by Sergey Lepilov
02/08/2006 | 10:48 AM
The processor cooling solutions using heatpipe technology start little by little moving from the high-end price segment to the mainstream category. As for the $30 price range, we could also sometimes come across heatpipe coolers here, although we have to admit that the solutions in this price group have always been considerably less efficient than the more expensive heatpipe products from Zalman, Thermaltake or CoolerMaster, for instance.
Nevertheless, inexpensive cooling solutions kept evolving and today we are going to introduce to you a pair of really attractive products from Arctic Cooling: Freezer 7 Pro and Freezer 64 Pro, that have come to replace their predecessors – Freezer 7 and Freezer 64.
Let’s find out now how much progress they have made today.
We will start checking out these babies with their packaging. Arctic Cooling products are shipped in identically designed carton boxes colored with glossy black:
The only difference between the coolers is the name on the box: one says Freezer 7 Pro for Socket T (LGA775) and the other - Freezer 64 Pro for Socket 754/939/940. In fact, the coolers only distinguishing features besides the name is the retention mechanism, which is actually quite logical. Therefore, we will continue our discussion of the cooler design peculiarities with only one of them: Freezer 7 Pro. The specific design of the retention mechanism of the Freezer 64 Pro will be discussed separately later in this review.
The front side of the box is decorated with a large photo of the cooler and the URL address of the company’s web-site. On the reverse side of the box you can actually see a detailed schematic representation of the air flow inside the system case when the cooler is installed. The detailed description is also available:
By the way, according to this description, the cooler fan grabs air not only from the side where the fan blades are located, but also from the bottom of the cooler and from the top. It means that the cooling fan is designed in such a way that it cools down not only the CPU itself but also the mainboard components and the system memory that appear to be in the air-grabbing zone. We will talk more about the fan design later today, of course. As for the removal of hot air away from the heatsink fins, the schematic provided on the box suggests that the warm air from the heatpipes and heatsink fins is ousted towards the system case cooler, while the warm air from the bottom part of the cooler is flowing towards the mainboard MOSFETs, which should also reduce their temperature a little bit.
But, let’s get back to the box. There is one more side that remains unmentioned. It is the side listing all the technical specifications of the cooler inside:
Besides the technical specifications for the Freezer series of coolers, there are also two performance diagrams. The first one illustrates the noise level of the given cooler model compared against the regular boxed Intel or AMD cooler (the results are all provided in Sone units, typical of Arctic Cooling). The second diagram shows thermal resistance of the cooler also compared against the standard boxed cooing solutions, which are available in several different modifications by both, Intel and AMD. Among them are a few one with heat-pipe technology. However, there is no mention on the box, what model is used as the opponent to Arctic Cooling solution in this comparison.
Inside the box there is a transparent plastic case shaped exactly following the cooler shape:
This packaging helps secure the cooler heatsink and fan and reduce the possible transportation damages to the minimum. As for the accessories, you will only get a cooler, a user’s manual and a few stickers with Arctic Cooling’s brand name.
Freezer cooler family boasts pretty traditional design for heatpipe cooling solutions: copper footing with heatpipes going through it. Both cooler models we are discussing today, Freezer 7 Pro and Freezer 64 Pro have 3 heatpipes each. All of them are made of solid copper. The heatpipes go through a set of thin aluminum plates (42 pieces total), and there is a seven-blade fan directing the airflow through these plates:
In fact, Arctic Cooling didn’t suggest anything revolutionary here, however there are still a few things that are absent even by such big names in the cooling industry as Zalman and Thermaltake.
Firstly, I would like to draw your attention to the shape of the three lower heatsink plates: they are bent downwards:
As I have already mentioned above, this simple but at the same time original solution allows directing the airflow towards the power elements on the mainboard thus reducing their temperature as well.
The second know-how from Arctic Cooling implemented in the Freezer series is the soft outboard retention mechanism for the cooling fan. Namely, the fan is fastened to a plastic frame not with screws or any other firmly holding retention. It is hung onto four rubber rods with fixing locks:
This way the fan and the retention frame have very “soft” contact, which reduces vibrations and noise quite significantly. The plastic frame is clipped to the cooler heatsink with the help of special lugs. The lugs catch to a slit on the aluminum plates. You can see it clearly if you look at the cooler from the side:
From this viewpoint you can also see very well where the heatsink plates meet the heatpipes. The contact between them is made exactly the same way as by Scythe Ninja cooler, which we have already discussed in great detail in our review called Battle of the Titans: Super Coolers from Scythe, Thermaltake and Zalman Face to Face . The contact surface between the plates and the heatpipes is the flat aluminum necks that may have been shaped like that on purpose or may have turned out like that when the preheated aluminum plates were threaded onto the heatpipes:
Note that there is a gap between the aluminum necks only where the three lower plates are bent. All the rest of the plates sit so close to one another that there is no copper heatpipe shining through anywhere.
The ends of the heatpipes at the top of the heatsink have been hermetically soldered, although not in the neatest way:
Moreover, the heatpipes bear some mechanical traces from the threaded plates that could have been left during the heatsink assembly.
