Noctua's NH-D15 Versus Five High End Closed Loop Liquid Coolers

[4960X]

This review has the nh-d15 benchmarked against the h100i gtx, nepton 240, kraken x61, deep cool captain 240, thermaltake water 3 and it seems like the nh-d15 did pretty well. It's much quieter than all the closed loop coolers but loses against the kraken x61.
Anybody here uses an NH-D15? Are they worth it?


http://www.relaxedtech.com/reviews/noctua/nh-d15-versus-closed-loop-liquid-coolers/1

 

[PliotronX]

I've been impressed with my D15's temps and noise levels it's been able to maintain but the awkward hugeness and chinsey fan clip in close proximity led to contact with my GTX780 and subsequent early death of it (admittedly in my haste of reassembly, did not see the clip like that). I'd probably check out the eVGA ACX that the one dude posted about here or the Scythe Ninja 4 because they're not too far off in temps and offer much better value and ease of installation/less problems with fans & RAM.

 

[lehtv]

Nice find.

Then compare HR-02 Macho ($50) to Noctua D15 ($100) and you see how little sense there is in buying an AIO liquid cooler. Techpowerup reviews: D15, HR-02. D15 is 2 Celsius cooler but 3dBa louder at load.

 

[Kenmitch]

Nice results....But for me at least I don't feel comfortable with a 3lb bobble head hanging off my motherboard. I'm sticking with AIO's most likely.

 

[Deders]

Am using one. Works great. Was expecting it to be quieter at full speed from previous experience using a noctua fan in my last setup, but it doesn't matter much. with the fan profile setup properly.

I set it to start ramping up from 50% (silent) at 45c, reaching 90% at 55c and 100% at 60c (motherboard socket readings, not cores.) and I find if there is a full core in use I will start to hear it so it's handy as a signal for different loads. Not really intrusive whilst gaming at all.

It does get a bit louder at 100% but I'm quite happy for it to be a bit noisy with extreme loads.

I also didn't notice any difference with having the 2nd fan on, so I've used it as an exhaust. I expect it has a better chance of shifting more air than my previous fan.

 

[lehtv]

Nice results....But for me at least I don't feel comfortable with a 3lb bobble head hanging off my motherboard. I'm sticking with AIO's most likely.

The motherboard transfers the load to the case because it's secured in it. Though technically the cooler is hanging off the motherboard, almost all of the weight is on the case, so the situation is practically the same as with an AIO.

 

[BonzaiDuck]

This review has the nh-d15 benchmarked against the h100i gtx, nepton 240, kraken x61, deep cool captain 240, thermaltake water 3 and it seems like the nh-d15 did pretty well. It's much quieter than all the closed loop coolers but loses against the kraken x61.
Anybody here uses an NH-D15? Are they worth it?


http://www.relaxedtech.com/reviews/noctua/nh-d15-versus-closed-loop-liquid-coolers/1

The thermal results cited for the OC'd chip at 4.5Ghz are pretty much consistent with my own experience with the D14. The D15 may perform maybe 3C better than the D14, from the reviews I'd read offering either direct comparison between the two or an inference you can make from two or more reviews of similar reliability.

That being said, you can probably match the AiO/CLC performance for units like the Kraken, H100 and H110i by managing case intake airflow and using a duct to port the exhaust-side of the D15 to either the case rear or case top. For the usual overclocked TDP power range, you might expect between 4 and 5C improvement this way.

 

[aigomorla]

im not even going to start with how much of a fail that test is.

no listing of delta ambient?
no mention of where ambients were taken? Intake of sink? in the bathroom next door? in the middle of the room?
how do we even know if the author did the tests all on the same day while the ambients were the same?

if ambient aren't held constant, and no delta ambient are listed, or no mention of climent controlled environment was used, you can not hold the test with any grain of salt.

 

[Kenmitch]

The motherboard transfers the load to the case because it's secured in it. Though technically the cooler is hanging off the motherboard, almost all of the weight is on the case, so the situation is practically the same as with an AIO.

All the weight is on the desk by your description. The coolers weight is strictly on the motherboard could cpu socket and surrounding area. Bobblehead effect could cause damage depending on rigs enviroment.

 

[dlerious]

All the weight is on the desk by your description. The coolers weight is strictly on the motherboard could cpu socket and surrounding area. Bobblehead effect could cause damage depending on rigs enviroment.

You could use a case (like my Caselabs S8) that has horizontal mount. I use custom loop.
I'd like to see them include some other coolers like the EK Predator or Swiftech H240-X, heck even the Thermaltake Water 3.0 Ultimate I had would be nice.

Just checked and the Swiftech H240-X is $150 - only $10 more than the X61.

 

[lehtv]

All the weight is on the desk by your description.

True. That's the whole point of securing the motherboard and other components into the case. The case carries their weight (therefore applying force on the desk).

The coolers weight is strictly on the motherboard could cpu socket and surrounding area. Bobblehead effect could cause damage depending on rigs enviroment.

Yet you cannot deny the fact that having the motherboard secured in the case means the force caused by the cooler onto the socket is transferred to the case frame.

You'd have a point if the motherboard was secured only using the lowest screw holes - just enough to keep the motherboard vertical. Then a heavy cooler would literally hang off of the motherboard and cause damage.

I can't think of any real world scenario where this "bobblehead effect" (which I've never heard of, but I think I know what you mean) could pose a risk to the motherboard, outside of maybe shipping a whole computer.

 

[aigomorla]

that is why motherboards of enthusiast class or gamer class have a back retention plate.
This plate spreads the pressure out on a larger area to avoid the bobble head effect.

They learned this back in S775 era when people were mounting almost 3kg sinks onto Q series cpu's.

 

[4960X]

Guru3d has the nh-d15 performing the same as the original H110 and beating the H100i Max Mode by 3c. This just proves that AIOs are simply not worth it.

http://www.guru3d.com/articles_pages/noctua_nh_d15_review,11.html