FrostyTech best CPU coolers : graphed temp vs noise (8-2014)

[halcyon]

Trying to make sense of the FrostyTech cooler rankings, I did the following plot today.

Perhaps it is of use to someone. My apologies for the bad graphing. Quick & Dirty Scatter plotting in Excel doesn't make pretty graphs.



To me the graph points out the general trend as well as the sweet spot areas, which I've handmarked (the first group highest on the Noise chart on the left, i.e. lowest temp rise, is a group of water coolers with external heatsinks/fans).

[quest_for_silence]

To me the graph points out the general trend as well as the sweet spot areas, which I've handmarked

Just a side note: I've always had more than a doubt about FT noise findings.
With reference to that, just one remark: I know the Eisberg 240 and IMO it's one of the worst offenders among AIO coolers, so it's a bit surprising to see that inefficient and overpriced AIO either besting or tieing lots of coolers, and among them its somehow more efficient CM Seidon siblings, but even the Noctua NH-D15 and the Scythe Ashura...
Now I'm not saying that those coolers are not capable to perform even loudly, but IMO those data somehow point out that FT methodology isn't probably appropriate for SPCR reader needs.

[halcyon]

I completely agree on the methodology and of course dB(A) != sone.

However, this can give a rough ballpark figure on the cooling efficiency of the heatsink with the stock fans (at certain airlow).

Now, if the results are +/-6dB(A) then it very very likely is louder/quieter. If it's +/-10dB(A) then the difference will be very likely to be clearly audible by even ear.

Further, if one keeps the same airflow (m3/min, CFM) and just finds better / quieter fans, then it is probably possible to tweak the sonics (esp. the frequency spectrum) one way or the other, without affecting the cooling significantly.

So, I agree and this should be used as a rough eye-balling measure to find clusters of potential candidates.

I've been fairly surprised by the Be Quiet Dark Rock 3 models. The last time I really spent time on coolers/PSUs (in 2009!) they were a crap brand and their efforts were quite half-baked. Looks like they've improved.

Of course, I'll still try and compare them against the best of Prolimatech, Scythe and Noctua coolers.

I was also considering trying the NoFan series (but with a very low spinning, large fan attached to the top of it), but it seems that NoFan CR-series has no LGA2011 mountings.

[quest_for_silence]

However, this can give a rough ballpark figure on the cooling efficiency of the heatsink with the stock fans (at certain airlow).

Yes, at certain airflow: at a different airflow (the one which matters for SPCR readers, for example) it can be different, so that prediction (the rough ballpark figure) cannot be used, or even better, that's what I meant.

Further, if one keeps the same airflow (m3/min, CFM) and just finds better / quieter fans, then it is probably possible to tweak the sonics (esp. the frequency spectrum) one way or the other, without affecting the cooling significantly.

Quieter fans have a different airflow (likely inferior), so that prediction about cooling prowess doesn't necessarily apply: with a better sounding one (bearing noise and the likes), YMMV.

Of course, I'll still try and compare them against the best of Prolimatech, Scythe and Noctua coolers.

The proof of the pudding is in the eating.

I was also considering trying the NoFan series (but with a very low spinning, large fan attached to the top of it), but it seems that NoFan CR-series has no LGA2011 mountings.

AFAIK the NoFan is not suited to cool any LGA 2011 CPU, so I guess that's why they don't supply such mountings.
You might give a look to Zalman FX-coolers also (I dunno whether they provide 2011 compatibility).