Testing low-end coolers. Spring 2007
Author:
Date: 30.03.2007 |
|
Tests
Since these coolers are aimed at varied CPU sockets, we assembled two test configurations differing in only motherboards and CPUs:
|
Test-bench configuration
|
|
CPU
|
? Socket T (LGA775) Intel Core2Duo E6400 (Allendale, B2) @3000 MHz / 1.325V
? Socket AM2 AMD Athlon 64 X2 3800+ ЕЕ (Windsor, F2) @2400 MHz / 1.4V
|
|
Motherboard
|
? ASUS P5B Deluxe rev.1.03G (Intel P965 Express)
? ASUS M2N-SLI Deluxe (NVIDIA nForce 570 SLI)
|
|
RAM
|
2 x 1024 DDR2 800 Corsair XMS2 Dual Kit (TWIN2X2048-6400C4)
|
|
Video Card
|
256 MB Sapphire ATI Radeon X1950XT
|
|
HDD
|
250 GB Seagate SATA II, 16 MB cache (ST3250620AS)
|
|
Housing
|
ThermalTake Xaser III (window, 5 housing fans 80-mm)
|
|
Power supply unit
|
FSP Optima 600 W (OPS600-80GLN)
|
Both the processors were subject to slight overclocking to imitate high-end models that offer higher heat emission.
The ambient temperature during the tests varied within 23…24°С. Among the special features, we note a very good ventilation in the housing and quite a hot video card. On the video card, we replaced the default cooler with the alternative cooling system Sytrin KuFormula VF1 Plus which does not expel heated air outside the housing but leaves it inside.
You must have already been familiar with the testing methodology through our previous materials, so we are imitating three operating modes of the computer:
- Simple. In this mode, there is no load upon the CPU, and all the power-saving technologies are disabled. In the end, we produce a CPU temperature which is typical of the "office" operating mode of the computer and "surfing" around the multimedia content and Internet.
- Games. This is already a serious load upon the CPU and a hard job for its cooler. The CPU itself in this mode is normally used at 80...100%, and the video card running at all the 100% creates a truly tropical climate inside the housing. To generate this type of load, we should simultaneously run two applications: a test for artifacts using ATI Tool which produces a 100% load upon the video card and approximately 80% upon the CPU, and archiving with WinRAR running in the single-threaded mode, in order to bring the CPU usage up to 100%.
- Maximum load. This is a stress test for the CPU and its cooling system. Since S&M has not yet been updated for the full-featured support for Intel Core 2 Duo, we used the no less "hot" module of stress tests in Everest which cyclically runs two most demanding FPU tests. For the AMD platform, the S&M utility is still the most convenient tool for heating, so we used just that. In the Max mode, the CPU is heated up to its maximum temperature which none of the "household" tasks is able heating it up to that.
For comparison, we added boxed cooler for AMD and Intel processors into the list, and even one of the Hi-End class – the cooler Thermaltake Big Typhoon VX.
The results are predictable enough, although it is a bit surprising that the cooler Igloo 5072 Silent lost a little to Intel's boxed cooler, which is strange because it has a powerful radiator and some higher rotational speed of the fan. Most likely, the cause is in the performance of the fans which depends not only on the rotational speed but on the design as well.
Almost all the low-end coolers have shown a performance similar to that of the boxed version. Only GlacialTech Igloo 5072 PWM stood out - at the maximum speed of the fan it showed up to 8 C lower, but at what price? .
The list of participants in the "AMD race" has been noticeably changed . SilverStone TN04 has demonstrated the weakest results - in fact, no one expected miracles from it because the compact and quite "rare" radiator is unable to dissipate heat effectively.
The other "low-enders" have demonstrated a performance greater or equal to that of the boxed AMD cooler, and especially powerful was Arctic Cooling Alpine 64 - it demonstrated an excellent performance in the maximum load mode while being almost unheard inside the housing.
It is interesting to see how the fan's automatic speed control works for ASUS Chilly Vent - just look at the diagram. As you remember, its rotational speed depends on the air temperature inside the PC housing. During tests under the maximum load upon the CPU the hot air from the cooler was immediately picked up by the housing fans and expelled outside, so the overall air temperature inside the housing did not differ much from the normal. The fan's rotational speed under the Max load was at about 2000 RPM, and 1600 RPM in the idle mode. In the gaming mode, ATI Radeon X1950XT expelled all of its heat into the housing, so the inner temperature went up substantially. The rotational speed of the fan in ASUS Chilly Vent went up to 3000 RPM at which the cooler proved one of the best performers in this operating mode. Although the noise level went up significantly.
