I was testing 3 Enermax ETS-F40-FS coolers (regular, black and ARGB) with Ryzen 5900X. All had +/- the same results which were +/- the same as Noctua NH-U12S/U12S chromax.black. I'm not sure why in this review, some test results are so bad on these Enermax coolers. I was testing it on a different platform so hard to compare but I'm just saying that something isn't right here.
on the Intel platfrom they were within a couple degrees of each other so not sure what your implying? The NH-U12S barely survived the OC tests in most situations. On the AMD side, with the various chiplets means the IHS varies a bit more, that variation can influence contact between the IHS and the heatpipes, While I did get good contact and overall coverage, It was not nearly to the same level as say an NH-U12S.
So while the cooler handles the stock CPUS just fine, the other aspect to keep in mind is the TDP of both processors under load, the Intel chip is 200-watts in the OC test that is not even the maximum this chip can do under stock turbo situations, and the AMD chip is pretty much at 150-watts. Most 4 heatpipe coolers are going to struggle here. As such the stock results on both chips would be comparable to running a lower tier CPU and the OC tests separate the good from the best.
Is the enermax a bad cooler? no it does the job its made for but its not a high end cooler. With rising costs and inflation this would normally in my own estimation be a 35-40 cooler at $50 the competition is just far harsher. Still it is using a cheaper base design and is more focused on aesthetics. The design also shows it works better on an Intel IHS with better pressure, and better overall contact. Maybe its my sample but everything is intact, the fan works through the entire RPM range, the heatpipes have no issues and the fins arent bent corroded or otherwise compromised. So maybe it works better on the 5900 series than the 3900 series. But not all CPUs are the same and results will differ. For example the Scythe Fuma 2 is still a damn good cooler but on the older test bench it readily excelled on the newer higher core count CPUs performance fell way off. Thats just how it is.
I am also testing everything 3 times in every test, with multiple mounts to verify tim spread etc. and in a controlled environment. Every test and mount gave results within a 1C margin of error. so the reported result is 55 i may see 54 and 56 but typically I dont see any change in peak readings (just an example). When mounting mechanisms are problematic or the IHS is real bad (Corsair A500) the swing can be brutal.)
