by Sergey Lepilov
12/16/2008 | 11:19 PM
If we look at the problem of creating efficient cooling for contemporary CPUs from an overclocker’s standpoint, we can say that in reality there is no problem at all. And it is certainly not the low heat dissipation of overclocked processors that explains it (on the contrary, overclocked CPUs run pretty hot). The reason for that is the fact that currently there are so many highly efficient coolers out there that even a fastidious user will find a good one easily. Here are a few coolers that I could think of on the spot: Thermalright IFX-14 / SI-128 (SE) / Ultra-120 eXtreme, ZEROtherm Nirvana NV120 Premium / ZEN FZ120, Cooler Master GeminII / Hyper Z600 / Hyper 212 / GeminII S, Ice Hammer IH-44xx, Noctua NH-U12(F) / NH-C12P, ASUSTek Silent Square EVO, XIGMATEK HDT-D1284, Scythe ZIPANG… As for the good old buddies, you shouldn’t forget about such great solutions of all times as Thermaltake Big Typhoon (VX), Scythe Infinity / Ninja (Copper/2), Tuniq Tower 120 (120-LFB), Enzotech Ultra-X and Zalman CNPS9700 LED. We’ve got about two dozen right here already. Pretty impressive, isn’t it?
And now let’s remember that there have recently arrived or will be coming out shortly such monsters of cooling as Cooler Master V12, V10 and V8, Zalman CNPS9900, Thermalright AXP-140, Thermaltake BigTyp14 ? Spin Q, ASUSTek Axe Square and Royal Knight. These are just a few contestants for the Super-Cooler title. As you can see, the manufacturers have no plans to be done with air cooling just yet, even though it seems that the potential of this technology has already been completely exhausted.
Today we are going to either strike out from the contestants list or add to the above listed recognized super-coolers two new solutions from ASUSTeK Computer Inc. and ARCTIC COOLING GmbH: Lion Square and Freezer Xtreme.
A not very big cube-shaped box that ASUS Lion Square comes in has a convenient carry handle and a cut out window in the front:
You can find the details about the cooler structure involving 8mm heatpipes. According to the manufacturer, it ensures a 30% advantage during heat transfer from the heatpipes to heatsink plate arrays compared to cooling solutions using 6mm heatpipes:
The back of the box bears technical specifications of the new cooler, the list of supported processor types and compatible sockets.
There is a clear plastic casing inside the cardboard box. It securely holds the cooler inside. By the way, we noticed that the plastic cracked in a few spots, probably during transportation, but the cooler remained intact:
At the bottom of this casing you will find a backplate for LGA775 platforms and a small box with other accessories. The accessories bundle includes two screw-retentions for LGA775, a swing-slip for AMD K8 and K10 processors and ASUS thermal compound:
The “lion” design reminds us of the previously reviewed ASUS Silent Square EVO cooler:
ASUS Lion Square measures 126 x 126 x 144 mm and is based on four copper heatpipes 8mm in diameter that go through copper base. The heatpipes hold 0.35 mm thick aluminum heatsink plates of two kinds. They are spaced out at a 2mm distance from one another:
The heatsink is topped with a shaped plastic cover decorated with an image of a lion head from the Japanese mythology:
The same symbolism is present in the mere name of the cooler – Lion Square. Lion is the symbol of power, courage and luck, and the second word “Square” is an already familiar abbreviation that reads as follows:
The top plastic cover can be easily removed (just remove a couple of screws first):
Here we see the fan retention frame that can also be easily dismounted. Finally we can meet up close and personal with the ASUS Lion Square heatsink:
As we see, the stamped heatsink plates have a special hole for the fan in the middle of them. In other words, the cooler heatsink doesn’t consist of two independent arrays, as you may have thought at first, but of 29 solid aluminum plates with a hole in the center. There are additional 5 plates added on each side at the bottom of the heatsink in the direction of the airflow from the cooling fan:
The edges of all plates are slightly bent downwards, which helps direct the airflow towards the mainboard PCB, according to the manufacturer. This way, it also cools the power elements in the area around the CPU socket. Note also that there are a few pairs of elongated holes in each heatsink plate in the way of the fan airflow. It is fairly hard to tell what they stand for. I dare suppose that this way engineers were trying to lower the noise generated by the fan at its maximum rotation speed.
