Testing Environment and Results
When we assembled the system, we installed Microsoft Windows Vista Home Premium SP1 on it with all necessary drivers. Further testing was performed in a closed system case with all system components running at their nominal frequencies with automatic fan rotation speed adjustment enabled in the mainboard BIOS for all system fans.
The ambient temperature stayed at 24°C throughout the entire test session. All system component temperatures were read with SpeedFan utility. For your convenience, the results charts show only the temperatures for the first graphics card GPU, first CPU core and CPU packaging.
We used Ondico Shanghai PM-300 wattmeter to register the system power consumption and make sure that we selected an appropriate power supply unit for our system.
In idle mode our system wasn’t totally silent, but it was quiet enough: it only produced quiet hissing sound mostly generated by the graphics card cooler joined by the case fans. I think we may have spent a little more time on finding the best 80-mm fans, but it is a secondary matter. First of all we wanted to see, if the relatively low-power fans we picked would be enough to cool down our system.
The total system power consumption was about 150W, the CPU temperature didn’t exceed 50°C, the graphics card temperature - 60°C.
The heaviest possible workload created by FurMark test heated our graphics card almost to 100°C, which is, however, quite acceptable for a powerful graphics solution like ours, especially keeping in mind that FurMark is a synthetic benchmark that created much heavier workload than any of the contemporary gaming titles. The CPU heated up not as dramatically, because FurMark doesn’t load it too much.
The maximum registered power consumption of our entire system was 453W. In other words, f we take into consideration the efficiency factor of about 90%, the power supply unit was loaded as 45 x 0.9 = 407.7W.the level of generated noise increased dramatically and of course, the graphics card made the biggest contribution here. However, those of you who have tried running FurMark at least once shouldn’t be surprised about it: this benchmark can speed up pretty much any graphics card fan to its maximum. Nevertheless, the acoustic range was comfortable enough, with air hissing being the prevailing sound.