Conclusion
GlacialTech Igloo 5072 (Silent / Light / PWM)
Coolers of this series will be quite good as a replacement to Intel's boxed cooler, although that will not add much in terms of the cooling efficiency. On the other hand, the tradeoff will be quite a high and annoying noise level, which especially applies to the higher-end models Igloo 5072 and Igloo 5072 PWM. All in all, the new fan has not proved to the most lucky buy in this line because the previous series Igloo 5071 offered more comfortable acoustic performance.
To install the coolers, you would have to dismantle the motherboard, but that gives a high fastening reliability. At the same time, the cooler does not bend the textolite of the motherboard, unlike the coolers which are fastened with plain clips. The advantage of the Igloo 5072 series is its low price and wide occurrence on the territory of Russia, which is typical of all the coolers made by GlacialTech.
Arctic Cooling Alpine 7 (64)
Low-end coolers by Arctic Cooling have stood out for their quiet operation coupled with excellent cooling level. So they can be safely recommended for those who are after reliability and quietness. Of no less importance is that the design of the fan provides blowing about the near-CPU space - the power transistors also need cooling. Unfortunately, the produce of Arctic Cooling is hardly made available in Russia, so it won't be easy to buy them. On the other hand, it price is also not the lowest - 15(17) USD. To install the Alpine 7 model on the LGA 775 platform, you would have to dismantle the motherboard, but that inconvenience is made up for by the fastening reliability. Installation on the Socket AM2 platform is straightforward and causes no issues.
SilverStone NT04
Regrettably, we can't praise this small nice-looking cooler. Its major advantage is perhaps no more than the nice design. By efficiency, it lags well behind even the boxed AMD cooler, let alone the other reviewed coolers. It can't be called neither noiseless nor powerful. Maybe cheap? But you still can't make it available anywhere.
This cooler is unlikely to appear soon on the shelves of our retail shops, but even if does, think twice before you buy it..
By the way, we can help recalling how difficult it was to remove the "silver" thermo paste off the CPU...
ASUS Chilly Vent
Quite a curious cooler, and its major "spice" is just in the fan's automatic speed control. While the motherboard most often adjusts the cooler's rotational speed depending on the CPU temperature, the ASUS Chilly Vent varies its speed depending on the air temperature inside the housing. If we compare these two methods, the most appropriate would be adjustment depending on the CPU temperature. But it often happens that CPU thermal sensors may misread which depends on the BIOS version, so the method used in ASUS Chilly may prove more trustworthy. At the same time, as we could see, it works reliably and consistently.
It is all fine about the cooling efficiency due to the copper base, the couple of heat pipes and a thick radiator - the cooler demonstrates quite a decent performance coupled with a powerful fan. There is a distinct rise in the cooler's efficiency with the rise in the fan's rotational speed. That is due to the high density of the radiator because a strong air flow is needed to provide good blowing.
The installation convenience is a bit shaded by having to apply a very sticky thermo compound on the CPU surface on your own. On the other hand, the margin of thermo interface leaves us the opportunity to reinstall the cooler several times if needed.
ASUS Chilly Vent Lux
This cooler is in fact a twin of the formerly reviewed ASUS Chilly Vent, but instead of the fan with RPM regulator, it is equipped with a permanent speed fan. You can hardly call it absolutely noiseless, and by the efficiency it proved to be better than AMD boxed cooler. To get a substantial performance boost, a much faster fan would be needed, but in that case you would have to put up with the high noise level. ASUS made a bid for the acoustic comfort, so the cooler has been dubbed as «Lux».
The installation traits fully meet those for ASUS Chilly Vent.
Thermaltake TR2-R1
Not a bad radiator and a huge fan made it possible for the new low-end cooler Thermaltake to demonstrate good efficiency, and most importantly – a very quiet operation. No doubt, this cooler has proved to be successful - at $9 in USA it appears to be a good replacement for the AMD boxed cooler, thus providing a quieter and effective cooling. Due to the bigger fan, the cooler does a good job blowing about all the near-CPU space.
The installation convenience is specific for all coolers for Socket AM2 which use a bracket with a clamp, so Thermaltake TR2-R1 has proved no exception.
We appreciate Sunrise-Rostov for the hardware presented for tests.
 |
Content: |
 |
|
|
Top Stories: |
|
 |
|
MoBo:
|
|
|
|
VGA Card:
|
|
|
|
CPU & Memory:
|
|
|
