Introduction

We tested the Noctua NH-D14, a huge CPU cooler with two tower heatsinks, six U-shaped heatpipes, and two fans, one 120 mm and one 140 mm. Let's see if this monster really cools a CPU.

The NH-D14 box is also huge, and made of cardboard, as you can see in Figure 1.

click to enlarge
Figure 1: Packaging

In Figure 2, you can see the accessories that come with the cooler: manuals, installation hardware, a case badge, and a tube of thermal compound.

click to enlarge
Figure 2: Accessories

In Figure 3, you see the cooler itself.

click to enlarge
Figure 3: The NH-D14

In the next pages, you will see this cooler in detail.

The NH-D14

In Figure 4, we have a front view of the cooler, where we can see the 120 mm fan and the six heatpipes.

click to enlarge
Figure 4: Front view

In Figure 5, you can see how big this cooler is. Each tower has its own fan.

click to enlarge
Figure 5: Side view

In Figure 6, you can see the back of the cooler. Note that the heatpipes are equally distributed inside the heatsink.

click to enlarge
Figure 6: Rear view

In Figure 7, you have a top view of the cooler. The fins have "teeth" in order to increase the turbulence and give better contact between the airflow and the dissipating surface.

click to enlarge
Figure 7: Top view

The NH-D14 (Cont’d)

In Figure 8, you can see the base of the cooler, in nickel-plated copper. Note it has no mirror-like finishing. You can also see the heatpipes and the NT-H1 thermal compound tube.

click to enlarge
Figure 8: Base

In Figure 9, you see the heatsink without the fans. Actually, you need to remove at least the middle fan, in order to attach the cooler in place. Note the silicon spacers that absorb the vibration of the fans.

click to enlarge
Figure 9: Without the fans

In Figure 10, you can check the fans. They have teeth on the edges of the blades, that help to reduce the noise level. The wire clips make the removal and installation of the fans easy. The fans have three-pin connectors (i.e., with no PWM speed control).

click to enlarge
Figure 10: Fans

The cooler comes with a Y extension, which allows you to connect both the fans to one motherboard (or fan controller) output. There are also two adapters that reduce the fan speed, if you prefer less noise. They can be seen in Figure 11 along with four rubber holders that can be used if you prefer to install one of the fans as a case fan.

click to enlarge
Figure 11: Adapters

Installation

The installation of the cooler is not simple. First, you need to prepare and install a backplate on the solder side of the motherboard.

click to enlarge
Figure 12: Backplate

After that, you will screw two clips on the component side of the motherboard.

click to enlarge
Figure 13: Clips

Then you put the cooler over the CPU and screw it to the clips. As we mentioned before, you must remove the center fan in order to do that.

click to enlarge
Figure 14: Installed

Then you can reinstall the fan(s). In Figure 15, note that the right hand fan extends over the memory modules, which can be a problem if your memories have tall heatsinks.

click to enlarge
Figure 15: Installed in our case

How We Tested

We tested the cooler with a Core i7-860 CPU (quad-core, 2.8 GHz), which is a socket LGA1156 processor with a 95 W TDP (Thermal Design Power). In order to get higher thermal dissipation, we overclocked it to 3.3 GHz (150 MHz base clock and 22x multiplier), keeping the standard core voltage (Vcore), which was the maximum stable overclock we could make with the stock cooler. Keep in mind that we could have raised the CPU clock more, but to include the stock cooler in our comparison, we needed to use this moderate overclock.

We measured noise and temperature with the CPU idle and under full load. In order to get 100% CPU usage in all threads, we ran Prime 95 25.11 (in this version, the software uses all available threads) with the "In-place Large FFTs" option.

We compared the tested cooler to the Intel stock cooler with a copper base (included with the CPU), as well as with other coolers. Note that in the past, we tested coolers with a socket LGA775 CPU, and we retested some "old" coolers with this new methodology. This means you can find different values in older reviews than the ones you will read in the next page.

