Miniature CPU Coolers: Four Models Reviewed

Not everyone needs super-efficient and expensive CPU cooling solutions, which we test on a regular basis. On the contrary, most users would be quite happy with a decent replacement to their boxed coolers that would run quieter. Today we are going to talk about four solutions like that from Scythe Thermaltake and Xilence.

by Sergey Lepilov
10/14/2009 | 10:16 AM

Looks like cooling solutions business is so profitable that they won’t stop developing new or pseudo-new coolers until someone introduces a revolutionary new approach that will change the way PC components are cooled. Now we are witnessing an endless flow of cooling systems of different efficiency, where you can easily get lost. Although we are constantly testing new cooling solutions, most of them still remain left out for understandable reasons, while our primary attention goes to technologically interesting products that usually cost quite a bit.  However, not all computer users out there need super-high-efficiency coolers targeted primarily at overclockers, because the percentage of overclocking fans is relatively small compared with the rest of the computer user community. That is why today we would like to review four compact and inexpensive CPU coolers from Scythe, Thermaltake and Xilence.

Testing Participants

Scythe BIG Shuriken (SCBSK-1000)

 

The second modification of the low-profile cooling solution, which name comes from the hand-held throwing weapon shaped like a star, is shipped in a compact box, which is completely covered in all sorts of information and data:

 

This time the box is designed in dark-purple colors, unlike the gray-red box of the first Shuriken. You will never mix them up in a store (if they ever happen to be side by side, as the old modification may be discontinued by then).

The cooler is bundled with retention kits for all contemporary platforms including LGA1156, SilMORE thermal paste and installation instructions:

Scythe BIG Shuriken is made in Taiwan and its recommended retail price is set at $34.95. The manufacturer guarantees that the cooler and its fan will work failure-free for a least 1 year.

The new cooling solution measures 125x135x58 mm, which means that it is 20 mm wider and longer than the first Shuriken on both sides, but 4 mm shorter. It has become 50 g heavier and weighs a total of 405 g.

 

Scythe BIG Shuriken now has one more heatpipe 6 mm in diameter, so there is a total of four heatpipes.

 

Cooler heatsink uses “Uneven Parallel Heatpipe” technology that implies that the heatpipes go beneath the heatsink instead of piercing it through. In other words, the heatsink sort of lies on top of the heatpipes that go through special grooves in it:

 

The heatpipes are soldered to the heatsink inside the grooves. There is also an additional aluminum heatsink above the base plate that not only partially offloads the upper part of the heatpipes in the base, but also serves as support for all retention kits.

BIG Shuriken heatsink consists of 78 aluminum plates, each only 0.35 mm thick. They are spaced out at 1.1 mm from one another:

The copper base plate is 2 mm thick and is covered with a thin layer of nickel alloy. Its finish quality is superb:

The surface is not only mirror-impeccable, it is also very even. Both thermal compound imprints – on the glass surface as well as processor heat-spreader – turned out perfect.

Scythe BIG Shuriken has not only a bigger heatsink, but also a bigger fan. It uses 120x120x12 mm Slim Slip Stream (SY1212SL12M-P) fan:

 

The fan rotation speed is automatically adjusted using pulse-width modulation method (PWM) in the interval from 650 (±300) RPM to 1600 (±10%) RPM. The fan should create 15.77-38.05 CFM airflow and generate 12.91-28.89 dBA of noise. They didn’t mention the static pressure in the specs (however, I assume that a fan of such thickness with the blades of such shape should have very modest static pressure numbers). According to the specifications, the maximum fan power consumption shouldn’t exceed 2.28 W.

Scythe BIG Shuriken should be installed onto mainboards with V.T.M.S. (Versatile Tool-Free Multiplatform System) that requires no tools or any kind or removing the mainboard from the system case. Just insert the appropriate retention brackets into the special grooves in the lower heatsink of BIG Shuriken and install the cooler onto the CPU. Despite its low-profile design, BIG Shuriken cooler co-existed perfectly fine with all the surrounding electronic components and heatsinks around the processor socket…

 

… but it turned out incompatible with tall heat-spreaders on the memory modules installed into the first two DIMM slots (on both platforms: AMD and Intel). That is why we had to remove them:

This cooler could fit just great if we used DDR2 memory modules without heat-spreaders, but I didn’t have any at my disposal at the time of tests.

