Hi All,
I have just ordered replacement parts from OCUK which should arrive tomorrow. All being well, I will be able to get the testing done then. This is the testing procedure I will be implementing:
- Intel Stock Cooler (Noctua P12 fan as exhaust) @ 2.67GHz
- Corsair H50 (Stock Fan) @ 2.67GHz
- Corsair H50 (Noctua P12 Fan) @ 2.67GHz
- Corsair H50 (Stock Fan) @ 3.80GHz
- Corsair H50 (Noctua P12 Fan) @ 3.80GHz
- CoolIT ECO (Stock Fan) @ 2.67GHz
- CoolIT ECO (Noctua P12 Fan) @ 2.67GHz
- CoolIT ECO (Stock Fan) @ 3.80Ghz
- CoolIT ECO (Noctua P12 Fan) @ 3.80Ghz
I will record both the CPU temperature and core temperatures, room temperature, NB temperature and ambient temperature for each of these.
With the Corsair H50, the stock fan will be setup so that it blows into the case as it is in a stock configuration. But with the Noctua Fan, it will be setup blowing out of the case so it is able to produce comparable results with the ECO. All coolers will have Arctic Cooling MX-3 thermal paste applied.
The test system will have the following specifications:
- Intel Core-i7 920 D0
- Asus P6T Motherboard
- 3 x 2 GB Corsair Dominator DDR3 RAM
- 500GB 7200rpm Seagate Barracuda Hard Drive
- Zotac GTX 275 Graphics Card (896MB)
- Samsung SH-B083L 8 x Blu-Ray Drive
- Thermaltake Toughpower XT 750W PSU
- Coolermaster CM-690 II Advanced Dominator Case
- Viewsonic VX2260WM 22" 1080P Monitor
- SteelSeries 7G Mechanical Keyboard
- SteelSeries XAI Gaming Mouse
The Case has one 140mm intake fan in the front, one 140mm exhaust fan at the top and one 120mm intake fan on the side panel.
To test load temperatures, I will run Prime 95 for 30 mins (would do longer but it would take far too long) and after the test has stopped, I will record the idle temperature after leaving the system on the desktop for 30 mins. Temperatures will be recorded using Everest Ultimate.
If anyone has any suggestions as to ways I can improve this procedure before I start, please comment below.
EDIT: Will actually use my Corsair HX 850W PSU instead - not that is should make too much of a difference to the results