Room temperature measurements were taken with a digital thermometer. The core temperature was read with the SpeedFan program (available from the CPU thermal sensors), using an arithmetic average of the core temperature readings. During the tests, the left panel of the case was open.

The sound pressure level (SPL) was measured with a digital noise meter, with its sensor placed 4" (10 cm) from the fan. We turned off the case and video board cooler fans, so they wouldn't interfere with the results, but this measurement is only for comparative purposes, because a precise SPL measurement needs to be made inside an acoustically insulated room with no other noise sources, which is not the case here.

Hardware Configuration

• Processor: Core i7-860
• Motherboard: Gigabyte P55A-UD6
• Memory: 4 GB A-Data AX3U1333GB2G8-2G (DDR3-1333/PC3-10700 with 9-9-9-25 timings), configured at 1,200 MHz
• Hard disk: Seagate Barracuda XT 2 TB
• Video card: Zotac GeForce GTS 250
• Video resolution: 1680x1050
• Video monitor: Samsung Syncmaster 2232BW Plus
• Power supply: Seventeam ST-550P-AM
• Case: 3RSystem L-1100 T.REX Cool

Operating System Configuration

• Windows 7 Home Premium 64 bit

Software Used

• Prime95
• SpeedFan

Error Margin

We adopted a 2 ^oC error margin, meaning temperature differences below 2 ^oC are considered irrelevant.

Our Tests

The table below presents the results of our measurements. We repeated the same test on all coolers listed below. Each measurement was taken with the CPU at idle and at full load. In the models with a fan supporting PWM, the motherboard controlled the fan speed according to core load and temperature. On coolers with an integrated fan controller, we set the fan at the minumum speed on the idle test and at full speed on the full load test.

		Idle Processor			Processor at Full Load
Cooler	Room Temp.	Noise	Speed	Core Temp.	Noise	Speed	Core Temp.
Intel stock (socket LGA1156)	14 °C	44 dBA	1700 rpm	46 °C	54 dBA	2500 rpm	90 °C
Cooler Master Hyper TX3 G1	14 °C	47 dBA	2050 rpm	33 °C	56 dBA	2900 rpm	62 °C
Zalman CNPS10X Extreme	14 °C	45 dBA	1400 rpm	27 °C	53 dBA	1950 rpm	51 °C
Thermaltake Silent 1156	14 °C	44 dBA	1200 rpm	38 °C	49 dBA	1750 rpm	69 °C
Noctua NH-D14	14 °C	49 dBA	1250 rpm	27 °C	49 dBA	1250 rpm	53 °C

In the graph below, you can see how many degrees Celsius hotter the CPU core is than the air outside the case, on full load. The lower this difference, the better is the performance of the cooler.

Main Specifications

The main characteristics of the Noctua NH-D14 are:

• Application: Socket LGA775, 1156, 1366, AM3, AM2+, and AM2 processors
• Fins: Aluminum
• Base: Copper
• Heat-pipes: Six copper heat-pipes
• Fan: one 140 mm and one 120 mm
• Nominal fan speed: 1,200 rpm and 1,300 rpm
• Fan air flow: 64.92 cfm and 54.33 cfm
  
• Maximum power consumption: 1.2 W and 1.08 W
• Nominal noise level: 19.6 dBA and 19.8 dBA
• Weight: 2.73 lbs (1240 g)
• More information: http://www.noctua.at
• Average price in the US*: USD 90.00

* Researched at Newegg.com on the day we published this review.

Conclusions

The Noctua NH-D14 CPU cooler is a monster, being one of the biggest coolers we tested so far. Because of this, we expected excellent performance.

It is not particularly beautiful. All its appeal lies in the fact it is big, and in the enthusiast world, big coolers are nice coolers. It is also quite expensive.

The performance level is very good, but not outstanding. The point here is that it kept our CPU very cool with a low noise level. The included fans are optimised for silence, not for high performance. This means that it can probably achieve higher cooling performance if you change one or both the fans for higher rpm ones.

In short, if you are looking for a big cooler which can deliver high cooling performance while maintaining an acceptable noise level, the Noctua NH-D14 is the cooler you are looking for.