Introduction

Central Processing Units, or CPUs for short, have not always needed extravagant cooling. There were times when a small heatsink, passively cooling the CPU, would be more than sufficient. Then as time went on, bigger heatsinks were needed and then eventually, fans were necessary to actively cool these ‘monsters’ of performance.

Things and times have changed but the more things change, the more they remain the same. We now have faster, more robust CPUs but we still need proper cooling in order to keep them functioning correctly. The majority of persons will simply use a computer and accept its performance for what it is worth. However, there are others that are in (believe it or not) a vast minority, that will venture into the realm of overclocking.

With the advent of overclocking, the need to squeeze out every last drop of performance comes at a price. That price is the additional heat generated and the need for better cooling. The stock heatsink that comes with a processor is good enough to cool the CPU at stock and maybe also with a mild overclock. However, for most overclockers, mild just is not going to cut it.

We all want the performance but not many are willing to do what it takes to achieve it. That is where aftermarket coolers come into play. There is a plethora of choices and to each his own, but today I will be reviewing two in particular.

Special thanks to Cooler Master for providing retail samples of the Hyper TX3 and the Hyper N620.

Cooler Master Hyper TX3

Design

The Hyper TX3 is an entry-level aftermarket cooler specifically geared towards being better than the stock heatsink. It utilizes a tower design that makes use of direct contact copper heatpipes and aluminum fins.

This has become a more or less standard design nowadays, simply because it is very effective at removing heat from the CPU.

Specifications

Model	RR-910-HTX3-GP
CPU Socket	Intel Socket LGA775/1156, AMD Socket 754/939/940/AM2/AM3
Heat Sink Dimensions	90 x 51 x 139 mm
Heat Sink Material	Aluminum fin + 3 heatpipes
Heat Pipes	3 pcs
Fan Dimension	92 x 92 x 25 mm
Fan Speed	800 – 2800 R.P.M. (PWM)
Fan Airflow	15.7 – 54.8 CFM
Fan Air Pressure	0.35 – 4.27 mm H2O
Fan Life Expectancy	40,000 hours
Bearing Type	Long life sleeve bearing
Voltage Rating	12 V
Noise Level	17 – 35 dBA
Weight	1.04 lbs; 470g

The Hyper TX3 retails for US$20 and is positioned to be a strong choice for the budget-minded buyer.

Installation

The installation process for the Hyper TX3 went well. However, it was not as smooth as I had hoped. In fact, mounting the cooler was the easy part – just as easy as the stock Intel cooler. Just simply ensure the pins are in the correct position and then press until they click firmly into place.

The hard part was actually attaching the fan to the heatsink itself. The included clips for holding the fan in place were a bit of a pain to catch properly. This could have been due to its tension design for keeping the fan in position. With a bit of patience, it all came together quite nicely.

The anti-vibration pads included on the fan was also a very nice touch. Also, keep in mind that Cooler Master has included an additional pair of clips in case you decide to add another 92mm fan to the setup in order to achieve a push-pull configuration. The single fan by default runs quietly even at full speed.

Cooler Master Hyper N620

Design

The Hyper N620 is positioned in the price category of high-end air cooling and it definitely fits the bill. With 6 copper heatpipes at the heart of numerous aluminum fins and a dual fan push-pull configuration by default, the Hyper N620 is a mean cooling machine. It is nice to see that the fans come pre-assembled so no tampering is necessary, unlike the Hyper TX3.

Specifications

Model	RR-B20-N620-GP
CPU Socket	Supports Core i7 with socket LGA 1366, Intel Socket T (LGA 775); Socket J (LGA 771 Skulltrail only), AMD Socket AM2+ / AM2 / 939 / 754
Heat Sink Dimensions	140.8 x 50.8 x 158 mm
Heat Sink Material	Cu base, Al fin, 6 heatpipes
Heat Pipes Dimensions	6 mm
Fan Dimension	120 x 120 x 25 mm (Blue LEDs) – 2 fans in push-pull configuration
Fan Speed	800 – 2000 R.P.M. (PWM)
Fan Airflow	83.6 CFM (Max)
Fan Air Pressure	4.43 mm H2O (Max)
Fan Life Expectancy	40,000 hours
Bearing Type	Rifle Bearing
Voltage Rating	12 V
Noise Level	16 – 28 dBA
Weight	1.87 lbs; 847g

The Hyper N620 retails for US$60 and is positioned in the high-end air cooling segment.

Installation

The installation process for the Hyper N620 was a bit more tedious than that of its smaller brother; not overbearing, just that there was more to do. This installation included mounting a backplate on the back of the motherboard. This is to help distribute the weight of the cooler and to ease the pressure that it places on your motherboard.

Anti-vibrations pads (some of which are also on the pre-installed fans) are placed under the screw heads on the front side, while nuts were used on the back side to tighten the cooler down. The necessary tool for the job was provided in the package. This allowed for the tightening of these hex nuts with a regular screwdriver.

It was not a difficult task to get done. However, I do think that the mounting screws could be made just a tad longer and the hex nuts a bit thicker. This would have made for an easier job at the back, securing the cooler.

When installed and running, the cooler boasts some nice, subtle blue LED lighting on both fans.

The Competition

For this review the competition is set up as follows:

A four-cooler shootout including the Intel stock cooler, the Hyper TX3, the Hyper N620 and the Tuniq Tower 120. Based on price, the Hyper TX3 will compete directly with the Intel stock cooler while the other newcomer, the Hyper N620, will go up against the tried and proven Tuniq Tower 120.

Intel Stock Cooler

Tuniq Tower 120

Methodology, Test System & Limitations

Methodology

For the duration of the testing, the system was running in an air-conditioned room, with the temperature regulated to maintain a steady 22 degrees Celsius ambient temperature. This was important because this would be the constant, while I test the variables (the coolers).

