Cooling Efficiency Tests
The following diagram shows the results obtained during the tests on AMD platform:
First of all let’s analyze the results obtained with the default cooling fans. ThermoLab bada2010 turned out to be the least effective one, which was quite an expected result, considering the cooler size and positioning. However, the mere fact that this little guy managed to cope with an overclocked six-core AMD CPU deserves our sincere respect. However, the next cooler didn’t prove up to our expectations. The all-copper Deep Cool Killer Whale outperformed bada2010 by only 2°C in quiet mode, and by 6°C in PWM mode when the fan sped up to its maximum. It may seem like a good result, but both its brothers, Ice Warrior and Ice Blade Pro turned out 2°C and 7°C more efficient respectively, although they cost less and generate less noise. We were a little surprised to see that Ice Blade Pro turned out 5°C more effective than Ice Warrior, although the latter is positioned one step higher by the manufacturer. Akasa Venom proved just as effective as Ice Blade Pro on AMD platform, but it has a faster fan. Thermalright Silver Arrow easily takes the lead in quiet mode at 810 RPM as well as at maximum rotation speed of its two TY-140 fans.
The results of our cooling efficiency tests with one and two Thermalright TR-FDB fans are just as interesting, because here we also have ThermoLab Baram2010 (it comes without any default fans at all). It wins in three tests out of four, but its advantage is minimal. It is closely followed by Akasa Venom and Deep Cool Ice Blade Pro, while the Deep Cool Ice Warrior is about 3-6°C behind. I have to say that ThermoLab Baram2010 equipped with two most suitable 120 mm fans at 2010 RPM is only 1°C behind Thermalright Silver Arrow with two 140 mm fans at 1280 RPM. Now let’s see what the results on Intel platform are going to be.
The following diagram and detailed table show the results obtained on an Intel platform:
We can clearly see a much greater difference in the performance of our air-coolers caused by extremely high heat dissipation of our six-core Intel CPU. Moreover, we see that things have changed within our test categories compared with what we have just seen on AMD platform. For example, during the tests with the default fan the all-copper Deep Cool Killer Whale in quite mode is the least efficient product, which yielded even to the modest ThermoLab bada2010. At the same time, when the rotation speed of its default fan is increased to the maximum, Killer Whale manages to reduce the peak CPU temperature by 6°C compared with only 4°C improvement we have just seen on AMD. As a result, Killer Whale outperforms bada2010 and even Akasa Venom at quiet 810 RPM. However, as we see from the obtained results, the latter cooler is extremely sensitive to the rotation speed of its fan, so its efficiency improves dramatically when the fan speeds up to 1100 RPM. At maximum fan rotation speed it becomes best cooler of the pack, yielding only to the Silver Arrow from Thermalright. As for the Deep Cool Ice Warrior and Ice Blade Pro, the second cooler proves to be 5°C more efficient.
During the tests with alternative Thermalright TR-FDB fans instead of the default ones, things are also different compared with AMD platform results. Both Deep Cool products can only compete against one another here, and Ice Blade Pro wins. It demonstrates 3-4°C better results. But the most interesting thing here is the fact that Akasa Venom wins in three test modes out of four. Only at maximum rotation speed of two fans ThermoLab Baram2010 turns out 1°C ahead. In all other test modes Akasa Venom managed to surpass the dangerous competitor by 1-3°C. Keeping in mind that ThermoLab Baram2010 is the best cooler with a single-array heatsink, Akasa Venom’s results look really impressive. Therefore, we decided to check how far we can overclock our processor using this remarkable cooling solution when tested with two Thermlaright TR-FDB fans at the comfortable 1100 RPM. Venom proved capable of cooling our six-core CPU overclocked to 4.46 GHz at 1.46875 V Vcore (in the BIOS). In this case the peak core temperature didn’t exceed 86°C:
It is a great result! However, in this case Baram2010 still turned out 2°C better than the cobra-marked cooler: