Not everyone wants a giant tower CPU cooler. Some people want dual fans and monstrous fin real estate, but some prefer something a bit squatter.
My latest CPU cooler reviews is the Cooler Master GeminII M4, a new low profile cooler. It’s a squat, copper heatpipe sporting, many aluminium finned heatsink fan combo that thanks to its unique shape could cause clearance issues; it’s going to be an interesting review.
- CPU Socket: Intel® LGA1366/1156/1155/775, AMD FM1/AM3+/AM3/AM2+/AM2
- Dimension: (W / H / D) 137 x 122 x 59mm (5.4 x 4.8 x 2.3 inch)
- Weight: 289g (0.64 lb)
- Heat Sink Dimensions: 120x 118 x 44 mm (4.7 x 4.6 x 1.7 inch)
- Fan Dimension: (W / H / D) 120 x 120 x 15mm (4.7 x 4.7 x 0.6 inch)
- Fan Speed: 500 – 1600 RPM (PWM) ± 10%
- Fan Airflow: 17.4 – 58.4 CFM ± 10%
- Fan Air Pressure: 0.18 – 1.76mm H2O ± 10%
- Fan Life Expectancy: 40,000 hrs
- Bearing Type: Long Life Sleeve Bearing
- Fan Noise Level (dB-A): 8 – 30 dBA
- Connector: 4-Pin
The box of this new cooler is typical of Cooler Master’s new box designs, very professional and clean while still clearly displaying the information you need to decide if this cooler is correct for you.
The back has information in multiple languages and some technical drawings of the cooler.
The bundled items include: a manual, AMD and Intel mounting parts, thermal compound and nuts, screws and a nut setter.
The provided mounting system is another variant of Cooler Master’s mounting solutions. I definitely like this version the most. For my AMD setup, the four included rubber rings needed to be stuck to the motherboard to ensure there is no metal contact with the motherboard. The system was easy to install and very firm once fixed.
The thermal paste is the same standard thermal paste Cooler Master provides with all of their coolers. The paste is of a high quality and has a good viscosity for a thermal compound, making sure it spreads well when you mount the cooler.
The form factor is slightly different to the ‘tower type’ coolers that we are used to. As a low profile cooler, the heatsink is spread out horizontally, often known as a ‘top down design’. The fan is also thinner than standard 25mm ones a, coming in at 15mm.
The 6mm heatpipes are laid out sensibly through the heatsink.
With a slightly different layout on the other side.
The base of the M4 is a heatpipe direct touch style base, which comes with some benefits – but a few downsides too. For example, having the heatpipe directly touching the CPU increases the efficiency of individual heatpipes. Instead of the heat having to travel through a base, then to the solder and then to the heatpipe. The downside is that because the actual CPU die is quite a bit smaller than the IHS (Integrated heat spreader, the top part you see of the CPU), the heat mostly comes from the middle of the IHS. This means the heatpipes nearest the core do most of the work, leaving the outside ones with less to do, decreasing overall performance. This is where a soldered on base helps to spread the work between heatpipes.
The core was flat and well machined.
This cooler is really thin, making it very useful for high end HTPC users.
Testing and Overclocking
CPU: AMD Phenom II 1090T
GPU: ASUS Reference 5870
Memory: 4GB G Skill Ripjaw 1600MHz 7-7-7-24 DDR3
Storage: Crucial C300 128GB
PSU: Cooler Master Silent Pro 850W
OS: Windows 7 Ultimate x64
Case: Corsair 700D equipped with many fans
To ensure the results were consistent, the heatsink was mounted three times and idle and load tests run each time. An average of the results is shown on the results page.
For idle testing, Windows 7 was allowed to run with no programs for five minutes.
For load testing I used Prime 95 Small FFTs and temperatures were taken after 20 minutes.
The standard thermal paste, provided with the cooler, was used.
The settings we used for our stock and overclocked tests are:
Stock – CPU 3200MHz Auto voltage (1.325v)
Overclocked – CPU 4000MHz 1.456v
NB. Due to the thermal probe built into the Phenom II CPUs being incorrectly calibrated, they temperatures displayed are 5C cooler than the actual temperature of the CPU. The temperatures are all taken from the same CPU so are all on the same scale.
[easychart type=”horizbar” title=”Temperatures – Stock” groupnames=”CM Gemini II M4, CM Gemini II SF542, CM Gemini II SF542 (Silent), CM Hyper 412S, CM Hyper 412S (Silent), Prolimatech Panther, Prolimatech Lynx, Antec 920 Silent, Antec 920 Extreme” valuenames=”Idle, Load” group1values=”23,45″ group2values=”22,37″ group3values=”22,45″ group4values=”14,31″ group5values=”16,35″ group6values=”20,32″ group7values=”20,38″ group8values=”21,38″ group9values=”15,26″ ]
The Cooler Master GeminII M4 did well in the stock frequency test. The fan did reach its maximum speed, which created some noise but only what you would expect from a normal cooler. Audible, but not loud.
[easychart type=”horizbar” title=”Temperatures – Overclocked” groupnames=”CM Gemini II M4, CM Gemini II SF542, CM Gemini II SF542(Vortex Fan),CM Gemini II SF542 (Silent), CM Hyper 412S, CM Hyper 412S (Silent),CM Hyper 412S (Two Fans) , Prolimatech Panther, Prolimatech Lynx, Antec 920 Silent, Antec 920 Extreme” valuenames=”Idle, Load” group1values=”28,76″ group2values=”23,67″ group3values=”21,64″ group4values=”26,74″ group5values=”19,49″ group6values=”21,58″ group7values=”18,45″ group8values=”24,48″ group9values=”25,55″ group10values=”21,46″ ]
When it came to the overclocked test, things went slightly downhill. No problem with idle but after a while of sitting at load, the CPU hit 77C which happens to be the thermal shut down point of my CPU. There is one main reason for this, the cooler simply isn’t designed to cool an overclocked Hexa core CPU. Something like an energy efficient dual or even quad core would be much more suited to this cooler. Using a lower wattage CPU also means that the fan would spin slower, making the cooler close to silent.
So the testing results were good while the CPU was set at stock speeds but went downhill when the overclocked load test was performed. These results were mostly expected. Mixing small/low profile with very quiet cooling normally equals lower performance.
The cooler does the job Cooler Master designed it to do which is cool a CPU in a small amount of space, it does that job very well.
I see the cooler retailing for between £25 and £27. That seems like a good price for the quality you get with this cooler. This is a more specialist device, designed for small spaces.
Performed well at stock frequency
Very low profile
Very quiet for most purposes
High quality thermal paste
Overall quality of the product is excellent
Mounting system works very well
Could not provide sufficient cooling when used in the overclocked test
Noise creeps up under load