Thought I'd post a separate remark instead of editing my last.
THIS looks like a VERY RELIABLE comparison review, with several top-end coolers:
Benchmark Reviews (.com) Prolimatech Megahelem test-results summary page
I'm running TRUE and Noctua NH-U12P coolers. This Megahelem must have a superior (lower) thermal resistance than all of these coolers and several water-cooling kits. I'm guessing that it registers at somewhere between 0.08 C/W and 0.09 C/W.
Usually, the reviews show only temperatures at load above room-ambient. It doesn't much matter how stressful the loading program is, if it pushes the processor to 100% capacity. The temperature differentials are more telling.
There's only about a 1+(C-degree) improvement over the TRUE cooler. For some reason, their test results show significant differentials between the TRUE, the OCZ Vendetta 2 and the Noctua, when other reviews show those units to be neck and neck. BUT -- the bench-configuration deploys an I7-920 OC'd to about 3.6 Ghz. The TDP for these cores, as I understand it, is about 135W, so the overclocking would push the thermal wattage higher.
The point here is that the higher the thermal wattage, the narrower the differences between good heatpipe coolers and the stock cooler, and the narrower will be the improvement over "second-best." This, because:
Thermal-Resistance = (T2 - T1) / W
Where the difference in temperatures at the "hot" and "cool" ends of the device is measured in Celsius or Kelvin as (T2-T1), and W is the thermal wattage. As W increases for a device of given TR, the difference in temperatures shrinks. This would also apply to the improvement-differentials between two different devices, I think.
I'm impressed with the Prolimatech Megahelem -- second-hand impression until I can obtain and test one myself.
I'd say that it's probably overkill for an E8400, but you sure would notice the difference between the Megahelem and the TRUE (or Vendetta-2, Noctua, etc.) because of the lower thermal wattage even under extreme over-clocking for an E8400 in comparison to this review's I7-OC test-bed. And fact is, if you're going to spend maybe $40 on another cooler, why not get the Megahelem instead?
Further, lapping the processor cap (for an E8400) might gain up to 5C, and diamond-thermal-paste another 2 to 4C degrees. Note especially their discussion of thermal conductivity of materials, and they don't include nano-diamond pastes in the comparison!!