The heatpipes are soldered to the grooves in the cooler base plate. They are also topped with a retention plate:
The base is very well finished. It is extremely even, although there is no mirror-shining polishing on its surface:
As I have already said, the base is very even: the thermal compound imprint from the evened out processor heat-spreader was nearly ideal:
ASUS Lion Square is equipped with a seven-blade 92 x 92 x 25 mm fan made by SUNON Inc.:
MagLev KDE1209PTVX fan model uses a so-called “Vapo” bearing. In reality, it is a type of slide bearing with longer life span (due to longer lasting materials used). The fan rotation speed can be adjusted using pulse-width modulation algorithm (PWM) between ~800 (according to the monitoring data) and ~2300 RPM. The cooler specifications, however, say that ASUS Lion Square shouldn’t be noisier than 18 dBA. These must be the readings taken at the minimal fan rotation speed, or the actual measuring methodology is truly “unique”. The fan is equipped with four blue LEDs.
The cooler installs very simply and intuitively. You can download the installation instructions from the official web-site (PDF file, 3.64 MB), but I doubt you will actually need them. If you want to install ASUS Lion Square onto an AMD K8 or K10 platform, you won’t need to remove the mainboard from the system case, because the cooler is pressed against the processor heat-spreader with the bundled swing-slip inserted into the slits in the cooler base:
However, if you have an LGA 775 mainboard, it will have to get out of the system case, because ASUS Lion Square will be attached to a backplate with two brackets with spring-screws:
In this case the cooler is pressed very firmly against the CPU heat-spreader, which works great for efficient heat transfer.
The distance from the lower heatsink plate of the cooler to the mainboard PCB is 39 mm, so the lion will not interfere with any components around the processor socket:
The cooler fits pretty compactly into the system case and doesn’t conflict with tall memory module heat-spreaders or relatively massive chipset cooling system of our new mainboard:
The blue fan highlighting is very pale and you can barely see it even in the dark, since the fan is hidden inside the heatsink.
ASUS Lion Square is priced at around $55.
Freezer Xtreme is a long awaited new cooler from the Swiss Arctic Cooling Company that hasn’t released anything really revolutionary for CPU cooling for quite some time already. The cooler package is a true representative of minimalism – clear plastic casing shaped exactly as the cooler with two cardboard inserts inside:
The information on these inserts tells about the cooler key features, lists its technical specifications and compatible processor sockets, promises 6-year manufacturer warranty. There are also two diagrams comparing the efficiency of the Freezer Xtreme cooler vs a standard “boxed” cooler from Intel. According to them, Arctic Cooling solution is 18°C more efficient and 1.25 Sone quieter than Intel Core 2 Duo reference cooler (not quite clear which generation, though) when tested on QX9770 CPU. The cooler is packed securely and the plastic is very sturdy, so the cooler is very well protected against any possible transportation mishaps.
The cooler is bundled with a plastic frame for LGA 775 mainboards, two retentions for contemporary AMD mainboards, and installation manual and a set of screws and plastic locks:
The simplicity of the cooler design wins your heart, but despite that, Freezer Xtreme looks very elegant:
The cooler is not very big. It measures only 130 x 100 x 131 mm and weighs 608 g. the cooler is built on four copper heatpipes 6mm in diameter. These heatpipes hold two aluminum heatsink arrays, 51 plates in each. The gap between the heatsink plates is pretty small and measures only ~1.8mm:
The plates are also quite thin (~0.25 mm), however, their edges are nicely finished, so you won’t cut yourself. The sides of the cooler are covered with a plastic panel with a paper sticker bearing an Arctic Cooler logo on it:
It looks pretty simplistic, even cheap, but it does its job 100% well, namely ensures more rational use of the airflow from the fan.
The cover at the top of the heatsink, the fan is attached to, also bears a paper logo sticker:
The fan can be easily removed without any tools, as the cover click-locks to the sides of the heatsink. So, you can really make sure that the heatsink consists of two identical parts:
Arctic Cooling Freezer Xtreme has a 120 x 120 x 25 mm fan in a round frame:
Look at the interesting shape of the fan blades: they are in fact bi-directional. It means that the fan should be equally efficient taking air flow in and exhausting it out. However, it isn’t the case for traditional 120-mm fans that work better for air intake. Considering the cooler heatsink design, using a fan of this type is absolutely justified and logical, because it has to create airflow cooling both parts of the heatsink. The fan rotation speed is adjusted using PWM method from ~500 to ~1500 RPM. It creates 35.7 – 60.7 CFM airflow and generates 17 – 21 dBA of noise respectively.
