by Sergey Lepilov
11/24/2008 | 05:45 PM
Contemporary air-coolers have reached the point in their evolution when the fight is on for every additional °C. In this situation the contribution made by thermal interface shouldn’t be underestimated. I am sure many of you know that using high-quality thermal compound instead of the substances usually bundled with the processor coolers may deliver over 10°C temperature improvement under maximum workload. Such tremendous difference may not be achieved even by new cooler revisions launched instead of the already existing ones, so, disregarding an opportunity like that would be absolutely unforgivable.
Now that we decided to check out a number of new thermal interfaces as well as some of the very well familiar ones, we discovered that last time we discussed this topic was over two years ago. It is unacceptably long for high-tech industry. There appeared about a dozen new thermal interfaces in the market since then, so altogether we are going to talk about 16 thermal compounds differing in efficiency as well as price.
Therefore, today we are going to discuss in detail not only the newest but also the very well familiar thermal interfaces from about a dozen manufacturers from all over the world. So, let’s get started.
In this part of our roundup we are going to talk not only about the new products, but will also recall a few thermal compounds we have already talked about before. The products are discussed in order they were tested in our labs.
Arctic Silver 5 thermal compound from Arctic Silver Company may be regarded as a veteran of thermal interfaces. It appeared in the market over 5 years ago and is still one of the best products out there. It is manufactured in the UZS and comes in 3.5g or 12g (in our case) syringes:
Dark-gray relatively thick substance consists of pure silver, boron nitride, zinc and aluminum oxides and also ester. The manufacturer claims that thermal conductivity of Arctic Silver 5 is 8.7 W/(m·K) operational temperatures lie between -50°C and 130°C (up to 180°C under short-term load). This thermal compound reaches its peak thermal conductivity after about 72 hours of operation (the recommended run-in period is from 50 to 200 hours).
The compound is very easy to apply and distribute over the processor heat-spreader, even despite its thick consistency. The imprint on the cooler base turned out very even (hereinafter the imprint on the processor heat-spreader is the mirror-image of the imprint on the cooler base):
Arctic Silver 5 thermal compound is not cheap: a 3.5g syringe costs about $10 and a 12g syringe is priced up to $25. You will definitely get a better price per gram if you go with a 12g syringe, provided you do in fact need that much interface.
We couldn’t find any information about new Pro-Thermal 81 thermal compound. There is nothing about it online, and the sticker on a small syringe says that it is made in Romania and weighs 1.5g:
This thermal compound is of light-gray color and relatively liquid consistency that is why it can be easily applied and removed:
We do not know the exact price of Pro-Thermal 81 thermal compound, but we believe it is around $1~$2.
This product from Innovation Cooling Company comes in a simple plastic bag that contains a 1.5g syringe. The name of this thermal interface – “Diamond 7 Carat” – suggests that there is some connection to diamonds. True, according to the manufacturer, it contains ground synthetic diamonds with up 40 µm particle size.
The manufacturer claims that the thermal conductivity of its product is 4.5 W/(m·K). However, synthetic diamonds are known to have 2000-2500 W/(m·K) thermal conductivity, which is 230 times higher than that of Arctic Silver 5, for instance. Nevertheless, you should keep in mind that we are talking not about pure synthetic diamonds, but about diamond dust with some additional filler that most likely has considerably lower heat conductivity. Diamond 7 Carat doesn’t leak, doesn’t conduct electricity and doesn’t change its features with the time. It requires only a couple hours run-in time to reach maximum efficiency.
This thermal compound is very thick that is why it why it contains special solvent that liquefies the compound in the first few minutes after application. Once applied, you should wait for 10 minutes before installing the cooler, so that the solvent could evaporate and the synthetic diamonds could crystallize. After that the thermal compound layer becomes glossy-gray. When we removed the cooler from the CPU, we saw that Diamond 7 Carat became very hard, but spread evenly over both surfaces:
This is a pretty interesting compound, I should say. However, only practical tests will show how efficient it actually is. Here I would only like to add that it is priced at $5 per syringe, which is quite acceptable compared with the other solutions discussed in our roundup.
