[sswilson]

Originally Posted by SKYMTL

This is a VERY good point and I am glad you brought it up. The reason for the recommendation is that the test I run for power consumption is VERY GPU intensive while the CPU is pretty much left at idle. As a matter of fact, if I look at the CPU useage log after the test, the CPU never goes above 5% on any of the cores. I do this so the CPU does not affect the overall power consumption figures and we can see a clearer picture of which GPU requires the most power.

I have also been thinking of adding a "worst case" scenario consumption test much like I did in the PSU reviews. Bacially running the Batch Size Test on the GPU while running OCCT in the background on the CPU. This would give a clearer indication of the power consumption in a GPU-intensive game that uses the CPU for physics calculations.

A good example of this would be in Crysis Warhead where I have routinely seen power consumption above 575W from the wall using the EVGA GTX 260 Core 216 Superclocked card.



They will be back but I am presently building the new PSU and GPU test system which will center around an X58 / i7 combo. The first PSU tested on it should be a little surprise from Corsair.

I'd definately want a good 750W if there was a possibility for that kind of draw... I get nervous whenever real power draw gets over 60 - 70% for extended periods of time.