The copper base plate has special grooves for the cooler heatpipes that are soldered to it, just like ASUS Lion Square. However, the thermal interface is pre-applied to the cooler base:
Arctic Cooling solutions use highly efficient Arctic MX-2 thermal compound, which is at least as good as Arctic Silver 5 that I am constantly using in my tests:
The finish quality on the cooler base is typical of Arctic Cooling solutions: no polishing, but also nothing to actually pick on:
The copper base surface is very even. There were no gaps on the thermal compound imprints from the processor heat-spreader obtained with Arctic Silver and MX-2 interfaces, although I have to admit that the manufacturer pre-applied too much paste:
Arctic Cooling Freezer Xtreme can be installed on all supported platforms without removing the mainboard from the system case. For LGA 775 mainboards you need to install the plastic retention frame using the retention holes around the processor socket:
Then it is locked in place with enclosed plastic “studs”:
And then you have to remove the fan and install the heatsink onto the CPU. Use two screws to attach it firmly to the plastic retention frame, as the installation manual suggests (you can download it here, 848 KB). After that all you have to do is insert the fan between the heatsink arrays and clip the plastic cover to its sides.
Arctic Cooling Freezer Xtreme looks as follows inside a system case:
The distance from the lower heatsink plate to the mainboard is only 2 mm shorter than that by ASUS Lion Square: 37 mm. The cooler is priced at $44.95, which is a very democratic price point compared to other super coolers. Although Arctic Cooling Freezer Xtreme hasn’t yet become one, at least until we see the results of our cooling efficiency tests…
The technical specifications and recommended retail price of the cooling solutions discussed today are summed up in the following table:
We tested new coolers and their competitor in two modes: in an open testbed when the mainboard sits horizontally on the desk and the coolers are installed vertically, and in a closed testbed with the mainboard in vertical position. Since today we are only testing tower-type coolers, they were installed the same way inside a system case: with the airflow directed towards the back of the case with a 120-mm case fan for air exhaust mounted on the panel.
Our testbed was identical for all coolers and featured the following configuration:
All tests were performed under Windows Vista Ultimate Edition SP3. SpeedFan 4.34 was used to monitor the temperature of the CPU and mainboard chipset, reading it directly from the CPU core sensor and to monitor the rotation speed of the cooler fans:
The mainboard’s automatic fan speed management feature as well as CPU power-saving technologies were disabled for the time of the tests in the mainboard BIOS. The CPU thermal throttling was controlled with the RightMark CPU Clock Utility version 2.35.0:
The CPU was heated up in two modes. First we used OCCT (OverClock Checking Tool) version 2.0.0a in a 20-minute test with maximum CPU utilization, during which the system remained idle in the first 1 and last 4 minutes of the test:
We have also created additional load with 15 runs of IntelBurnTest v1.6 (by AgentGOD) that uses Linpack 32-bit stress-test algorithm:
I performed at least two cycles of tests in both test modes and waited for approximately 20 minutes for the temperature inside the system case to stabilize during each test cycle. The stabilization period in an open testbed took about half the time. Despite the stabilization period, the result of the second test cycle was usually 0.5-1°C higher. We took the maximum temperature of the hottest processor core after two test cycles for the results charts (if the difference was no bigger than 1°C – otherwise the test was performed at least once again).
The ambient temperature was checked next to the system case with an electronic thermometer that allows monitoring the temperature changes over the past 6 hours. During our test session room temperatures were pretty low and varied between 19.0~19.5°C. It is used as a starting point on the temperature diagrams. Note that the fan rotation speeds as shown in the diagrams are the average readings reported by SpeedFan, and not the official claimed fan specifications.
We chose the cooler of the same type to compete against ASUS Lion Square and Arctic Cooling Freezer Xtreme in our today’s test session. It will be Thermalright IFX-14 (without the additional HR-10 heatsink) with one Scythe Slip Stream 120 fan installed between the heatsink arrays and working at ~910 RPM and 2040 RPM. Of course, the cooler we chose is more expensive than both today’s main characters (Freezer Xtreme sells at half the price of the Thermalright solution). However, it is especially interesting to compare our today’s testing participants against Thermalright IFX-14, not only because it is the best air cooler out there, but also because they all have very similar design. Therefore, I believe it makes more sense to perform this particular comparison rather than look for a similarly priced competitor. However, we have already tested IFX-14 quite a few times at Xbit Labs, so you will be able to easily draw parallels between the today’s and previous results.
During OCCT tests inside a closed system case using the “weakest” cooling system of the today’s testing participants we managed to overclock our quad-core processor to 3.95 GHz (+31.7%). The nominal processor Vcore was increased to ~1.55 V in the mainboard BIOS (+24.0%).
