Review: Noctua NH-U9, U12S, U14S TR4-SP3 Threadripper coolers

Published by

Click here to post a comment for Review: Noctua NH-U9, U12S, U14S TR4-SP3 Threadripper coolers on our message forum
data/avatar/default/avatar36.webp
Nice review! Page 9 both images have title of idle when the second should say load
data/avatar/default/avatar04.webp
Nice review, but it is missing Noctua U14S two fan push-pull test. My guess is that TR OC@4.0Ghz temps would be under 80 Degrees C. Hilbert, is there any chance to add U14S push-pull performance results to review? Tnx.
data/avatar/default/avatar22.webp
DaSick:

Nice review, but it is missing Noctua U14S two fan push-pull test. My guess is that TR OC@4.0Ghz temps would be under 80 Degrees C. Hilbert, is there any chance to add U14S push-pull performance results to review? Tnx.
I was thinking the same thing, DaSick and Hilbert. I'm not sure about below 80 degrees, but they'd be close to that.
https://forums.guru3d.com/data/avatars/m/245/245459.jpg
While these threadripper Noctua coolers aren't bad, what I don't understand is why they didn't just take the NH-D14 or NH-D15 CPU cooler & modify the contact plate for threadripper - those 2 coolers are a lot beefier than the ones in this article, and being twin tower are also more efficient. Considering threadripper can chew through a significant amount of Watts I really don't know why they didn't adapt the monstrous NH-D14/NH-D15!
https://forums.guru3d.com/data/avatars/m/262/262208.jpg
Robbo9999:

While these threadripper Noctua coolers aren't bad, what I don't understand is why they didn't just take the NH-D14 or NH-D15 CPU cooler & modify the contact plate for threadripper - those 2 coolers are a lot beefier than the ones in this article, and being twin tower are also more efficient. Considering threadripper can chew through a significant amount of Watts I really don't know why they didn't adapt the monstrous NH-D14/NH-D15!
I would agree there with adapting NH-D15 or NH-D14 for TR socket,maybe they'll release these coolers for TR but if its not needed, I would assume they're done few in house tests with NH-D15 or D14 I have run NH-D15 on 5820k with 4.5Ghz or 4.6Ghz and no issues,then tried NH-D15 on my 5960x with 4.5Ghz or 4.6Ghz OC and still temperatures are very good and very similar to the my custom water loop, temperatures are higher by 4-5°C on PKG in OCCT than on my custom water loop Hope this helps Thanks, Jura
data/avatar/default/avatar18.webp
Above the NH-U9, the Threadripper 1950X with its 16 cores @ 4.0 GHz / 1.38v.
Why do I see 1.440v core voltage on CPU-Z then?
https://forums.guru3d.com/data/avatars/m/245/245459.jpg
linoliveira:

Why do I see 1.440v core voltage on CPU-Z then?
Oh yeah, just checked what you were referring to - yep CPUz stating 1.44V instead of the 1.38V quoted in the article text & graphs. I wonder if the other coolers in the comparison graphs were tested at 1.44V too, because otherwise it would not be a fair test if the CPU was at different voltages for different coolers. Hopefully Hilbert can comment on your observation.