Question Temps in mid-80's on Prime95 a cause for concern with 10600K?

Jul 21, 2021
3
0
10
0
Prime95 v30.3.1
12 threads
Small FFTs
Disable AVX2, Disable AVX

i5-10600K with the Reeven Justice:
https://www.tweaktown.com/reviews/7169/reeven-justice-rc-1204-cpu-cooler-review/index.html

Motherboard is Z590 Gigabyte Gaming X. Case is mesh with 4xfans (2 front, 1 rear, 1 top/rear).

Temps are peaking at 86C while usually holding around 84C across most of the package after 15min. This is unexpectedly high for this cooler, isn't it? Is it unacceptably high? Should I order some thermal paste for an attempt to re-seat?
 
While that is quite high, it's nothing to worry about. Unless you get to around 92+ degrees in a heavy load you shouldn't really be in any concern.
It would probably help reseating the cooler and reapplying thermal paste, but is not that necessary.
 
Jul 21, 2021
3
0
10
0
While that is quite high, it's nothing to worry about. Unless you get to around 92+ degrees in a heavy load you shouldn't really be in any concern.
It would probably help reseating the cooler and reapplying thermal paste, but is not that necessary.
It seemed high based on what I'd seen reported anecdotally across the web for this CPU with similar air coolers running at stock frequency. In fact, some users have claimed peak temps as many as 20 degrees lower on coolers like the Dark Rock Pro 3 or NH-U12. This bothered me, so I went out and grabbed some paste, returned, and reinstalled the cooler.

Identical. Package peaked at 86C with the rest of the cores peaking from 80-85C. I could hear the fans kick up, but just to eliminate this as a potential issue, I manually calibrated them all to 100%, including the CPU fan, but it made no difference. I even tried swapping in a higher end Noctua fan. It actually did worse. Peaked at 89C. Never in any of these tests did the CPU voltage exceed 1.310V.

Gotta admit. This is a bit depressing.
 
It seemed high based on what I'd seen reported anecdotally across the web for this CPU with similar air coolers running at stock frequency. In fact, some users have claimed peak temps as many as 20 degrees lower on coolers like the Dark Rock Pro 3 or NH-U12. This bothered me, so I went out and grabbed some paste, returned, and reinstalled the cooler.

Identical. Package peaked at 86C with the rest of the cores peaking from 80-85C. I could hear the fans kick up, but just to eliminate this as a potential issue, I manually calibrated them all to 100%, including the CPU fan, but it made no difference. I even tried swapping in a higher end Noctua fan. It actually did worse. Peaked at 89C. Never in any of these tests did the CPU voltage exceed 1.310V.

Gotta admit. This is a bit depressing.
I can tell you 1 thing, that cooler is worse than the bequiet and noctua ones. you can't compare coolers out of size.
Those 2 use my higher quality heatpipes, they are much flatter on the bottom and are generally just much better air coolers.
 
Jul 21, 2021
3
0
10
0
I can tell you 1 thing, that cooler is worse than the bequiet and noctua ones. you can't compare coolers out of size.
Those 2 use my higher quality heatpipes, they are much flatter on the bottom and are generally just much better air coolers.
How have you assessed they use higher quality heatpipes? I wouldn't doubt it, but I've seen nothing from reviewers evaluating this. Is there metallurgical analysis to show the Reeven uses a less conductive copper, or something like that? Is there some other explanation?

Because as best as I have been able to divine it, drawn from correlation to hundreds of reviews over the years, per the heatsink, size certainly does matter. You have total surface area (dimensions + fin count); air gap height for better airflow through the cooler; heat pipe count + diameter; contact of heat pipes to the CPU as well as the consideration of their proper spacing for dispersion through the heatsink; metal type & treatment; finally, the total weight. I'm only discussing the heatsink, here. This is separate from the fan itself.

I realize it isn't direct contact, but that isn't the end of the world, and typically makes only a small difference in temps. You can see from reviews how well the Justice performs.

It trailed the mighty Noctua NH-D15 by only 1.33C in the Overclocked Load AIDA64 test on Tweaktown I linked above. This margin was roughly repeated, this time 1.75C hotter, in the review by the now-defunct PC Gameware UK. It bested the Noctua NH-D14 by 2C, the NH-U14 by 3C, the NH-U14S by 4C, and the NH-U12P by 8C in the Overclocked Load Prime95 test in the OC Club review. Contradicting this, it lost to the NH-D14 by 0.6C in the 150W test on Frostytech, but still beat the NH-U14S by 0.3C. It lost to the Noctua NH-U12S by 1C in the overclocked test on Enostech, but beat it by 1C in the Bjorn3D review which appears to have used the original uncoated heatpipe version of the cooler. It bested the be quiet! Dark Rock Pro 3 by 2C in the Eteknix review, and again by 1C at stock frequency in the Hardwareslave review.

I cite all of this to demonstrate it is a peer to the finest air coolers on the market in terms of peak thermal performance. So I'm just a bit puzzled what the issue could be. Even losing the silicon lottery doesn't explain these discrepancies.
 

ASK THE COMMUNITY

TRENDING THREADS