Thermaltake ISGC-100 (CLP0537)

The second new cooler we will be discussing today belongs to Thermaltake and is the junior solution in the ISGC lineup. This abbreviation stands for “Inspiration of Silent Gaming Cooling”. ISGC-100 ships in a small flat box with a girls holding a sword on the front if it, ready to defend the cooler (though it is not quite clear from whom):

 

The back of the box contains a brief list of cooler’s key features and the schematics of the airflow it creates. At the top of the package there is a small box with accessories bundle. Among them you will find retention kits for LGA775 and Socket AM2(+)/AM3, washers and screw-nuts, manual and warranty slip, and a 1 g pack of SilMORE thermal compound.

The cooler is surprisingly small: only 124x96x70 mm (including the heatpipes). And the heatsink itself fits completely under the 92x92x25 mm fan:

 

ISGC-100 is only 70 mm tall and weighs 335 g. despite its modest size, this cooler uses three copper heatpipes 6 mm in diameter that come out of the copper base in two directions and pierce the aluminum heatsink:

In fact, the heatsink is so small, that it seems there are too many heatpipes for it:

 

The heatsink consists of 37 aluminum plates. Each is 0.45 mm thick and the gap between them measures 2 mm:

The heatpipes are soldered to the copper base plate, which is a little less than 2 mm thick. The base surface is finished pretty well, it lacks just a bit to be as shiny as a mirror:

The base is also impeccably even:

Thermaltake ISGC-100 is equipped with a 92 mm nine-blade fan that is a smaller version of an ISGC 12 fan, which we have tested before:

It uses PWM-method to control its fan rotation speed in the interval from 600 to 1600 RPM creating 37 CFM maximum airflow and running as loud as 17 dBA. The claimed static pressure of this fan is 1.22 mmH2O. The fan uses a fluid dynamic bearing that should last for at least 50,000 hours. The maximum power consumption of this fan is only 1 W.

Thermaltake ISGC-100 is compatible with LGA775/1156 and Socket AM2(+)/AM3. Unfortunately, this cooler doesn’t support LGA1366 I was a little surprised that a small and lightweight cooler like that is installed onto LGA775/1156 using a screw-on retention that goes through the board. However, the explanation is evident: the cooler is so compact that it is simply impossible to use standard plastic push-pins on it. That is why Thermaltake engineers could only use mounting spindles and large screw nuts that are tightened at the bottom of the PCB. As for the AMD mainboards, the cooler is installed with a common swing-clip with a locking tab.

During cooler installation onto a Socket AM2 mainboard, it will block the first memory DIMM slot, no matter which way of the two possible ones you turn it (even if there is a memory modules without the heat-spreaders on it):

Also, this cooler didn’t fit at all on one of the LGA775 mainboards that I had at my disposal, because the cooler interfered with the chipset and MOSFET heatsinks:

 

Mainboards designed for HTPC systems, which are the primary target platform for Thermaltake ISGC-100, often don’t have any large heatsinks in the area around the processor socket that is why there shouldn’t be any problems like that there. And in our case, we used Albatron NF 650i Ultra for tests on an LGA775 platform.

No matter how shocking it may seem to you, but the MSRP of this little fellow is $49.99. What are they asking so much money for? It is really pleasing that according to Google product search you can now buy Thermaltake ISGC-100 cooler for $33, although even this price is way too high, in my opinion.

Thermaltake ISGC-200 (CLP0538)

The next new cooler from Thermaltake – ISGC-200 (CLP0538) - belongs to the same family, but is a tower-cooler. It comes in a larger box, but the design and information on it are exactly the same as by ISGC-100:

 

The accessories bundle includes the same components, except for the screw-on retention for LGA775/1156 that has been replaced with the traditional push-pins:

Thermaltake ISGC-200 is a cooler of larger size than ISGC-100, but is nevertheless quite compact: it measures 90x96x140.5 mm and weighs only 475 g.

The cooler heatsink consists of two arrays with a 92 mm fan between them:

 

Our experience with super-coolers suggests that this design proves very efficient (although in coolers of different size). I wonder how the small ISGC-200 will perform?

The new solution uses three copper heatpipes, each 6 mm in diameter, that are soldered to the 2 mm copper base plate. These heatpipes hold two separate arrays of aluminum plates (a total of 68) that are pressed onto the heatpipes. Each plate is 0.45 mm thick and they at 2.5 mm distance from one another:

 

There is an additional aluminum heatsink above the heatpipes in the base of the cooler that also serves as support for the retention brackets during cooler installation.

If you remove the fan you can clearly see the sides of the heatsink plates of variable height on entry and exit of the fan airflow:

Although this is no new solution, it is a smart one in this case, especially since ISGC fans can’t boast high static pressure. By the way, this cooler uses the same exact fan as the ISGC-100.

The base surface of ISGC-200 is finished exactly the same way as that of the previous solution, but unfortunately, it turned out salient:

As a result, the thermal compound imprints of AMD and Intel processor heat-spreaders turned out poor:

 

Nevertheless, we tested the cooler as is, and didn’t do anything to level out its base. And at this point I would like to say that the results turned out pretty interesting. However, we are going to talk about it a little later, and now let’s take a look at the ISGC-200 inside the system case:

As you can see, the cooler doesn’t block the memory slots in any way, even if you are using modules with tall heat-spreaders. I don’t know if the cooler will do well in the market with the recommended retail price of $54.99, but luckily it is already available in retail for as low as $37.

Xilence Icebreaker 775

The last miniature cooling solution that we will discuss today is made by Xilence and has a promising “icebreaker” name. Well, let’s find out what kind of beast is it…

The cooler comes in a clear plastic blister packaging, which is just as small as the cooler itself:

Despite this fact, the cooler is pretty well protected against external damages. The cardboard insert with the model name on it lists the brief cooler specifications, mentions the PWM support and compatible CPU sockets – one. Xilence Icebreaker 775 comes bundled only with no-name gray thermal compound:

Oh, yes, there is also a plastic backplate attached to the cooler.

Xilence Icebreaker 775 measures 110x100x81 mm and weighs 419 g. it uses four copper heatpipes, 6 mm in diameter:

 

The heatpipes have 61 aluminum plates pressed to them. Each plate is 0.35 mm thick and the distance between them is about 1.4-1.5 mm. There is a frameless 92x92x32 mm fan installed in a plastic frame above the heatsink:

 

In fact, the fan is 25 mm tall, and the 32 mm height in the specs takes into account the plastic mounting spindles of the retention frame (for purely marketing reasons):

The heatpipes come out of one side of the cooler base that is why Icebreaker 775 will be way more compact than Thermaltake ISGC-100 discussed above:

 

The heatpipes lie in the grooves in the copper base plate and are soldered to them.

This is what the heatsink looks like without the fan:

You can even see traces of soldering on the top of the heatpipes in the base through the heatsink plates. The base surface is finished quite nicely, although there is no polishing to mirror shine here:

However, most importantly it is even over the entire surface area except the very edges that do not contact the LGA775 processors heat-spreader anyway.

The frameless cooler fan has no marking of any kind:

According to the specifications, it is PWM controlled, but there is no mention of the minimum speed and the maximum is claimed to be 2800 RPM. At this speed the fact creates 55.78 CFM airflow and should generate 19-22 dBA of noise. We don’t know how long the fluid dynamic bearing of the fan should last. Looks like the only distinguishing feature of this fan is the use of soft silicon mounting spindles that absorb the rotation vibrations from the frame and the blades.

Since Xilence Icebreaker 775 is designed only for LGA775 platforms the entire installation procedure consists of attaching the cooler to the backplate on the back of the mainboard PCB. However, I failed to accomplish it on my DFI LANPARTY DK X48-T2RS mainboard: I experienced the same exact problems as with Thermaltake ISGC-100:

 

The latter could at least be installed onto the AMD platform, but this didn’t work with Xilence Icebreaker 775, because it was only compatible with LGA775. Therefore, I had to work with Albatron NF 650i Ultra mainboard during the cooling efficiency tests, just like with Thermaltake ISGC-100.

Technical Specifications and Recommended Pricing

 

Testbed and Methods

Today all tests were performed on two platforms: with an Intel CPU and an AMD CPU. Therefore, we assembled two testbeds with the following configurations:

All tests were performed under Windows 7 RTM x64. We used the following software during our test session:

The CPU was loaded with two consecutive Linpack test runs with the settings as indicated above. The stabilization period for the CPU temperature between the two test cycles was about 10 minutes. We took the maximum temperature of the hottest processor core of the four for the results charts. The ambient temperature was checked next to the system case with an electronic thermometer with 0.1 °C precision that allows monitoring the temperature changes over the past 6 hours. During our test session room temperature was unusually high and stayed at 22.1-22.4 °C.