Now we would like to introduce to you Cool Silver compound from a not very well-known company called AI Technology Inc. This thermal compound comes in a small package with a cardboard insert and a plastic cover that contains a syringe with the solution:
The insert tells you that this thermal compound is designed to help fight high heat dissipation of overclocked processors, has a picture showing how to apply AiT Cool Silver to the CPU heat-spreader. It also has a temperature monitoring chart showing that AiT Cool Silver is 2-3°C more efficient than “other leading thermal interfaces” (they did not specify which ones, though).
According to the manufacturer, AiT Cool Silver contains 90% of silver particles and has no silicone fillers. They didn’t disclose the heat conductivity of their interface but promise 0.0045°C-cm2/W thermal resistance. This thermal interface comes in a small 3.5g syringe:
This silverfish substance is pretty hard to apply and spread over the processor heat-spreader, because it just wouldn’t smear evenly and doesn’t stick to the surface. Although it did in fact spread quite OK when we installed the cooler:
AiT Cool Silver doesn’t conduct electricity, doesn’t leak or separate with the time. Unfortunately, there is no mention of its MTBF. One syringe costs $8.99, which is quite expensive for the efficiency it offers… Anyway, we are going to discuss the test results later and now let’s move on to the next testing participant.
Another thermal interface from Arctic Cooling Company is based not on silver, but on aluminum oxide. It also contains polysynthetic tars, but no silicone. Arctic Alumina doesn’t leak or dry and maintains its quality over a long period of time. The manufacturer of Arctic Alumina claims 4.0 W/(m·K) thermal conductivity (half of what Arctic Silver 5 has), 0.010 °C-cm2/W thermal resistance and -40°C to +180°C operational temperature range.
This thermal compound retails in 1.75g, 3.5g and 14g syringes. We received the largest for our test session:
Arctic Alumina is white. It is not very thick, but pretty viscous, although you can still easily apply and spread it over the processor heat-spreader surface:
The run in time required for this thermal interface depends on the operational temperature and number of heating-cooling cycles and lies between 50 and 200 hours. Arctic Alumina sells for $5 per 3.5g syringe.
OCZ Ultra 5+ thermal compound from OCZ Technology has been discontinued from production and replaced with a newer solution that we are going to discuss next. Nevertheless, it will be pretty interesting to see how much progress they have made and what the differences are between the old and the new thermal compounds from the same maker.
OCZ Ultra 5+ ships in a small 3g syringe:
This thermal interface contains silver particles (99.9% pure), zinc oxide, aluminum oxide and some boron nitride particles. Its claimed thermal conductivity is 350 W/(m2·C) and thermal resistance – 0.0045 °C-cm2/W. We don’t have any data on the operational temperature range and recommended stabilization period.
There are applications instructions that can be downloaded from the official company web-site, however, even without them OCZ Ultra 5+ is very easy to apply and spread over the surface. It is of dark gray color and very thick and viscous consistency (similar to Arctic Silver 5):
This thermal interface could also be easily removed from the processor surface. We don’t know its price, and even though you can still buy it, it is not very important any more: OCZ Ultra 5+ has already been officially discontinued and replaced with a new product, which we are going to discuss next.
New OCZ Freeze Extreme thermal interface comes in a small plastic packaging with a cardboard insert inside:
The packaging promises you new non-toxic formula, high thermal conductivity and low thermal resistance. OCZ Freeze Extreme contains no silver particles, and their chemical content is obviously not disclosed. Nevertheless, the manufacturer claims that this new thermal compound is 10% more efficient than the predecessor – OCZ Ultra 5+. According to the specifications, OCZ Freeze Extreme boasts 0.0045°C-cm2/W thermal resistance and 3.8 W/(m·C) thermal conductivity.
