Review: Corsair Obsidian Series 550D

Specification and test methodology

To find out how well the chassis can cool, we've loaded it up with an X58 test platform and run a series of CPU and GPU benchmarks for comparison against a variety of mid-tower alternatives.

Firstly, CPU cooling performance is measured by setting fan speed to low and recording CPU temperature when the system is idle. We then apply a 15-minute load of the Prime95 stress test and record CPU temperature again. To demonstrate the chassis' optimum cooling capability, we also set all fans - including the CPU cooler - to high speed mode and re-run the Prime95 torture test for a further 15 minutes.

Secondly, GPU temperature is also recorded in three states; idle, load with fans at standard speed and load with fans at high speed. In this test, GPU load is defined as a 15-minute stint of Furmark.

And finally, to find out how loud the chassis can get we're using a PCE-318 noise meter placed at the front of each chassis to take decibel readings with the system idling in two modes; low fan speed and high fan speed.

All chassis are tested only with the standard manufacturer-supplied fans, and to take into account the fluctuating ambient temperature, our graphs depict the delta temperature - that's actual CPU/GPU temperature minus the ambient. Just so we're clear, room temperature at the start of testing was recorded for each chassis as follows: