by Sergey Lepilov
10/27/2011 | 02:34 PM
The air-coolers for processors designed in the so-called “top” form-factor (when the airflow is directed towards the mainboard surface) are about 10 times rarer than the tower-coolers. Even if you take, for example, our reviews, then there will only be one top-cooler review per every ten or so tower-cooler reviews. The reasons are obvious: lower cooling efficiency, or, to be more exact, its greater dependence on the system case and the airflow organization inside it. Moreover, the efficiency difference compared to the tower coolers is so dramatic that the only advantage of the top-coolers, namely the cooling of the components in around-the-socket area, is often not significant enough to make the choice of a top-cooler preferable. Today, Cooler Master Co., Ltd. and Deepcool Industries Co., Ltd. will try to confirm or shatter these statements with their new products. Let’s meet them.
The first cooler is called GeminII S524. It ships in a cardboard box designed in Cooler Master’s traditional color scheme and style:
The cooler model name and image are on the front of the box, while its technical specifications, description of the key features, dimensions and another photograph are spread over the back and side panels of the packaging:
The cooler sits between the polyurethane foam inserts, which will protect it against transportation mishaps. The accessories including retention kits, thermal paste and installation instructions in 20 languages are neatly packed into an individual smaller box:
The cooler is made in
GeminII S524 is a continuation to the relatively compact GeminII S model and looks very similar to it:
The new cooler became 17.7 mm taller, 20 mm wider and 23.2 mm deeper. Now its dimensions are 105.4 x144.2x144 mm (LxWxD):
As for its weight, it gained only 34 grams reaching a total of 594 g, which is nothing, according to today’s standards :)
As for differences in design, we should definitely mention that the entire heatsink and all heatpipes of this cooler are nickel-plated. There are still five heatpipes 6 mm in diameter. Moreover, the heatsink profile makes this cooler compatible with memory modules featuring tall heat-spreaders:
The heatsink array consists of 62 fins, each 0.4 mm thick and pressed against heatpipes at 2 mm distance from one another. The top of the heatsink is framed with steel plates. The fan is attached to this plate with four retention screws:
If you remove the fan, you will see that when the airflow enters the heatsink, the fin edges haven’t been optimized in any way:
And I think they could use some optimizations, because it is very nearsighted to rely solely on the high fan pressure.
However, when it comes to the cooler base, we couldn’t find any flaws here. The heatpipes are all in special grooves about 1 mm away from one another and are soldered to the base plate:
The thinnest part of the copper nickel-plated base is 4 mm. Its contact surface is exceptionally even and gets a solid “B” for the finish quality:
Cooler Master GeminII S524 is equipped with a seven-blade 120x120x25 mm fan:
This fan is constructively very similar to the new XtraFlo (R4-XFBL-22PR-R1), which impeller generates “extra high airflow”, according to Cooler Master. However, the fan on GeminII S524 has different speed characteristics and is built on a regular frictionless bearing with 40,000 hours MTBF instead of a brand-new DynaLoop bearing used in the original XtraFlo. The fan supports PWM rotation speed control feature that is why its speed may be adjusted automatically between 800 and 1800 RPM creating 34.2-77.7 CFM airflow and generating 15.1-31.6 dBA of noise. The static pressure of the fan is also mentioned among the cooler specifications and equals 0.43-2.46 mmH2O. The maximum fan power consumption shouldn’t exceed the official 3.9 W, and our tests showed it at 3.7 W.
Another peculiarity of the Cooler Master GeminII S524 cooler is its ability to accommodate a 140x140x25 mm fan, which will be fastened using special additional holes in the corners of the steel frame:
Overall, we would like to point out that the use of screws for attaching the cooling fan to the heatsink is gradually being replaced by contract-free method with silicone mounts, which also allow lowering vibrations and noise. It’s a pity that Cooler Master hasn’t yet adopted this simple and at the same time very efficient approach.
GeminII S524 is compatible with all contemporary platforms, and once LGA2011 mainboards come out, Cooler Master promises to supply all of their current coolers with the proper retention kits. The installation procedure is technically simple and is described in detail in the following manual (PDF-file, 1.72 MB). The cooler is fastened with screws, or threaded mounts, to be more exact. Their position in the retention loops of the brackets can be adjusted to match the right processor socket:
These brackets are screwed on to the base of the cooler. For extra convenience it is better to put the cooler on the desk upside down, place the mainboard on top of it and then tighten the screw-nuts on top of the backplate:
This procedure is not really difficult, but mostly rather inconvenient, but the actual retention is extremely secure and presses the cooler very tightly against the CPU.