A small syringe contains 3g of the compound:
This thermal compound is of light gray color and comparatively thin consistency. However, it is extremely adhesive, so you can easily apply an even and thin layer over the CPU heat-spreader. Here is what thermal interface imprint on the cooler base looks like:
By the way, the manufacturer recommends in the user’s instructions to this thermal compound to apply a drop to the center of the processor heat-spreader instead of spreading it evenly over the surface. Then put the cooler on and rotate it slowly in different directions applying a little pressure to it at the same time. However, you have to make sure that both surfaces are ideally even, otherwise, this trick will not work.
OCZ Freeze Extreme should remain functional during a period of two years. The range of operational temperatures is not revealed. The thermal interface is priced at about $8.
Now let’s take a closer look at thermal compound from the Korean Zalman Company. We have already checked out Zalman ZM-STG1 thermal compound in one of our previous cooling solution reviews. This thermal interface is included with some Zalman coolers, such as Zalman CNPS9700 NT/LED and is also available separately.
A small 2.5 ml bottle contains about 6.5g of thermal interface. There is a small brush built into its cap that helps apply this compound easily onto any surface:
The manufacturer doesn’t disclose mush on Zalman ZM-STG1 specifications. It has relatively low heat conductivity of 1.2 W/(m·K) and remains fully functional at up to +125°C.
This thermal compound is very liquid, I would even say watery. Having gathered the thickest part of it onto the applicator-brush and smeared it over the CPU heat-spreader, I got the following imprint on the cooler base:
The manufacturer doesn’t say how long Zalman ZM-STG1 will last. The compound is priced at $12, which is pretty expensive. However, we know that Zalman solutions have always been pricy: brand costs money.
ZEROtherm ZT-100 thermal compound from APACK ZEROtherm Company, whose coolers we have already reviewed before, is available as part of the top cooling solutions bundle as well as an individual product. The syringe with the compound lies on a small cardboard base covered with clear plastic top:
This thermal interface is made in the USA. The syringe contains 3.5g of compound. ZEROtherm ZT-100 has 3.1 W/(m·K) heat conductivity and 0.014 °C-cm2/W thermal resistance. Its operational temperatures range from -40°C to +150°C. It doesn’t leak or dry out and remains fully functional over the entire lifetime.
ZEROtherm ZT-100 comes bundled with a rubber fingerstall:
You can use it to evenly distribute thermal compound over the desired surface, like the instructions show:
Thermal interface is of dark gray color and pretty thick consistency, but is very easy to apply:
The manufacturer claims that ZEROtherm ZT-100 is 1°C more efficient than Arctic Silver 5. We couldn’t find out the recommended pricing, but it is currently selling for about $7.
Thermal compound from the Austrian Noctua released in the end of last year can be considered a new solution in the thermal interfaces market. NT-H1 was developed by the specialists from the Austrian Institute for Heat-Transmission and Fan Technology. It is not only bundled with new Noctua coolers, but also sells separately.
Noctua NT-H1 has the biggest packaging of all our today’s testing participants. It is a plastic casing with a cardboard insert and a special section for the syringe:
The insert contains some info on the peculiarities and features of this thermal interface, such as guaranteed functional stability over the entire usage time of 3 years. Noctua NT-H1 can be stored for 2 years. Also, it doesn’t conduct electricity, doesn’t cause corrosion, is highly plastic and boasts low thermal resistance.
This gray thermal interface comes in a 1.4g syringe that should be enough for 15 cooler installations, according to the manufacturer:
Thermal compound is really very thick, but despite that, very plastic. It can be easily applied and spread over the CPU/GPU heat-spreader:
Noctua NT-H1 can be removed as easily as applied. The lowest operational temperature for this solution is -40°C, however, it can go down to -50°C for a short period of time. The maximum operational temperature is also no record-breaker: +90°C, with up to +110°C for short-term operation.
This syringe is relatively expensive and is priced at $9. As we have already said, Noctua coolers come bundled with this thermal compound, so our today’s tests will show how good this bonus actually is.