The best competitors for our today’s testing participants will be the so-called “boxed” coolers that come with retail AMD and Intel processors. Unfortunately, I didn’t have any at my disposal at the time of tests. That is why I could only use the price point as the criterion for finding the rivals for today’s race and included an inexpensive Cooler Master Hyper 212 Plus that costs only $29.99. Of course, its size is incomparable with the size of our today’s heroes and it will hardly fit into an HTPC case. But that was the only choice I had. AT least it will allow us to determine the advantage a full-size tower cooler has over the miniature solutions.

Since the noise level is of utmost importance for home theater PC systems, we tested the coolers not only at their maximum fan rotation speeds, but also in pretty quiet mode at 1000 RPM, which is nearly silent operation for 92 mm Thermaltake or 120x12 mm Scythe fans. Cooler Master Hyper 212 Plus was tested at quiet 850 RPM and at maximum 2220 RPM fan speed.

Performance

LGA775 Platform

During this test session we managed to overclock our 45nm quad-core processor with the multiplier set at 10x to 3.65 GHz using the weakest cooling system of the today’s testing participants. The nominal processor Vcore was increased to 1.45 V in the mainboard BIOS.

All other parameters available in the mainboard BIOS and connected with CPU or memory overclocking remained unchanged (set to Auto).

Let’s check out the obtained results:

On the one hand, Thermaltake ISGC-100 was the least efficient on Intel platform. But if we remember that the mission of this cooler is the use in compact HTPC systems, where you will never see an overclocked quad-core processor loaded with Linpack, then this little fellow really inspires some respect. Scythe BIG Shuriken looks even better, as it outperforms its competitor by 3 °C in both test modes. But we were mostly surprised with the results demonstrated by the little Thermaltake ISGC-200 tower that turned out very efficient even in quiet mode. Xilence Icebreaker 775 yielded about 2 °C to the latter in quiet mode and outperformed it by the same number of degrees at maximum fan rotation speed, which isn’t surprising at all, because the icebreaker’s fan rotates by about 1000 RPM faster. And the results of the included Cooler Master Hyper 212 Plus show 5-6 °C advantage of a full-size tower cooler over the compact systems in both test modes.

Socket AM2+

Our Phenom II X4 940 BE processor with the weakest cooling solution of the today’s tested participants could be overclocked to 3.63 GHz with the Vcore increased to 1.475 V:

Here are the results:

The situation on AMD platform hasn’t changed a bit: in terms of cooling efficiency the products lined up in the same exact order as on the Intel platform. The interesting thing is that the gaps between the participants have grown bigger, although it may seem that AMD Phenom II X4 working at such frequency and voltage should dissipate less heat than Intel Core 2 Extreme. Once again the best here was Thermaltake ISGC-200 even though its base surface was far from even.

This time we didn’t measure the coolers acoustic performance, but I would like to share my personal subjective observations about the maximum fan rotation speed mode. The quiet mode that we used during this test session is indeed very quiet and hence there is nothing to comment on here. As for the maximum fan mode, Scythe BIG Shuriken appeared to be the quietest of all. It must be its thin fan with non-aggressive blades that determined this outcome. The next coolers are two Thermaltake ISGC solutions, with the 200 model being slightly quieter than 100 one. And the noisiest of all four is Xilence Icebreaker 775, as 2780 RPM is pretty loud even for a 92 mm fan on silicon shock-absorbing mounting spindles. The fan makes a slight whistling sound at this rotation speed. Up until 1650-1700 RPM Xilence Icebreaker 775 operates pretty quietly.

Conclusion

Our today’s test of miniature cooling solutions performed on two different platforms showed that these systems can also cope with overclocked processors, although they may rarely have to do it. In other words, if you are really looking for a highly-efficient CPU cooler, then these little fellows are not for you. However, they will be a perfect fit for the increasingly popular home theater computer systems and I am sure that they will perform superb even in quiet mode. That is why it is natural that they appeared in the market.

And speaking of the particular solutions tested today, I personally liked Thermaltake ISGC-200 best of all, as it proved to be the most efficient cooler for its size of the four. Its younger brother, ISGC-100, can’t boast the same efficiency. Moreover, it is not quite clear why the recommended retail prices of Thermaltake coolers are so high. There is another top-cooler that looks really attractive here – Scythe BIG Shuriken, which has fewer compatibility issues than ISGC-100. I am sure that some will like Xilence Icebreaker 775, but you should keep in mind that it is only compatible with LGA775 platforms and hence has much narrower target audience.