You can check the reviews / data I didn't bookmark any of them.
One of the charts here they have it at 165w 38db noise and NH-D15 is only 1c better
Why people not do Aida FPU only? Cpu + FPU is actually a softer test... Well, at least it wasn't cpu + FPU + cache, that's even softer.
At the 165w chart, it does 1C better while being almost 2 times louder(at minimum), and most users don't run their fans at 100% anyway. So while there's extra performance to be had, not many are going to utilize it.
That price is definitely hard to beat... though Thermalright's support and availability is a bit lacking in the US, which stops me from diving in on their products.
If I wanted to get replacement fans, or use some as case fans, I'm screwed.
My curiosity has been disappointed by the Thermalright Noctua Killer claims before: I did a Vs using the Silver Arrow IB-E Extreme Rev. B and True Spirit 140 Power against the NH-D15S on the 7820X, and they didn't really 'kill' it, from what I remember.
Swapped fans between the 3, tried different case fan configs... D15S came out on top almost every time. The almost comes from the IB-E just barely beating out the D15S in single fan mode while being much louder to do it.
Add a second fan to D15S, and it lost in cooling too - though just as slightly.
The TS140P was far behind both by almost 10 degrees(C)... At least it was quieter than the other 2? Until I tried the 2 TY-143, IPPC 3000 140mm on it...
Some of these older reviews out there were testing with quad core i5s and whatnot, in which some of the Thermalright coolers are more efficient than Noctua's at lower heat loads.
When testing a cooler, one should at least be using a hot and heavy cpu, or a dummy heater...