Turns out Tuniq Company makes not only Tuniq Tower 120 coolers, but also Tuniq TX-2 thermal compound. Despite a similar name to that of Arctic Cooling MX-2, these are two completely different thermal compounds. The manufacturer makes it clear on the official web-site. Syringe with Tuniq thermal compound is sealed in a plastic bag with some of its technical specifications mentioned on the upper part of it:
Only the manufacturer knows what’s in this thermal interface, nevertheless, like other today’s testing participants it doesn’t leak, doesn’t separate, doesn’t conduct electricity and is stable over the entire lifetime. Tuniq TX-2 thermal conductivity is claimed at 4.5 W/(m·K). The syringe comes filled with 3.5g of solution.
Tuniq thermal compound is of light gray color, thick and viscous consistency, without any alien specks. It is very easy to apply and distribute over the surface.
Tuniq TX-2 functions between -45°C and +200°C temperatures. According to the official info, this thermal compound should be 2~3°C more efficient than Arctic Silver 5 (note that a lot of manufacturers use Arctic Silver 5 as a reference). One syringe of Tuniq TX-2 is priced at $9.
The next solution to be discussed is one of the leaders in the thermal interfaces market – Arctic Cooling MX-2 thermal compound from the Swiss Arctic Cooling Company. The product is packed in an attractive clear plastic package with a black insert:
The front bears product name in large print and its major features. The back – the same, but in greater detail together with brief technical specifications and a chart comparing Arctic Cooling MX-2 against the competitors. It shows that MX-2 is 2°C more efficient than Arctic Silver 5. This thermal interface doesn’t have any specific distinguishing features to boast. Almost like all the other testing participants of our today’s session, it doesn’t leak, doesn’t cause corrosion, doesn’t conduct electricity, boasts high thermal conductivity and low thermal resistance. It is extremely stable.
The syringe contains 4g of thermal compound:
It is of light gray color, and its consistency is almost identical to that of Tuniq TX-2, so it is also very easy to apply and evenly distribute over the surface:
Unfortunately, the manufacturer doesn’t mention the operational temperature range for this solution, but I assume it is similar to that of Tuniq TX-2. One 4g syringe retailing for ~$9 should be enough for 15 cooler installations. CPU and cooling solution testers may be more comfortable with a 30g syringe selling for ~$40.
Coollaboratory Liquid MetalPad is still a newcomer in this market, compared with other solutions discussed today as well as with its own counterpart – Coollaboratory Liquid Pro. It is shipped in a clear plastic packaging:
Besides the actual thermal interface and user’s manual, it is bundled with a damp napkin and a piece of rough material for cleaning away the compound after use:
Liquid MetalPad kit includes six squares of the liquid metal in solid state, i.e. as very thin foil. Three squares measure 38 x 38 mm and are intended for CPUs, while the other three are a little smaller, 20 x 20 mm, and are intended for graphics processors:
This is what these MetalPads look like:
They are very easy to use: just place the foil sheet on the processor or GPU heat-spreader, carefully put the cooler on top making sure it doesn’t shift sideways and then heat up the processor to 60°C in any appropriate application. It would be best to run several heating/cooling cycles to make sure that MetalPad burns in fine. Nevertheless, I failed to get MetalPad on the CPU heat-spreader to burn in completely, even despite long-term testing. However, on a graphics processor running at considerably lower temperatures it did in fact stick very securely to the cooler base:
So, the test results were different in both cases, but we are going to talk about it later today. After I already finished testing, I managed to find some advice about MetalPad application in the forums: it’s better to warm it up with a blow-drier before installing the cooler, and if the base surface is not very even, you may as well use two or three MetalPad foils. I doubt that it is the best way-out, but it is available to you.
MetalPad films cannot be used over and over again, and they are in fact pretty expensive: one package with 6 MetalPad sheets costs $15.
