The best way to compare the performance of similar parts is to use the same test bed for each. Unfortunately, our test bed simply isn't elaborate enough for one of the cases being evaluated. A look at the thermal results consequently requires some explanation.
Rosewill, Enermax, and Thermaltake presented variations on a theme, with Thermaltake using the power supply’s intake fan to do the same job as Rosewill’s and Enermax’s top-panel fan. As such, a power supply with a bottom-mounted 140 mm intake would likely have boosted Thermaltake’s cooling a little. The Antec HCP-1200 used in this test has flow-through cooling with an 80 mm rear fan instead.
This is also the point where we must buck the myth that bigger is better in regards to keeping case temperatures low. Airflow is the key to keeping a CPU cool, and some of the cases with the highest air velocity at the CPU cooler are fairly small. In the case of Azza’s Fusion 4000, open space above the lower system’s motherboard most likely reduced the exhaust fan’s ability to create velocity at its CPU cooler. Had a mini-ITX motherboard been mounted in this case’s upper portion, it may have acted as a diverter to increase velocity in this critical part. Thus, this case should really be tested in the dual-system or dual-radiator configurations for which it is designed.
The few reviewers who have called Rosewill’s Thor V2 a quiet case are obviously not testing four GeForce GTX 580s at full fan speed. Though the difference is small, it does allow slightly more graphics noise to escape when compared to its competition.
What the noise chart doesn’t show is pitch, and that’s a shame because Thermaltake’s Armor+ VH6000BWS emits an annoying groan from its side-panel fan. We found that simply sticking a napkin on half the fan was enough to stop its soft (yet conspicuous) tone.
To compare cooling and noise, we first made a percent scale of each system’s actual performance. Dividing the total average temperature of all systems by each system’s average temperature creates a scale that awarded a higher performance “score” for lower temperatures. Dividing the average noise level of each system by the total average noise level of all systems creates an inverse performance score for low noise. Finally, dividing the heat “score” by the noise “score” creates an “Acoustic Efficiency” number where the average is 100%.
Of course, nothing is 100% efficient. Subtracting the 100% baseline from the resulting calculations gives chartable values that reflect how much more or less efficient a case is relative to the average of all cases. The Thor V2’s cooling prowess puts it at the top of this chart in spite of its slightly worse noise suppression.
The Armor+ VH6000BWS again suffers from our choice of power supply, since its intake vents are nowhere near the CPU cooler, and would be best paired with a power supply that has a 140 mm intake on its bottom. The Fusion 4000 suffers in the same way, though it has far more room for additional fans.