A” @ 11 * 192
- Epox 8k5a2+ motherboard
- Thermalright AX7
- Artic Silver 3
- 2 sticks of Corsair PC3200 256MB RAM @ turbo timing
- AOpen GeForce4 Ti4200 128MB RAM
- Maxtor fluid bearing 60G Harddrive
- Antec True Power 430W dual fan power supply
- Coolermaster 200 case
The Coolermaster 200 case has 5 case fans. 2 Panaflo L1A’s are used for intake, 1 Panaflo L1A is used for side blow hole, 1 Sunon 31 cfm is used for top blow hole, and finally 1 Sunon 39 cfm is used for back exhaust.
- CPU voltage setting: 1.9V in BIOS, 1.95 in Motherboard Monitor & Magic Health at post
- RAM voltage setting: 2.7V in BIOS
Testing Procedures:
All testing was performed with ambient room held at a constant temperature of 20ºC ±1º. The idea of case cooling is to use fans to draw ambient air to the heatsink and dissipate heat produced by the CPU. Due to this, temperatures obtained will always be the same or higher than the ambient room temperature.
Both full load and idle temperatures were measured through Motherboard Monitor. Though there are many CPU intensive programs such as Prime95, SETI@home, Folding@home, and Genome@home, I decided to test the full load temperature by running Distributed Folding because I am one of the proud members of the AMDMB.com Distributed Folding Killer Barbarian Team. Full load temperature was obtained by running CPU at 100% for a period of 1 hour. Then idle temperature was measured by running computer idle for 1 hour. Fan noises were recorded so readers can compare them.
Just as expected, high cfm fans result in lower CPU temperatures. As you can see, the Vantec Tornado has the lowest temperature, follow closely by Delta. It’s rather interesting to point out that Vantec Tornado is only 2 degree away from the default system setting using H1A. Considering the voltages and the overclocking settings used on the testing system, anything under 40C is quite good. At 33C full load and only producing 35 dB, Enermax Adjustable fan has shown itself to be quite worthy in this benchmark. Interestingly, though YS-Tech adjustable is only rated at 48.7 cfm, it actually out performed Mechatronics.
The case temperature seemed to change depending on the fan that was used. The higher the fan’s cfm rating was, the higher the case temperature. After discussing with several members on AMDMB.com forum and testing with different settings, I came to the conclusion that the case temperature sensor on Epox 8k5a2+ is located somewhere close to the CPU socket. The high cfm fans push more hot air toward the sensor, hence the higher case temperature.
Again both the Vantec Tornado and Delta show very good results. But does this mean everyone should use these fans?
Benchmark Data
On the above diagram we can easily see a trend. The fans that have a ratio higher than 2 all have high fan noise, while fans that have a ratio lower than 1 are all quiet. In the ratio of 1 range, Enermax came out on top follow closely by Mechatronics. I am rather disappointed that H1A only has a ratio of 1.05.