Now we would like to reintroduce to you another thermal interface from the Coollaboratory Company that we have already mentioned in our articles before. Now, however, it is not in the OEM, but in a retail package:
The bundle includes rough porous pad, damp napkin and two Q-tips for removing the liquid metal from surfaces:
A small syringe (0.3 CC) contains 1ml of liquid metal. It contains gallium, indium and other non-toxic non-ferrous compounds:
Although this is a fairly small syringe, it should last you for at least 5 cooler installations provided you apply the compound the right way. The application procedure is very simple, just squeeze a small drop into the center of degreased processor heat-spreader and spread it evenly with a cotton ball or soft cloth over the entire surface. Besides, I would also recommend doing the same thing to the cooler base:
When both surfaces have been prepped like that and run through several heating/cooling cycles, they get a better “grip” than in case thermal interface has been only applied to one of them. Now that they included rough porous pad for compound removal (I don’t know its exact name), it has become a very easy to remove Liquid Pro from the surfaces. It will take you less than a minute to clean it off, although you will have to forget about nice polishing. Remember, Liquid Pro shouldn’t be used with aluminum surfaces.
The manufacturer promises thermal conductivity of 82 W/(m·K). Liquid pro should remain fully functional at operational temperatures between -273°C and +1200°C. If it is true, it is truly impressive, but I doubt that anyone needs such a wide temperature range. This thermal interface contains no mercury or any other toxic substances. One syringe costs about $12.
The Hong-Kong Gelid Solutions Ltd. is a new company in this market. Created in 2008 by former Arctic Cooling employees, it is based out of Switzerland. The name “Gelid” comes from the Latin word “geldius” that means “very cold, icy”. In the future, gelid is going to manufacture cooling systems and fans; the fans have actually already appeared in the market. But today we would like to talk about Gelid GC1 thermal interface.
Gelid thermal compound comes in a small plastic package with a cardboard insert inside. The front of the packaging mentions the solution’s key features, displays three award logos, and promises 25% of the compound for free:
You can check out the application fields for Gelid GC1, its brief technical specifications and comparison chart showing that this thermal compound should be 2°C more efficient than Arctic Silver 5 on the reverse of the package.
Together with a small 1g syringe you get a small spatula that should help evenly spread the compound over the desired surface:
Thermal compound is gray, has very thick and viscous consistency, but spreads easily in a perfectly thin layer over the surface:
According to the manufacturer, Gelid GC1 doesn’t conduct electricity, doesn’t leak or cause corrosion and boasts low thermal resistance. Unfortunately, they didn’t mention the operational temperature range for it, but it should last you 10 years without losing any of its features. Its recommended retail price is $6.99.
We tested all thermal interfaces discussed in this article inside a system case with the side panel removed. Our testbed configuration remained unchanged for the entire test session and included the following components:
All tests were performed under Windows Vista Ultimate Edition x86 SP1. SpeedFan 4.36 beta 15 was used to monitor the temperature of the CPU, reading it directly from the CPU core sensor:
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 Linpack 32-bit with very convenient LinX shell version 0.4.9 to heat it up to its maximum. We manually set the RAM capacity at 1850MB and recorded 15 runs.
Besides, since Linpack heats the CPU significantly, we also performed an additional test by running OCCT v2.0.0a CPU test for 23 minutes (with maximum priority):
The testing procedure on CPUs looked as follows. I applied the tested thermal interface, installed the cooler and started the system. I let it run for ~20 minutes for all system components to reach their nominal temperatures and stabilize. As a rule, this period of time there were some office applications running. Then I launched OCCT processor test. Once the test was completed, I let the system stabilize for another ~10 minutes and repeated OCCT once again. The next stabilization period was 20 minutes. After that I launched LinX test. When the test run was completed, a 10-minute stabilization period followed, and then I repeated LinX once again. In both cases, we recorded the results of the second test cycle, because the temperature readings in this case were often about 0.5~1°C higher than in the first test cycle.
However, this wasn’t the end of our test session on a CPU. After a complete cycle of tests with two applications we dismounted the cooler, removed the remaining thermal interface from the processor heat-spreader and the cooler base and degreased them with alcohol. Then we applied a new layer of thermal interface, installed the cooler and ran all the tests over again. This way, we tested each thermal interface during two 3-hour test cycles. The only exception was Arctic Silver 5 which was the first to be tested and which has already been in use for four days. The final result that you see on the diagrams is the average temperature of all four processor cores. We took the best result of the two test cycles for each thermal compound. To be fair I have to say that the results during first and second application of the thermal compound differed by 0.5~1°C at the most.