Since GeminII S524 is so compact at the base and provides 47 mm clearance for the electronic components with tall heatsinks, it won’t ever interfere with voltage regulator parts or DIMMs featuring tall heat-spreaders:
If you install the cooler with the heatpipe ends facing up, one of its sides will go 40 mm past the edge of the mainboard, which is important to keep in mind in small system cases where the PSU is located in the upper part of the case:
However, this distance depends not only on the cooler, but also on the distance between the processor socket and the edge of the mainboard PCB.
The manual doesn’t point out the preferred installation position for GeminII S524, but we did run our traditional preliminary tests for two possible installations – with the heatpipes going perpendicular to the memory DIMM slots and parallel to them (with the ends of the heatpipes facing up):
It turned out that GeminII S524 efficiency doesn’t depend on the way it is installed on the processor and in both cases the cooler works… but we will get to that a little later. This is what the new Cooler Master GeminII S524 looks like if we replace its default 120 mm fan with a 140 mm Scythe Slip Stream:
I believe we’ve told you enough about our first testing participant, and now let’s spend some time on the second newcomer.
Deep Cool Company has always pleasantly surprised us with their packaging. The new Fiend Shark cooler is also no exception. It arrived in a box the size of a cooler:
However, it wasn’t particularly informative, providing nothing particularly important besides the scarce technical specs of the cooler:
When we opened the box, we didn’t find the traditional soft polyurethane foam inserts, but a plastic molded casing with a lid holding the cooler and two boxes with accessories beneath it:
Looks very impressive, doesn’t it? They could have probably done more by making an individual compartment for each washer. :) Anyway, Fiend Shark is manufactured in China and priced at $59. It comes with a 1-year warranty.
Deep Cool Fiend Shark definitely looks more interesting than the Cooler Master cooler we have just discussed. Take a look:
Very beautiful cooler. It is also considerably larger and measures 157x156x131 mm. Fiend Shark weighs 1142 grams.
The blue fan and inserts of matching color on the sides are the only non-nickel-plated parts of the new Fiend Shark:
The cooler is built on six copper heatpipes 6 mm in diameter, which go through the copper base plate:
The heatsink body is composed of a total of 150 fins soldered to the heatpipes at 2 mm distance from one another. They create six independent heatsink arrays – 25 fins in each:
Since Deep Cool engineers had to lead the heatpipes into the heatsink arrays somehow, there appeared quite a bit of empty space in the center of the heatsink making its effective surface significantly smaller than it could have been otherwise. Moreover, we would like to point out that two hottest heatpipes going through the very center of the cooler base are assigned to the two smallest heatsink arrays, which will hardly be good for efficient heat transfer and dissipation.
There are special grooves cut out in the cooler base plate and the heatpipes are soldered to them:
The thinnest part of the copper nickel-plated base beneath the heatpipes is 3mm thick and the gaps between the heatpipes in the base do not exceed 1.5 mm.
The contact surface of the base is decently finished. Although you can visually identify machine marks on the base surface, it is very smooth:
It is also impeccably even:
Deep Cool Fiend Shark is equipped with 140 mm UF140 fan, which our regular readers already know from one of our fan roundups:
Actually, the Fiend Shark UF140 fan has been slightly “enhanced”: its maximum rotation speed is 1400 RPM, instead of 1200 RPM by the original model. At this speed it should create 80.28 CFM airflow and generate no more than 32 dBA of noise. The minimal fan speed is 700 RPM thanks to PWM support, and the level of generated noise in this case is 18.2 dBA.
The fan is attached to the heatsink with two wire clips catching on to the retention holes in the fan frame:
Since UF140 has special anti-vibration lining, there is no need for any additional vibration-absorbing solutions between the fan and the heatsink.
The new Deep Cool Fiend Shark is compatible with all contemporary processors and is installed with a backplate on all mainboards except for the LGA1366:
As for the LGA1366 installation, instead of a backplate you will have to use rubberized screw-nuts with pads and threaded bushes that go into them:
These are the retention brackets that you will have to attach to the cooler itself before installation:
After that you place the cooler on top of the CPU and tighten the spring-screws. The retention provides pretty high pressure hold, although it is somewhat inconvenient to tighten the screws under the horizontal heatsink. You can find more details on the cooler installation in this manual (PDF-file, 1.6 MB).
This is what Deep Cool Fiend Shark looks like when it is installed onto a mainboard:
The distance between the cooelr base and the lowest heatsink fin is 55 mm.
It turned out that the cooling efficiency of Fiend Shark doesn’t depend on the way it is installed, and it looks very catchy inside the system case:
Now let’s compare the coolers technical specifications side by side and move on to the actual tests.
