| CPU Cooler Roundup
This is one of the few cooler round-ups done using an Intel
processor. Most roundups are based on AMD's Socket A processors.
AMD has been the overclocker's choice because of the ability to
unlock the multiplier and the infamous AXIA core which is highly
overclockable.
But there are still plenty of Intel users like me. And AMD
boards have been plagued with inaccurate temperature readings
because of its external thermistor. This can create false
readings. Even external thermistors like the Senfu thermometer
give misleading results when used with socker coolers like the
Thermoengine because the fan is actually cooling the thermistor,
instead of the processor!
So, the coolers in this roundup will be tested using an Intel
processor, with its on-die sensor. This will give the most
accurate reading. I will be using Senfu Thermometer to monitor the
room temperature and the CPU temperature will be recorded using
Motherboard Monitor 5.
Test Setup
| Processor |
750MHz Intel
Pentium III o/c to 930MHz (1.75V) |
| Motherboard |
ABIT BX6-2
(modified to monitor CPU on-die sensor) |
| Slot Converter |
MSI MS6905
Master V2 slot converter |
| RAM |
128MB PC100
SDRAM |
| Power Supply |
300W |
Other stuffs are pretty irrelevant. So, they are not listed
here.
Most of the socket coolers in the market can be used with both
Socket A and Socket 370 motherboards. Here are the coolers covered
in this roundup :-
Alpha PEP66
Global WIN FOP38
Global WIN WBK32
Global WIN WBK38
Kanie Hedgehog 238-M
OP Experimental Cooler
Taisol CGK742092
Thermalright SK-6
Thermaltake SuperOrb
Thermaltake Volcano II
Malaysia has a hot climate with ambient temperature ranging
from 28 to 35°C. That's HOT!!! So, the results will be very
useful for those who want to know
how well these coolers perform during summer, if you are not from
Malaysia.
The tests were done during the night, when the ambient
temperature is stable at around 30.0ºC to 30.5ºC. The ambient temperature is monitored constantly using
the Senfu thermometer.
The objective of my test methodology is to simulate a condition
as similar to that experienced by many people out there. All the testing
was done inside my casing with all the case fans turned on except
for the
GlobalWin FA420 (which is on the back of the Voodoo3). All the
coolers were tested as is - No modification like lapping,
bending the clips, etc. were done. |