Intel Core i7 2600K
G.Skill Pi 2x2GB 2400mhz 9-11-9
MSI R6970 Lightning
CPU Cooling: Thermalright Ultra 120 Extreme
Silverstone Strider 1000w PSU
Testing everything at stock. Like the setup is being run for the first time out of the box, other than setting memory speed, timings, and voltage.
Next, I bumped the CPU multiplier up to 42x to check for CPU scaling with stock card clocks. Obviously some benchmarks will scale more than others with only a CPU frequency increase. I was actually surprised to see such a gain in Heaven though, given how GPU-dependent the benchmark is. That would seem to indicate a bottleneck somewhere between the ~3.5ghz used in the first round of testing, and the ~4.2ghz used in this one.
Next I started testing some card clocks. Obviously, 1000mhz was a piece of cake with stock voltage. I still have the BIOS switch set to "Silent" and the card is exactly that. With the CPU still running on the 42x multiplier, I did some quick tests at 1000mhz core, 1500mhz memory.
And with the CPU on the 50x multiplier.
Edited by sno.lcn - 3/28/11 at 8:09am