DeepCool Assassin III – Silent killer of noise and temperature

Pavel Sekerka

2 years ago

Key Features

Today I have the pleasure of introducing the Assassin III premium air cooler from DeepCool. Its size and construction is majestic, yet its design is pleasantly minimalistic without unnecessary bling. It also boasts high cooling efficiency (claimed TDP is up to 280 W). So I’m really curious to see how it performs in real tests on a hot CPU and what its acoustic performance will be like.

Key features of the cooler

The initial impression of the cooler is simply excellent. Right out of the box, it is clear that you have purchased a truly premium product. Even without the fans installed, you are holding more than a kilogram of aluminum and copper in your hand. The workmanship is of a high standard, and everything is nicely complemented by a pleasantly simple design.

The cooler consists of two dense stacks of fins, which are connected by a total of seven copper heatpipes with a diameter of six millimetres. The heatpipes converge in a solid copper base with a nickel plated surface. The contact surface of the base is polished to a high degree of flatness (to within 100 µm, according to the manufacturer) and its shape is slightly convex for maximum effective contact with the processor’s heat spreader, which is usually slightly concave.

The upper surfaces of both heatsink towers are nickel plated and polished to a mirror shine, which contrasts nicely with the black covers of the heatpipes. Although the manufacturer refers to them as spoilers, their function is purely aesthetic. The passive part of the cooler is complemented by a pair of matte black fans, which make the whole cooler look really massive. Although no part of this cooler features RGB backlighting, its size and appearance will certainly make it a striking element of any build.

After a long time, I am really excited about the accessories in the cooler package. This is just what a premium product should look like. The mounting systems for both Intel and AMD sockets are accompanied by a well-crafted screwdriver with a magnetic tip. In addition to thermal paste paste with applicator, you also get a disposable cleaning cloth for perfect preparation of the IHS surface before installation. And as an added bonus to the fan cable splitter, you get a pair of adapters to lower the maximum fan speeds if you’re aiming more for silence than low temperatures. You’ll even find an extra set of clips for a possible third fan, though I don’t think adding one will increase the efficiency of the cooler in any way, rather it will only increase the noise level.

The mounting of the cooler is simple and there are no tricky parts. The whole system is very robust and apart from the (very strong) plastic backplate, all components are metal. Overall, I rate this part of the cooler an absolute A-star.

Pump, radiator and fan parameters

When comparing the Assassin III cooler with its direct competitor, the Noctua NH-D15, it can be noticed that the coolers are very similar in size and weight. The Assassin III is slightly slimmer thanks to the standard 140 mm fans, but it is 5 mm taller and heavier overall. Price-wise, however, the Assassin III doesn’t mess around either, making it one of the most expensive air coolers on the market.


Cooler		DeepCool Assassin III	Noctua NH-D15 chr.blk	Noctua NH-D15
		DeepCool Assassin III	Noctua NH-D15 chr.blk	Noctua NH-D15
Supported sockets		Intel 115x, 1200, 1700, 20xx; AMD AM4	Intel 115x, 1200, 1700, 20xx; AMD AM4	Intel 115x, 1200, 1700, 20xx; AMD AM4
Height		165 mm	160 mm	160 mm
Width		140 mm	150 mm	150 mm
Depth with fan		161 mm	161 mm	161 mm
Weight		1464 g	1320 g	1320 g
Maximum RAM height		32/49 mm *	32/64 mm *	32/64 mm *
Warranty	Záruka	3 years	6 years
MSRP		€ 113	€ 114	€ 101

/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1177" as a base selector for example: #supsystic-table-1177 { ... } #supsystic-table-1177 tbody { ... } #supsystic-table-1177 tbody tr { ... } */

* Maximum RAM height when the front fan position is unoccupied. I recommend using only low profile memory modules when this position is occupied. Alternatively, the front fan can be positioned higher, but this may lead to reduced cooling efficiency and also increase the overall height of the cooler, which should be kept in mind when selecting a computer case.

You get a pair of TF-140S fans with the cooler. These fans offer very similar performance to their Noctua counterparts on the NH-D15 cooler with very solid airflow and static pressure values. The speed range is also similar. So I expect a very close battle in terms of temperatures and noise levels.


Fan	DeepCool TF-140S	Noctua NF-A15-HS chr.sw	Noctua NF-A15-HS
	DeepCool TF-140S	Noctua NF-A15-HS chr.sw	Noctua NF-A15-HS
Bearing	FDB	SSO2	SSO2
Speed range	400–1400 rpm	300–1500 rpm	300–1500 rpm
Max. airflow	153.5 m³/hod	140.2 m³/hod	140.2 m³/h
Max. static pressure	1.79 mm H₂O	2.08 mm H₂O	2.08 mm H₂O

/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1178" as a base selector for example: #supsystic-table-1178 { ... } #supsystic-table-1178 tbody { ... } #supsystic-table-1178 tbody tr { ... } */

Testing methodology

As a test processor and heat generator, I am now using an Intel Core i7-11700K on the ASUS TUF Gaming Z590-Plus WIFI motherboard. The entire setup is mounted on a Streacom BC1 frame with no fans to provide airflow. Acoustic shielding (acoustic foam on OSB boards) is installed around the test setup to eliminate ambient noise.

The Voltcraft SL-100 calibrated noise meter is placed perpendicular to the cooler in a fixed position relative to the motherboard at a distance of approximately 35 cm from the CPU socket. Stress tests are performed using Prime95 with a duration of 30-35 minutes. The coolers are tested at three standardized noise levels (36, 39 and 42 dBA) and finally at maximum speed. The tests are run in two sets according to the heat output of the processor.

The first set of tests is at 125 W, the enforced PL1 limit of the processor. The second set of power consumption is then about 205–210 W (4500 MHz at 1.285 V). However, this second level represents a very high concentration of heat per mm² due to the small size of the chip itself. Thus, not all coolers can be expected to pass this set of tests at all noise levels. The highest temperature achieved is always recorded at the end of the test.

All data regarding temperatures and heat output of the processor is obtained from the HWiNFO64 application. The measured values are then corrected against the room temperature as the resulting Delta T value. If the heatsink fails a load with a Delta T value below 75 °C, the test is marked as a failure. Any higher value means a processor temperature over 95–100 °C and a risk of thermal throttling.

Results – 36 dBA

Results – 39 dBA

Results – 42 dBA

Results – maximum speed

Conclusion and rating

I have a very good impression of the DeepCool Assassin III cooler in all aspects. Right out of the box you know you have a premium product in your hands. This continues with practically flawless workmanship and pleasantly rich accessories. All complemented by a minimalist yet luxurious design.

Most importantly, DeepCool Assassin III can handle very high heat output without being excessively noisy. The only flaws remain the relatively short warranty and the high price tag. However, we can only conclude that this cooler is overall very successful and rightfully wins the editorial award „Approved“.

English translation and edit by Jozef Dudáš


DeepCool Assassin III
+ Excellent cooling efficiency
+ Solid workmanship
+ Extensive accessories
+ Low noise level
- Higher price
- Shorter warranty compared to competitors

/* Here you can add custom CSS for the current table */ /* Lean more about CSS: https://en.wikipedia.org/wiki/Cascading_Style_Sheets */ /* To prevent the use of styles to other tables use "#supsystic-table-1179" as a base selector for example: #supsystic-table-1179 { ... } #supsystic-table-1179 tbody { ... } #supsystic-table-1179 tbody tr { ... } */

Continue: Measurement methodology