During our test session we paid special attention to ambient temperature, because it was one of the key factors to influence the final result. It 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 varied between 24.5~25.0°C. So, the temperature deviations didn’t exceed 0.5°C. Taking into account the above described testing methodology I allow a measuring error of no more than 0.5°C.
In order to increase the dependence of the CPU cooling efficiency on the thermal interface used, we employed a highly efficient ZEROtherm ZEN FZ120 cooler. We replaced its default fan with two 9-blade Scythe Minebea Silent IC fans at ~1130 RPM each. They were attached to the heatsink for air intake/exhaust:
We decided on a ZEROtherm cooler, also because its base is not nickel-plated, like on some other cooling systems. Besides, its surface is well-finished, but not polished:
The cooler base is almost impeccably even. And the processor heat-spreader has long been perfected to the maximum (I evened its surface with 1000x sanding paper). In other words, both contact surfaces are made of copper and have been very well finished. The surfaces haven’t been polished. The cooler was always installed the same way and was fastened with retention screws evenly tightened in diagonal pairs.
Besides the tests on a CPU, we also checked the efficiency of our thermal interfaces on RV770 GPU of Radeon HD 4870 graphics card. This GPU die size is 256sq.mm. Although the graphics processor and the graphics card itself boast pretty high heat dissipation already, we increased the frequency to 830MHz (300MHz in 2D mode). According to the results of our latest VGA cooler tests, reference cooler of ATI Radeon HD 4870 is pretty efficient, however, only if you are eager to put up with relatively high noise. Since I wasn’t ready for sacrifices like that, I replaced the default reference cooler with Arctic Cooling Accelero Twin Turbo with two fans working at their maximum speed of ~2090 RPM:
I still used the lower heatsink plate from the Radeon HD 4870 reference cooling system, so it helped with cooling the PCB voltage regulator components and ensured higher stability during overclocking. We used the backplate from the Radeon HD 4870 reference cooler, the pressure was applied not to the GPU corners but to the entire die.
Let’s take a look at the cooler base and GPU top surface:
There is nothing I could add about the GPU (this is the only way it can be without the heat-spreader). As for the base of Accelero Twin Turbo cooler, it is even, has average finish quality. I didn’t improve it in any way.
Radeon HD 4870 was warmed up by two 12-run cycles of Firefly Forest benchmark from the synthetic 3DMark 2006 suite in 1920x1200 resolution with activated x16 anisotropic filtering but without FSAA. Moreover, to create additional GPU workload we used FurMark version 1.4.0 in stability test mode in 1024x768 resolution (window mode). We ran the tests twice for each thermal compound with 10~12 minutes stabilization period between the test cycles. We installed the cooler twice for each thermal compound, the same way we did during our CPU tests – after removing the remaining thermal interface and degreasing the surfaces with alcohol. We monitored the graphics card temperatures using RivaTuner v2.11 utility (created by Aleksey Nikolaichuk aka Unwinder). We took the average reading from four GPU diodes: "Core t", "Core t, display IO", "Core t, memory IO" and "Core t, shader core".
Besides 15 thermal interfaces discussed in our today’s roundup we also added the results of the 16th participant: SilMORE thermal compound:
We decided to include this common thermal compound into our today’s test session because most manufacturers use SilMORE thermal interface to bundle with their cooling solutions. For example, Scythe, Thermaltake, Xigmatek, Ice Hammer, OCZ and several other less popular makers include a 1g pack of SilMORE thermal compound with their coolers. So, the results will help us determine how many extra degrees we can win by replacing the standard SilMORE thermal interface with one of the reviewed today.
