Test setup

All of today's testing has been run on the following:

- Intel Core 2 Quad Q6600
- 4GB (2x 2GB) Corsair PC6400 DDR2 RAM
- Gigabyte GA-X38-DS5
- 250GB Western Digital Caviar SE16 hard drive
- Sony DVD-ROM
- NVIDIA GeForce 8800 GT 512MB
- Arctic Cooling MX-2 thermal paste
- Noctua NT-H1 thermal paste
- 1000W Thermaltake Toughpower power supply
- Windows Vista Home Premium (64-bit Edition)

CPU cooling performance

Now we get on to the real test - Just how much does Noctua's NT-H1 thermal compound improve the CPU cooling performance of a system?  To test this, we've employed our Intel test system to see how the application of NT-H1 compares to using both some generic thermal paste, as well as Arctic Cooling's MX2 compound, in each system.  To compare the properties of each type of paste, we've examined the CPU temperature first from idling at the Windows Vista desktop for a few minutes, then again after running the CPU-intensive Prime95's in-place FFT stress test for a thirty minute spell to produce the maximum amount of heat possible from the processor.

Let's start off with our idle temperature readings, where we can see the temperature of each CPU core in our Core 2 Quad system:

Overall, both of our third-party thermal pastes deal with heat transfer better than the generic no-name brand even at idle, but it's Noctua's offering which offers the most striking numbers, reducing temperatures quite considerably even with no load on the processor.

Once we fire up a multi-threaded version of Prime95 for a half-hour period, we find that Arctic Cooling and Noctua's thermal compounds offer almost identical results, with both reducing temperatures around seven degrees compared to our no-name thermal paste.  Considering that the feel of the MX2 and NT-H1 paste is quite similar overall, these close results shouldn't be particularly surprising.