We performed all cooler tests inside a closed system case. Here is our testbed configuration:
We overclocked our six-core processor (with its default non-lapped heat-spreader) with the multiplier set at 25x and “Load-Line Calibration” (Level 2) enabled to 4.2 GHz. The nominal processor Vcore was increased to 1.35 V in the mainboard BIOS:
Turbo Boost and Hyper-Threading technologies were disabled during our test session. The memory voltage was at 1.6 V and its frequency was 1.38 GHz with 7-7-7-16_1T timings (Extreme profile). All other parameters available in the mainboard BIOS and related to CPU or memory overclocking remained unchanged.
All tests were performed under Windows 7 Ultimate x64 SP1 operating system. We used the following software during our test session:
So, the complete screenshot during the test session looks as follows:
The CPU was loaded with two consecutive CST test runs with the settings as indicated above. The stabilization period for the CPU temperature between the two test cycles was about 8-10 minutes. We took the maximum temperature of the hottest CPU core for the results charts. Moreover, we will also provide a table with the temperature readings for all cores including their average values. The ambient temperature was checked next to the system case with an electronic thermometer with 0.1 °C precision that allows hourly monitoring of the temperature changes over the past 6 hours. The room temperature during our test session varied between 24.6-25.2 °C.
The noise level of each cooler was measured between 1:00 and 3:00 AM in a closed room about 20 m2 big using CENTER-321 electronic noise meter. The noise level for each cooler was tested outside the system case when the only noise sources in the lab were the cooler and its fan. The noise meter was installed on a tripod and was always at a 150 mm distance from the cooler fan rotor. The tested cooling systems were placed at the edge of the desk on a sheet of polyurethane foam. The lowest noise reading our noise meter device can register is 29.8 dBA and the subjectively comfortable noise level in these testing conditions was around 36 dBA (do not mix it up with low noise level). The fan(s) rotation speed was adjusted in the entire supported range using our in-house controller by changing the voltage with 0.5 V increment.
We are going to compare our newcomers against the modest and inexpensive Thermalright True Spirit cooler ($30). Te cooler was tested with its default 120 mm fan.
Now let’s check out the obtained results.
The results of our cooling efficiency tests are summed up in the diagram and table below:
The difference in cooling efficiency between a tower-cooler and two new top-coolers is simply shameful. Even at minimal 800 RPM rotation speed, True Spirit is 1°C more efficient than the best top-cooler of the two tested at the maximum fan speed. I don’t even want to mention the best results of the Thermalright cooler, when its advantage increases to impressive 13°C. The only thing we can do at this point is compare the efficiency of Cooler Master GeminII S524 against that of Deep Cool Fiend Shark. And here the Cooler Master product wins, though mostly due to a faster fan than due to some unique design advantages. I would also like to add that it makes sense to use a 140 mm fan only at 1000-1200 RPM, because after that the results even out for the most part.
Heavier and larger Fiend Shark cooler didn’t prove up to our expectations, but once we removed the case side panel, it really started to shine:
Deep Cool Fiend Shark (1490 RPM)
Closed system case
Case side panel removed
As we see, this simple modification lowers the maximum CPU temperature by as much as 8°C. Moreover, with the removed case side panel we managed to overclock our test CPU under Fiend Shark cooler even more:
Deep Cool Fiend Shark (1490 RPM)
However, if we look at the results of the two today’s newcomers compared with the products we have tested over the past few months, the new top-coolers will make a very modest appearance:
Well, they are actually in the very end of the ranks.
We measured the acoustic performance of our today’s testing participants in the entire rotation speed range of their fans. The results are summed up on the following graph:
Looks like the unimpressive efficiency is now paired with a far not the lowest level of noise. For example, Fiend Shark can be considered acoustically comfortable only up to 950 RPM, and GeminII S524 – up to 1050 RPM. As you have just seen, both coolers do not do that well at these fan speeds.
Unfortunately, we didn’t see a miracle again. The cooling efficiency of top-coolers remains at the same level, i.e. significantly lower than the efficiency of the tower-models. In fact, of Cooler Master GeminII S524 and Deep Cool Fiend Shark didn’t bring anything new to the table and just joined the army of processor coolers available in the market today. Both of them are universal, have very reliable retentions, are compatible with memory DIMMs featuring tall heat-spreaders, should provide efficient cooling to the components around the processor socket, and Fiend Shark also looks great on top of all of the above. To be fair I have to say that the latter is pretty expensive and is not justified by its efficiency. So, if your system case side panel doesn’t have a large fan or at least a meshed window, then you should better consider tower-coolers.