The cooler footing is made of two copper plates, each about 5mm thick. The heatpipes go through these plates. The footing plates and the heatpipes are evidently soldered together, since we can clearly see alloy traces on the lower ends of the heatpipes:
Again, I have to admit that this could have been done neater. Although, the finish of the cooler sole leaves even more to be desired. It would be more correct to say that the cooler sole hasn’t been finished at all. The mechanical milling traces can be seen with a naked eye and moreover, can be even felt to the touch. I have to add here that despite the absence of any final polishing on the sole, it is ideally even: the footprint from the cooler sole on the glass that we obtained with a very thin layer of thermal paste appeared outstandingly even, without any empty spots.
The cooler sole is pre-covered with Arctic Cooling MX-1 thermal paste and a protective plastic cap:
The picture above has been taken after we removed the cooler from the system, so do not be concerned with the missing thermal interface in some spots. The Arctic Cooling’s own thermal paste was very thick, I would even say dry, and the layer was quite generous as well. After 4 days of testing this cooler with its default thermal paste we decided to replace it with our KPT-8 paste, and the improvement didn’t keep us waiting for long: the temperature dropped by 3o C in burn mode compared with what we were seeing with the standard MX-1 thermal interface before. Maybe we couldn’t really get better results with Arctic Cooling’s authentic thermal interface because we haven’t used it long enough. According to the manufacturer’s instructions, the interface should be exploited at least for 200 hours (8 days) in order to reveal its full conductive potential. Unfortunately, we didn’t have so much time at our disposal during this test session, so we cannot really make any judgment about the efficiency of the MX-1.
The fans used on Freezer coolers are built with Arctic Ceramic bearings, are identical for both cooler models and differ only by the rotation speed and the power connector. The connector on Freezer 7 Pro is four-pin, so that the mainboard could control its rotation speed (PWM).
As you may have already noticed, the Freezer 7 Pro is installed onto the CPU in a standard way. It is fastened just like any boxed Intel cooler, i.e. with four plastic “pins” that should be pushed all the way into the holes around the processor socket until you hear a clicking sound.
The cooler can be removed very easily, by simply rotating the nail heads counter-clockwise with a flat screwdriver and removing them from the holes. Since the holes are symmetrically placed around the Socket T (LGA775), the cooler can be installed facing either way. The ideal position however is when the airflow from the cooler fan is directed to the system case rear fan or towards the PSU fan taking air in. This is a great advantage of this cooler. I would also like to add that when you install the cooler it is very convenient to remove the fan first, fasten the heatsink on top of the CPU and then return the plastic frame with the fan back in its place.
As for the other cooler model, Freezer 64 Pro, it is installed in a classical way with the help of a traditional metal bracket catching to the plastic frame around the processor socket. After that the cooler is pressed to the CPU with a locking latch:
You may get the impression that the installation of the Freezer 64 pro cooler may depend on the position of the plastic frame around the CPU socket. However, it is not quite like that. Arctic Cooling offers us a lot more flexibility here. The thing is that the cooler retention bracket can be moved from the standard position “along the fan” into the position “across the fan”. It can be easily done by simply removing the locking latch:
You can download a detailed installation guide for Arctic Cooling Freezer 64 Pro from the company web-site (pdf-file, 176KB ).
Arctic Cooling Freezer 64 Pro and Freezer 7 Pro are already selling for about $32.
Well, now that we have introduced to you all the smallest details about new arctic coolers design, let’s sum up the technical specs in a table.
Feature | Freezer 7 Pro | Freezer 64 Pro |
Manufacturer | Arctic Cooling (Switzerland) | |
Dimensions (fan size), mm | 107 x 96.5 x 126.5 | 107 x 96.5 x 126.5 |
Nominal voltage | 12V | |
Nominal current | 0.16A | |
Power consumption | ~1.92W | |
Heatsink material | Aluminum | |
Fan rotation speed | 300 ~ 2500rpm (PWM) | ~2200rpm |
Airflow, CFM | 45 (max) | 40 (max) |
Noise, dBA | 0.9 Sone | 0.8 Sone |
Fan bearings | Arctic Ceramic | |
MTBF | 137 thousand hours (at 40°C) | |
Weight | 520g | 528g |
Supported CPU sockets | Socket T (LGA 775) | Socket 754/939/940 |
Maximum CPU | All Intel Pentium 4 | All AMD Athlon 64 FX |
Additional | Cooler sole covered in MX-1 thermal paste | |
Approximate price | ~$30-$32 | |
Now let’s pass over to our actual test session.
All tests were carried out on two platforms: Intel and AMD. We tested the systems in open stands as well as with closed system cases. We used the following hardware components to assemble our testbeds:
All the tests were conducted in Windows XP Professional Edition Service Pack 2 with NVIDIA nForce system driver version 6.82. I would like to point out specifically, that we used KPT-8 thermal paste for all our tests today. I have already mentioned the results of the efficiency comparison between the Arctic Cooling’s own and our thermal pastes. The automatic fan rotation speed control option was disabled in the BIOS Setup of both mainboards.