To increase the heat dissipation our 45 nm quad-core processor was overclocked to 3.75 GHz (+25.0%). The nominal processor Vcore was increased to ~1.5 V in the mainboard BIOS (+20.0%):
The cooler we used this time together with the best thermal compounds can cool the CPU overclocked to 3.85GHz (the workload created by Linpack). However, I lowered the CPU frequency on purpose, to ensure that the least efficient thermal compounds will be able to pass. And turned out, I was right to do it this way.
Here is the table with the detailed results and the diagram with the average temperature readings (thermal interfaces have been sorted according to their efficiency and temperature increase):
Click to enlarge
The results of our test session on a CPU show that there is a group of 8 highly efficient solutions starting from the leaders – IC Diamond 7 Carat and Coollaboratory Liquid Pro – and finishing with Noctua NT-H1. Arctic Cooling MX-2 and Gelid GC1 fell just a little bit behind them, and liquid Zalman ZM-STG1 loses another 0.5°C to them. However, the temperature difference between IC Diamond 7 Carat and even Zalman ZM-STG1 is 2.5°C at the most, so we can recommend any of these 11 thermal interfaces for use with CPU coolers.
And further on the efficiency differences are much more dramatic and reach up to 10°C. SilMORE thermal compound bundled with many cooling solutions is the worst of the 16 products tested today, so we would strongly advise you to replace it immediately. Silver-colored AiT Cool Silver and Romanian Pro-Thermal 81 also cannot boast much, and Arctic Alumina becomes the best among the worst. I would like to particularly dwell on Coollaboratory Liquid MetalPad: its unimpressive performance in our tests must be connected with not ideally even contact surfaces, so the micro-grooves didn’t get filled with the compound even despite relatively long testing period and high operational temperatures. However, the photos above as well as the results of our GPU tests prove this assumption very clearly.
As I have already mentioned above, thermal interfaces were tested on the graphics card processor overclocked to 830 MHz:
The frequency lowered to 300 MHz and the voltage – to 1.1V in 2D mode. The table with detailed results and the diagram with average values follow:
Click to enlarge
It is interesting that the tests on a graphics processor changed a few things within the groups, but overall the picture remained the same. However, I have to draw your attention to a serious breakthrough demonstrated by Coollaboratory Liquid MetalPad that yielded only to liquid metal. It is certainly the result of impeccably even contact surfaces. The leaders are followed by a group of 9 highly efficient thermal compounds starting with OCZ Freezer Extreme and finishing with OCZ Ultra 5+ (by the way, the new thermal interface from OCZ is really better than the predecessor in this test session). And the last ones are still the solutions with liquid consistency including Arctic Cooling, SilMORE and Zalman ZM-STG1.
I don’t think it makes sense to hunt down every half a degree in the conclusion to our today’s roundup. Especially, since a small difference like that lies within the accepted measuring error. Therefore, I would like to say that out of all thermal interfaces we tested today we wouldn’t recommend such solutions as AiT Cool Silver, Pro-Thermal 81 and Zalman ZM-STG1. SilMORE thermal compound is the worst solution of all we checked out this time and we strongly advise to replace it immediately. And those cooling solution manufacturers who bundle their products with a pack of SilMORE compound should definitely reconsider if they want to score high. Arctic Alumina turned out the best of the worst thermal interfaces, however, it delivers high thermal conductivity at extremely low temperatures, which might be handy for extreme overclocking fans.
As for the winners, I can certainly single out liquid metal from Coollaboratory and Coollaboratory Liquid MetalPad with a few remarks regarding its proper use that should be taken into account. The new OCZ Freezer Extreme performed extremely well as a highly efficient, relatively inexpensive and very easy to work with. The good old Arctic Silver 5 still holds on to its leading position successfully, although now it has at least eight worthy competitors to worry about. Overall, it is very pleasing to see that there are more than 2-3 worthy thermal interfaces around. And of course, do not forget that thermal interface is simply a substance to fill into the micro-pores between the cooler base and the processor heat-spreader. You have to make sure that both these surfaces are very even, because no “millimeter” layer of thermal compound usually applied by newbies and several “experienced” testers out there will guarantee efficient cooling and objective cooler comparison.