Testing Methodology

All comparison testing was done on an open test bench keeping ambient temps 23 deg C. ± 1 C. . If these conditions could not be met, testing was ceased and carried out at a later date when conditions were within stated parameters.

Recorded temps were as reported in Speedfan (same readings as DFI’s Smartguardian monitoring software). Load temps were taken after 1 hour Orthos blend, and are the average of the last 30 seconds disregarding high and low spikes. Idle temps were taken 15 minutes after 1 hour Orthos Load ceased, using the same criteria.

As with all software monitoring software, baseline temperature readings in speedfan cannot be verified as accurate, and thus should be considered only as a subjective comparison against other temperatures recorded on the same test bed using the same software/methodology. It's by no means a perfect solution for benchmarking, but unless CPU & Motherboard manufacturers can devise an auto calibrating temperature sensor system for CPUs, it remains the only way to consistently record results without affecting CPU/Heatsink mating surfaces as current thermal probes do.

Arctic Cooling MX-2 thermal paste was used for all comparisons unless otherwise stated. Application of thermal paste was according to manufacturer’s instructions, and allowed to cure for a minimum of 48 hours under moderate load prior to testing. Installations not meeting expectations were re-applied. Only best results are represented.

Control temps with AC 7 Pro were recording using the same methodology with the fan speed manually set to 100% in bios. MX-1 thermal paste was used.

In cases of testing procedures outside of the norm, any deviation from the standard testing methodology will be clearly indicated as part of the set-up for that particular test.

OC settings were Clock speed 3370 Vcore 1.55V.

• Intel E6600 lapped
• DFI P965-S Dark
• 2 X 1 Gig Supertalent PC6400
• Ultra Xpro 600W
• Arctic Cooling AC 7 Pro
• BFG 6600GT
• Highspeed PC Tech Station

This chart will set the stage for all of the following charts. Since the NT06 ships with a fan controller, and cannot be controlled by bios PWM, the full benchmark test was carried out at Minimum RPM (1045 as recorded in bios), Medium RPM (1800 ± 15 as recorded in bios), and 100% RPM (2566 ± 10 as recorded in bios).

Already evident in the stock idle CPU temperatures is the obvious inaccuracy of baseline readings which are below ambient temps. As mentioned in the testing methodology, software temperature readings should be considered as accurate only in respect to other readings taken from the same test bed using the same software / methodology.

As it stands, no great surprises in the stock idle temperatures. The NT06 holds its own against the current budget favorite AC 7 Pro, but doesn't walk away with any prizes.




Once again, the NT06 doesn't wow us under stock load conditions. Minimum RPM is dead quiet, but sacrifices some cooling performance. Medium range 1800 RPM is on par with the control unit, and subjectively puts out around the same db (hard to compare the two using rated specs as the AC unit rates according to sone vice db). 100% RPM manages to eke out a slight advantage, however the noise from the fan at 100% is clearly the most prominent sound in the room.

So far, stock temperatures on the NT06 have failed to impress. The only outstanding results are the minimum RPM sound levels.









As mentioned in the overclock idle temperature testing, the lowest setting failed to complete the overclocked load benchmark. A total of four tests were carried out on two different days, re-attempts included re-seating and re-application of thermal compound. In each case, the benchmark failed at or around the 50 minute mark of Orthos Blend. Due to CPU/Core/Northbridge temperatures within ranges which have previously completed the overclock load test, the most likely suspect for the failure is a lack of airflow to the voltage regulation circuitry around the CPU socket.

Setting aside the failed benchmark, it's interesting to note that the NT06 finally seems to distinguish itself from the control unit under the medium and 100% RPM overclocked load tests. However slight, the results are enough to call it a clear win for the NT06 over the AC 7 Pro.

The NT06 may have come out a winner in the overclocked load tests, but considering the size of the cooler, and the volume of air the fan is able to move through it, better results were expected. Could there be a mitigating factor in the less-than-stellar results?

Read on to find out.