AMD Athlon 64 3000+ (1800MHz) was overclocked to 2700MHz at 1.575V Vcore:
The Intel CPU marked as SL8J9 and manufactured in Malaysia was overclocked to 4019MHz without raising the Vcore at all:
We used S&M utility version 1.7.6 (beta) for CPU warm-up, temperature monitoring (with SpeedFan 4.26 utility) and fan rotation speed control. We warmed up the CPU for 15 minutes in the FPU test under 100% workload.
Moreover, keeping in mind that S&M loads the CPU very heavily, which is not typical of most software applications, we performed the CPU warm-up tests with two simultaneously running copies of Super PI benchmark, which is very popular among overclockers today. The calculation of the PI value up to 32 million digits takes about 26-27 minutes on a platform like ours.
The thermal protection of the Intel Pentium 4 Prescott processors was again managed with the help of ThrottleWatch tool version 2.02:
The coolers were tested in absolutely equal conditions in two systems types: in an open stand and in a closed system case equipped with a pair of quiet 120-mm case fans. The winter weather hasn’t spared us these days, so the room temperature during our test session was pretty low: only 19o C. So, we took this temperature value as a reference po0int on our diagrams.
We didn’t have any cooling solution price within the same $30-$32 range at that time, so we decided to compare them against the best representatives of the today’s air cooling systems, such as Thermaltake Big Typhoon ($41) and Zalman CNPS9500 LED ($65). Let’s take a look at the results now.
Before we start talking about the actual numbers obtained in an AMD platform we have to point out that the diagrams were built using the results taken from two thermal diodes: PWM (the thermal diode located next to the processor socket on the mainboard) and CPU (built-in processor thermal diode).
So, first come the results taken from the PWM diode on AMD platform built around ABIT AN8 SLI mainboard:
* - the fan rotation speeds are taken not from the specification list, but are calculated as the average rpm value according to S&M measurements.
On the left-hand side you can see the results from the closed system case, and on the right-hand side – from an open stand. Note that the results for Zalman CNPS9500 LED cooler are given not only for the maximum fan rotation speed of 2940rpm, but also for the medium rotation speed of 2050rpm, which was determined as an audible border between the noisy and quiet operation. As for the minimal rotation speed, we decided not to include the results for this mode.
So, how do the new coolers cope with the overclocked processor? Firstly, I have to stress that our today’s hero from Arctic Cooling coped well with this task. Also, both Arctic Cooling models, Freezer 64 Pro and Freezer 7 Pro generate quite little noise and the bearing sounds quite tolerably, unlike many other bearings that produce very unpleasant noise.
As for the Thermaltake Big Typhoon and Zalman ZNPS9500 LED, which have become very popular among computer users these days, provide about the same thermal performance at a comparable level of generated noise. In an open stand, Zalman cooler appears 6o C more efficient, while Thermaltake cooler improves the thermal conditions by only 3o C. I have to say though that we reinstalled Thermaltake Big Typhoon three times before we managed to obtain this result.
Nevertheless, everything we have just said has been based on the results reported by the mainboard diode. Now let’s take a look at the results obtained from the built-in processor thermal diode:
Except for the results of Zalman CNPS9500 LED, the situation remains the same. Only the gaps between the results grew somewhat smaller this time. As for the results of the Zalman cooler, it is probably the poor contact between the cooler foot and the processor heat-spreader that are to blame for the failure here. Yes, the thermal paste footprint on the cooler foot was very even, but it seems to be not enough this time for efficient heat dissipation. Unfortunately, we couldn’t eliminate the problem neither by reinstalling the cooler, nor by removing the thermal paste completely and putting a new layer on. The remarkable thing though, is that with Zalman CNPS9500 LED cooler the CPU was working stably at the frequency of 2800MHz and 1.675V Vcore, which are the maximum supported values for this type of CPU with the air cooling system. So, it would be really interesting to see what our heroes will be able to do when we install it onto an overclocked Pentium 4 processor working at 4GHz.
The diagram was built only from the results reported by the thermal diode built into the CPU. The results are actually quite interesting. Arctic Cooling Freezer 7 Pro looks very good on an open stand, however, in a closed stand its results appeared less impressive. However, despite this fact, even when our overclocked CPU temperature reached 70o C it wouldn’t throttle and continued running at its full power. Here I would only like to say that the leadership on the Intel platform still rests with Zalman CNPS9500 LED cooling solution.
To tell the truth, we have pretty mixed impressions from these two cooling solutions. Yes, they do cool the CPUs quite efficiently, and you can also do some overclocking with them. However, they still fall a little bit behind the best contemporary cooling solutions out there. On the other hand, however, keeping in mind that these coolers are currently selling for around $30 in most stores, they have all the chances to find their customer. Especially since Arctic Cooling Freezer 7 Pro and Freezer 64 Pro are much smaller in size than the monsters from Thermaltake, for example, and can easily fit into any system case.
Anyway, you always have the last word here, but we hope our article will help you in this uneasy decision making :)
Highs:
Lows: