Noctua not supporting TR4 socket with existing coolers.

[doyll]

What Rroff said.

The review he linked rto found a 10-11c difference between normal cooler base and TR4 base on NH-U14S coolers.

To me that is a huge difference in load temps.

 

[Steampunk]

OP should probably change the thread title, as it makes it sound like Noctua are not supporting TR4. In fact, they are actually supporting it properly (with their new products) rather than palming customers off with a bodge based on their old products that don't fit the new socket correctly.

 

[doyll]

OP should probably change the thread title, as it makes it sound like Noctua are not supporting TR4. In fact, they are actually supporting it properly (with their new products) rather than palming customers off with a bodge based on their old products that don't fit the new socket correctly.

I agree he should .. or a moderator should. Noctua is one of a very few that have Threadripper compatible coolers out. ..

And I don't think the Asetek are any more compatible than a stock based air cooler .. even though AMD is supplying their mount. Their base is 53mm in diameter so no bigger than stock NH-U14S base, and under stock NH-U14S Threadripper runs 10-11c hotter than under NH-U14S TR4.


Here is the test results Rroff link refers to:

Steve Burke on GamerNexus did a test comparison of normal NH-U14S base and NH-U14S TR4 base coverage and cooling on Threadripper Gamers

Testing was fairly straight-forward in this one – we’re not reviewing the cooler today, just looking at coldplate A/B performance.

• TR 1950X CPU fixed to 34x multiplier, 1.243 Vcore fixed voltage
• ASUS Zenith Extreme X399 motherboard
• 32GB GSkill Trident Z RGB 3200MHz
• GTX 1080 Ti FTW3
• Thermaltake Toughpower 1200W
• Cooler: This is what we’re testing

He even used the same 140mm fan (NH-U14S TR4) on each of the coolers instead of fan included on each cooler to help eliminate variables, and used three primary tools for CPU load, ultimately eliminating one of them:

• Prime95 28.5 LFFT (eliminated: The LGA115X cooler couldn’t handle the thermal load, and throttled to a point where differences became immeasurable)
• Blender 8K render until steady state is achieved and maintained
• Handbrake 4K transcode until steady state is achieved and maintained

We take measurements at steady state under each workload. Ambient is measured second-to-second during each test, aligned with thermal data from the cooler passes, and then used to offset the output as necessary.

Threadripper’s 1950X throttles at 68C Tdie, or 95C TCTL and the stayed below that temperature.

While not a full review it is very interesting how much difference full coverage makes compared to partial coverage.

Full article can be read at
http://www.gamersnexus.net/guides/3029-quick-ab-test-impact-tr4-coldplate-size-with-noctua

And here is image Steve used showing Asetek contact area on Threadripper.

 

[Rroff]

Doing a bit more digging it seems one of the reasons they aren't just providing kits is that it shifts a lot of the weight/pressure of the cooler (especially if overtightened) away from the designed load bearing layout of the IHS with a higher potential for causing damage.

 

[doyll]

Doing a bit more digging it seems one of the reasons they aren't just providing kits is that it shifts a lot of the weight/pressure of the cooler (especially if overtightened) away from the designed load bearing layout of the IHS with a higher potential for causing damage.

While that makes perfect sense I don't think that is a serious concern .. at least Asetek and AMD don't seem to think it is.

But I'm not putting any faith in either of them after AMD started supplying Asstek mount for a thin 53mm diameter base having a tiny microfin area.