Add to Reddit
Testing The Armada
The Testbed
| Processor | Intel Core 2 Duo E6600 (2.4GHz) |
| Motherboard | MSI P965 Neo |
| Memory | Two Mushkin 1GB XP2-5300 DDR2 Memory Modules |
| Graphics Card | NVIDIA GeForce 7600 GT (256MB) |
| Hard Drive | 400GB Seagate Barracuda 7200.10 (SATA) |
| Thermometer | Omega HH506R with two Type K thermocouples |
Testing Methodology
We set out to test the VIZO Armada's performance with a standard Intel Core 2 Duo system with two memory modules. We installed the Armada on the memory module furthest from the CPU cooler (which we will call First DIMM ) and angled it to blow at the second module as well.
) |
The Omega HH506R thermometer was used to monitor both module and ambient temperature. The T1 thermal probe was inserted next to the top-most memory chip of the first memory module. The T2 thermal probe was placed just above the blades of the top-most fan of the Armada cooler.
The system was then started up and set to run the Company of Heroes benchmark to stress the memory modules. The HH506R thermometer was set to monitor and record the maximum stable T1-T2 delta temperature throughout the test. The test was repeated twice and the results averaged.
The system was then powered down, and the T1 thermal probe inserted next to the top-most memory chip of the second memory module. The T2 thermal probe was left in its original position. The system was restarted and then set to run the Company of Heroes benchmark.
The HH506R thermometer was once again set to monitor and record the maximum able T1-T2 delta temperature throughout the test. The test was repeated twice and the results averaged.
The entire test was then repeated for both memory modules with the Armada removed.
The Results
Please note that the results shown below are T1-T2 delta temperatures, which give us the actual differences in temperature between the memory chip and ambient air (as measured at the fan inlet).
With the Armada's three fans blowing onto the heatspreaders of the Mushkin PC2-5300 memory modules, we recorded very low temperature deltas of 5.9 °C for the first DIMM and 7.5 °C for the second DIMM. Because the average ambient temperature of 31.3 °C, this means the memory chip on the first DIMM only hit a high of 37.2 °C while the memory chip on the second DIMM hit a high of 38.8 °C. That's really not hot at all.
But when we retested the modules with the Armada removed, we noted a marked rise in temperature. The first DIMM recorded a jump of 8.7 °C while the second DIMM saw its temperature increase by 5.8 °C. With the ambient temperature at 31.3 °C, the first DIMM was as hot as 45.9 °C while the second DIMM was about 44.6 °C.
The slight reduction of the second DIMM's temperature when tested without the Armada can be attributed to its close proximity to the CPU cooler.
Even though the Armada is nothing more than just three fans attached to the top of a memory module, the air flow it produced significantly reduced the chip temperature of both memory modules. It's a clear testament to the effectiveness of an active cooling solution, as opposed to a passive one.
