Testbed and Methods
So, having this abundance of system cases varying both internally and externally, we couldn’t help checking how their features affect their performance from the end-user’s point of view.
We didn’t invent sophisticated testing methods for once, but chose a simple and reliable approach: we assembled a top-end computer configuration in each system case, powered the system on, and read the temperatures of the system’s main components. Since the system cases were all without preinstalled power supplies, we didn’t have to evaluate the latter (it is rather inexpedient to buy a system case with a preinstalled power supply today, considering the multitude of existing PC configurations).
The system was assembled “as is”. That is, we didn’t change the default speeds of the preinstalled fans and performed our tests in a closed and fully assembled case at a constant ambient temperature, maintained with the help of an air conditioner. We took care to lay all the cables inside the case in such a way as to avoid their hindering air circulation.
We put the following hardware parts in the tested system cases:
- Intel i955X mainboard;
- Intel Pentium Extreme Edition 2.8GHz CPU (Smithfield core);
- Thermaltake Big Typhoon cooler;
- 2 x 512MB Kingston KHX5400D2/512 DDR2 SDRAM;
- ATI Radeon X850 XT Platinum Edition 256MB graphics card;
- Western Digital WD1600JB hard disk drive;
- Lite-On SOHW-1693S (White) and NEC ND-3540A (Black) DVD-recorders;
- Zippy HP2-6460P power supply.
We took a Big Typhoon cooler for two reasons. First, although the Smithfield processor works at a rather low frequency (2.8GHz), it is a dual-core Extreme Edition all the same. Its heat dissipation is quite high even at the seemingly low clock rate. And second, the Big Typhoon is a typical top-end monster of an air cooler as they are today. So we were just curious to see how this really huge and heavy thing was going to fit into system cases from the same manufacturer.
This cooler was also very suitable for the testing purposes because of its low rotational speed, 1300rpm. The fans preinstalled in the reviewed cases have similar speeds, so the noise from the CPU cooler won’t interfere with our estimating the noise from the system fans proper.
We performed our tests in the following modes:
- CPU Burn (two copies of the CPU Burn utility running)
- HDD Burn (two sets of files, a 7GB folder with small files and a 40GB folder with large movie files, are being copied simultaneously from one partition of the drive to another)
- VGA Burn (Unreal Tournament 2004 is running with the maximum graphics quality settings)
We read the temperatures of the CPU and mainboard through the Motherboard Monitor utility of the latest version; the temperatures of the graphics card’s PCB and of the graphics processor were read with RivaTuner. We didn’t measure the temperature of the memory modules as it didn’t depend on the case design. The modules we used were equipped with heat-spreaders and needed just some weak airflow to work normally. In our tests, however, they even got a strong stream of air from the CPU cooler and remained cold to the touch even in the hardest operational modes.
The temperature of the hard disk drive was read with HDD Thermometer.
The room temperature was 20°C at the time of our tests and remained constant throughout them.
The temperatures of the system components are read only after they have fully stabilized.
So, that’s the end of the introductive section, and we can now get closer to the system cases proper.