This temperature was verified using a temperature probe connected to a digital multimeter. The probe was placed adjacent to the 200mm intake fan at the front of the case.

To measure actual CPU temperature, Real Temp 3.00 was used because over time it has been proven to give accurate readings for Intel Core based CPUs. It was also easy to record the minimum temperature at idle and also the peak temperature during load.

Idle temperatures were recorded at the desktop with minimal system activity, resulting in the CPU being at less than 2% load across both cores.

For loading the CPU, OCCT 3.1.0 was used. Two separate tests were used to load the cores of the CPU to 100% for a sustained period of time. These are as follows and are described as stated by the program’s help section.

• CPU: OCCT – OCCT CPU test, where you’ll find the famous OCCT test that have been around for years now.
• CPU: LINPACK – Linpack CPU Test, based on a library provided by Intel. Similar to IntelBurnTest. CAUTION, it is VERY extreme and will make your CPU real hot.

Default settings were used for all tests.

Test System

The test system consisted of the following:

• Case: Cooler Master HAF 922
• PSU: Cooler Master UCP 1100w
• CPU: Intel Core 2 Duo E7200
• MB: ASUS P5E Deluxe X48
• RAM: G.SKILL 4GB DDR2-1000 PQ
• GPU: SAPPHIRE HD4870 1GB
• 4 Hard Disk Drives
• 1 Optical Drive

The purpose of keeping all four hard drives in the test system is to further simulate a real life scenario, instead of using only one so as to cut down on internal system heat.

Limitations

The stock settings for the CPU are 2.53GHz with its VID of 1.175V and this was used throughout the testing period.

The overclocked and overvolted settings were chosen based on a limitation. The CPU itself can attain a maximum stable clock speed of 3.8GHz. However, all the coolers except the Intel stock cooler was able to keep the system stable at that speed. Given this limitation, 3.60GHz with 1.400V was used throughout the testing period.

Now for the results of the testing.

Results – Stock

The results at stock speed and voltage are pretty straight forward and are shown in the chart below.

A point worthy of note is that all the coolers tested were able to fully dissipate the heat output from the CPU at idle. Hence, no difference in the temperatures recorded and this translates to saying that this processor in particular, might not run any cooler on air.

The first load test, CPU: OCCT, is where we see the first case of disparity. It shows all the aftermarket coolers reasonably outperforming the stock solution which registered a temperature of 44C. The Tuniq Tower comes out on top at 35C with the Hyper N620 right on its heels at 36C. The Hyper TX3 does well on a budget, registering a maximum load temperature of 38C.

The second load test, CPU: Linpack, kept more or less the same trend, each just one degree warmer than the other. The Intel cooler slipped further away from the pack with a reading of 51C. This is still acceptable for regular use though, keeping in mind that the strain that Linpack puts on the system most likely will never be replicated in real life daily use.

Results – Overclocked

The overclocked results depict a clearer image, a separation of men from boys really. The results are shown in the chart below.

Again at idle, despite the additional voltage, all coolers were able to keep the CPU at a very respectable 30C, which seems to be this processor’s minimum operating temperature on air cooling.

The first load test, CPU: OCCT, shows the Tuniq Tower outperforming its rivals by a clear six degrees with a temperature of 45C. Both of today’s featured coolers from Cooler Master managed to keep the peak temperature to 51C. The Intel stock cooler, as expected started to fall further from the pack, registering a maximum temperature of 66C.

The second load test, CPU: Linpack, poses an interesting scenario. Aside from the Intel cooler being totally out of the contest at 90C (way to hot for consistent usage); the budget Hyper TX3 falls behind the bigger boys now at 71C.

The interesting part is that the Tuniq Tower 120 and the Hyper N620 now trade blows. The Hyper N620 now comes out on top registering only 62C, while the Tuniq Tower 120 was only able to manage 65C. This was rather surprising considering the previous results and as such, this test was run several times and the same result was achieved every time. Therefore…it stands.

As mentioned before, the Linpack test puts much more pressure on your system than regular daily use (including gaming) and serves to show how these coolers will perform under maximum strain.

Conclusion

Not much needs to be said here, seeing as though the result charts speak for themselves. However, it needs to be said that this was an interesting review as the entry-level Hyper TX3 was a pleasant surprise and the Hyper N620 performed up to expectation.

In concluding, here is a breakdown of the coolers separately.

Intel Stock cooler

I would not recommend the stock cooler to any overclocker. However, for stock users, it performs as expected considering that you get it ‘free’ with your retail CPU.

Tuniq Tower 120

The venerable Tuniq Tower 120 shows that it is still capable of standing at the top among the newer high performance coolers.

Hyper TX3

A pleasant surprise, the Hyper TX3 provides exceptionally great value for a budget cooler. Small and affordable, this cooler packs a serious punch is capable of delivering a knockout blow on processor heat. The Hyper TX3 receives my full recommendation.

Hyper N620

Very impressive, the Hyper N620 stands tall among the top air coolers. It does very well to keep processor heat under control and does so in style with a good-looking and very functional design. The Hyper N620 receives the Editor’s Choice Award and also my full recommendation.

HAF 922 REVIEW UPDATE: As mentioned in my HAF 922 Review, the CPU retention hole in the case’s motherboard tray, really does save a lot of time when changing CPU coolers. It also helps a great deal with the installation, especially seeing as though the tray is not removable.

Leon Hyman
Senior Editor

Please join us in our Forums

Follow us on Twitter

For the latest updates from ABT, please join our RSS News Feed

Join our Distributed Computing teams

• Folding@Home – Team AlienBabelTech – 164304
• SETI@Home – Team AlienBabelTech – 138705
• World Community Grid – Team AlienBabelTech