Just a followup on the heat issue; it's off topic, but info could be relevant for Matt and possibly others.
Start of this week I got my 2nd hand hexacore CPU; Core i7 980 (not the Extreme 980X version). I initially mounted it with my existing cooler; the Zalman CNPS 9900A. As seen above, my earlier i7 930 was running at 3.25 GHz, reaching temps of 84ºC with Prime95 max heat test. Started out at the stock speed of 3.33 GHz with the 980 CPU. In the same test, temps climbed to 90ºC. Disturbingly high, and obviously no scope for overclocking.
Had a brand new Cooler Master Hyper 212 EVO CPU cooler, given to me by a colleague who wasn't going to use it. Tried it instead. Idle temps seemed a few degrees lower, but was virtually shocked when I fired up the Prime95 test, and saw temps as low as 53-54ºC
Lots of scope for overclocking here then. Haven't found the limit, but currently it's running at 4.25 GHz, reaching temps of 71-72ºC after 1 hour of Prime95 max heat test. Appears to be rock stable. Ambient temp at testing has been around 22ºC
Looking at the two coolers, you wouldn't expect such a great difference. They're roughly the same size, albeit different shapes. Zalman is all copper, while the cooling fins on the 212 is aluminum, and copper is supposedly the better heat conductor. Zalman has a mirror smooth contact piece for the CPU, whereas the 212 is much rougher. The 212 has 'direct contact' heat pipes - the tubes are flattened in the part in contact with the CPU, whereas the Zalman has a separate contact piece, with the heat pipes running through them. Don't know if the 'direct contact' design is that much better - if not, I'm inclined to think the Zalman is defect. For both I used Arctic MX-4 Thermal compound, applied with the 'vertical line' method.
In other words; the cooler can make a huge difference, and based on the above, I can certainly recommend the Hyper 212 EVO wholeheartedly.
Cheers,
Erling