According to monitoring data, the CPU core voltage was a little lower than what was set in the mainboard BIOS, namely around 1.50~1.54 V. When we launched Linpack 32-bit we had to lower the frequency to 3.85GHz to ensure stability at the same Vcore setting:
Before we start checking out the obtained results we need to draw your attention to Arctic Cooling Freezer Xtreme cooler. The reason for that is pretty simple: the fan of the cooler sample we received for testing wasn’t working correctly. No matter if we enabled or disabled PWM adjustment mode (we even physically disconnected the fourth pin in the four-pin connector), the fan rotation speed remained constant and stayed at the minimal ~500 RPM. This speed ensured extremely quiet operation of the cooler, but was critically low for efficient cooling of an overclocked CPU. In this quiet operational mode the new Freezer Xtreme could only cope with a CPU overclocked to 3.75GHz at 1.4V Vcore, which is a fairly modest result against the background of other today’s testing participants. So, we decided not to limit the potential of the Freezer Xtreme cooler by an improperly working fan and tested it with a Scythe Slip Stream 120 fan working at the same speeds as on Thermalright IFX-14. The fan was fastened between the heatsink arrays using two flexible rubber bands.
So, let’s check out the obtained results:
The results of our today’s test session turned out very interesting. ASUS Lion Square with automatically adjusted fan rotation speed turned out as efficient as Arctic Cooling Freezer Xtreme with an alternative fan in quiet mode at ~910 RPM. Overall, the “lion” doesn’t impress us with its efficiency, although it keeps a quad-core processor stable at 3.85GHz even under severe Linpack conditions. As we have expected, the winner in quiet mode is Thermalright IFX-14, outperforming its rivals by 5-6°C inside the system case and by 7°C in an open testbed. Note that the efficiency difference between the coolers of the same type inside the system case and in an open testbed remained the same (taking into account overall lower temperatures in the open testbed).
However, as soon as you double the rotation speed of the Arctic Cooling Freezer Xtreme fan, the cooling efficiency with an overclocked processor improves by 10-11°C and it even outperforms Thermalright IFX-14 by 1°C inside the system case and gets dangerously close to it in an open testbed! Freezer Xtreme has a very dense heatsink (only ~1.8 mm between the heatsink plates) that is why we did expect its efficiency to improve with the increase of the fan rotation speed. Note that both leading cooling solutions use the same fan, so in fact it is the competition of two heatsinks with similar design. Freezer Xtreme features 6-mm heatpipes, while IFX-14 – 8-mm ones. Besides, the latter is almost twice as expensive as the Arctic Cooling solution and is larger in size. Well, I have to admit that after a simple modification the new Arctic Cooling Freezer Xtreme managed to really surprise us. I only wish I could fit a 32/38 mm fan between its heatsink arrays…
I am not going to offer you the results of our acoustic measurements this time, because everything here is fairly simple. In quiet mode you cannot hear the 92-mm fan of the ASUS Lion Square at its modest ~800 RPM. And under workload when it speeds up to ~2280 RPM, you can hear it, but it is not annoying in any way (36.2 dBA at 1m). As for the acoustics of the Scythe Slip Stream fan, we have already discussed it in our 120-mm fans roundup. As I have already said before, you can’t hear the original Arctic Cooling Freezer Xtreme fan in its quiet mode at ~500 RPM. Unfortunately, it didn’t work and at maximum ~1500 RPM.
I don’t think that the conclusions I am going to draw today will take anyone by surprise. ASUS Lion Square turned out just a good cooler. I would suggest that in the next cooler modification the manufacturer replaces the 92-mm fan with a 120-mm one with nine aggressive blades, increases the supported rotation speed range and allows manual adjustment. They will have to make the hole inside the heatsink larger, they may even have to make the cooler larger as well, but only in this case the “baby lion” could grow up into a real beast.
Arctic Cooling Freezer Xtreme seems to be one of the most interesting solutions we have reviewed lately. A relatively compact and light-weight cooler boasts truly remarkable potential. If overclockers manage to uncover it to the full extent, they will be more than happy with the results. You will need to replace the default fan with the one featuring higher pressure and higher rotation speeds, however, in this case you will get the efficiency of Thermalright IFX-14 for about ~$45 (assuming, you can tolerate certain level of noise). It is a pity that the manufacturer didn’t improve their cooling solution on their own (a 38-mm fan with 700~2500 RPM could be a nice addition). And as for the problem with the fan rotation speed adjustment that we experienced during our tests, I believe it was a flaw of our particular cooler sample.