Noctua NH-D14 CPU Cooler

Performance Test Platform

The hardware chosen was picked to reflect that of a relatively modern computer system that an enthusiast may own. The configuration of the system is as follows:

  • Intel Core i7 2700k
  • ASRock Z68 Extreme3 motherboard
  • Crucial 2x2GB DDR3-1600 memory kit
  • Asus Radeon 4870 512MB video card
  • OCZ Vertex 3 240GB solid state drive
  • Western Digital Black 640GB SATA hard drive
  • Liteon iHAS124B SATA DVD-writer

Testing Methodology

Noctua's NT-H1 thermal compound was used for all CPU cooler testing results.  In our last thermal compound roundup, we found it to be one of the best performing thermal compounds tested.  Application of the thermal compound was done as recommended by the installation guide provided by the cooler manufacturer.  If no method was recommended, then a pea sized amount was applied to the center of the CPU's heatspreader (with pressure from the installed cooler to spread it).

We employ an open test bed (not installed in a case).  This is to remove positive/negative factors that a case can apply to coolers depending on design.  Keep in mind that temperatures achieved in this article will be lower than what a user will achieve when the cooler is installed in a case.  The ambient environmental temperature is maintained at 24 degrees celcius (plus/minus 1 degree celcius).

CPU temperatures are recorded using OCCT.  Since the temperature of each core is recorded individually, the average of these measurements is used to represent the 'Average' measure in the results.  The highest temperature recorded during testing among the individual cores is used to represent the 'Maximum' measure in the results.  Stress testing measurements are conducted using OCCT's CPU Test with the Small Data Set.  This is left to run for 30 minutes.  Idle measurements are taken 30 minutes after the the stress test has been completed (same method as the 'Average' stress test results).

The following motherboard configuration was used:

  • Hyperthreading disabled
  • Speedstep disabled
  • Turbo disabled
  • C1E, C3 & C6 State Support disabled
  • CPU Thermal Throttling disabled
  • CPU Load Line Calibration set to maintain level voltage regardless of load
  • Stock mode: default voltage, CPU Ratio set to 35 = 3500MHz
  • Overclocked mode: 1.3v, CPU ratio set to 45 = 4500MHz

Performance Test Results

Noctua's NH-D14 was a great performer.  It easily beat Corsair's and Intel's cooler.  This shouldn't come as a surprise given the difference in price tag for the coolers.  With idle CPU load, the NH-D14 stays in the same range as the budget cooler, but with increased heat load it pulls away easily.  The overclocked loaded CPU temperatures achieved by the NH-D14 were more than 10 degrees celcius lower than that of the boxed Intel cooler while the CPU was running at stock speeds, and 4 degrees celcius lower than that of the budget cooler range.

Conclusion

Subjectively, the NH-D14 heatsink and fan cooler package was moderately quiet.  At full speed, I found it to produce around the same amount of noise as the Corsair A50 cooler we previously looked at.  But when used with the ULNA connectors, the two fans are almost silent.

As far as air coolers go, Noctua hit it out of the park with the NH-D14.  It's quiet and the cooling performance it delivers is great.  While it's not cheap, you get what you pay for with Noctua's NH-D14: top of the line air cooling performance.