Noctua NH-D15 Intel and AMD CPU Air Cooler Review @ [H]

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,533
Noctua NH-D15 Intel and AMD CPU Air Cooler Review - The Noctua NH-D15 is firmly footed in the excellent performance of the NH-D14 model. The NH-D15 is not a "new" air cooling technology but rather an evolution of the old. The new claim to fame with this heatsink unit is that it has an expanded heatpipe layout, plus some other bells and whistles that Noctua hopes make it king of the hill.
 
Thanks for the review. My D14 has been rock solid for nearly 4 years now. Incredible air cooler, especially because I paid $70 for it new.

The D15 slightly improves so hopefully I can catch a sale and bring it down to $80.
 
What this shows is for air cooled the NHU14S just gets better as prices have drop considerable , 55 to 66$ depending where you look .
lighter , more ram compatible .
 
Completey agree pricing makes this not even a bronze award. WIll be interesting if they adjust there prices as way better alternatives out there.
 
One thing I wonder on D15 , does that front fan really need to be 140mm, is it much better than the old way of 140/120mm fan option .
reason why I say that, first time i looked at new D15 that 140mm is not holding on much of HS with that cut-out for ram clearance, so since it covers all ram slots to me there not much adjusting up and down with a 140mm .
 
Would it be possible to review these coolers on FX series processors? I don't think I have ever seen an FX-8350 in a heatsink review, with the Intel proc's they are meant to run at the threshold so you really don't see any major differences, it's most about how far you are able to OC, that depends on many other factor's though. With the AMD proc's, it really is about keeping it under the 60c mark, as there are a lot of us running these budget chips, wouldn't it be nice to see how the cooler performs without relying on forum posts from other users? You run into too many variables; ambient temps, paste, application, BIOS settings etc. Thanks!
 
It looks to have the same almost no clearance with the first PCIe slot that the DH-14 has.

I actually put electrical tape on the fan clips because I had about 1/32" clearance between the clip and my video card. One bump without it being insulated and you are just asking to fry something.

Also, you should do a review of this cooler with a 4930k or 4960x. I am curious if it is up to the task of cooling an overclocked hex-core.
 
Kyle, I wanted to expand on a point that you raised in the review and that EdKiefer touched on above.

The RAM clearance issues with the NH-D14 were well documented. And as was stated in this (NH-D15) review, this was addressed in the design of this new cooler. I'm glad you covered the height of the second fan in relation to clearing the memory and yet still managing to clear the side (or top) cover of the case. This could be a larger issue than most people may think depending on the memory used and which slots are populated. I've seen it written that the height of the cooler could easily exceed 185mm with the second fan installed over the memory. Which in turn could necessitate the need to relocate that second fan to the I/O side of the cooler (possibly leading to other clearance issues with an exhaust fan) or actually leading the new owner to consider replacing (with say a 120mm or even a 92mm unit) or possibly eliminating the second fan from the setup altogether. Thus degrading the performance and defeating the purpose of the acquisition in the first place (the dreaded "Why didn't I just buy the NH-D14S?"syndrome OY!).

I've seen some recent reviews where this topic was completely ignored. Kudos to you for at least bringing up the possibility of case clearance issues.

As always, "your milage may vary" (depending on your case, mobo and memory).
 
Would it be possible to review these coolers on FX series processors? I don't think I have ever seen an FX-8350 in a heatsink review, with the Intel proc's they are meant to run at the threshold so you really don't see any major differences, it's most about how far you are able to OC, that depends on many other factor's though. With the AMD proc's, it really is about keeping it under the 60c mark, as there are a lot of us running these budget chips, wouldn't it be nice to see how the cooler performs without relying on forum posts from other users? You run into too many variables; ambient temps, paste, application, BIOS settings etc. Thanks!

I would like to second this request. I don't understand why AMD was left out of the review in the first place.
 
A nitpick: on page 4 you have references to silence:

It met our expectations when it comes to silent operation. The fans are tuned beautifully for performance and silence ...

Should that not be 'quiet' instead of 'silent' and 'quietness' instead of 'silence'?
 
I cannot hear my D14 so while its not technically "silent", I cannot audibly hear it so for me it would be.

Put this up against the H100i and you will agree that it is silent :)
 
Would it be possible to review these coolers on FX series processors? I don't think I have ever seen an FX-8350 in a heatsink review, with the Intel proc's they are meant to run at the threshold so you really don't see any major differences, it's most about how far you are able to OC, that depends on many other factor's though. With the AMD proc's, it really is about keeping it under the 60c mark, as there are a lot of us running these budget chips, wouldn't it be nice to see how the cooler performs without relying on forum posts from other users? You run into too many variables; ambient temps, paste, application, BIOS settings etc. Thanks!
A heatsink review is all about performance relative to other heatsinks using the same hardware. That's how you gauge its performance. Changing the processor underneath the heatsink is not going to change how it performs compared to other heatsinks. The temps may be different, but which heatsink performs better than the others should not change. You are basically asking them to duplicate their work for no real gain in knowledge.

This is the same reason why they don't review heatsinks on other platforms that these heatsinks can use (e.g. lga 2011, lga 1366, lga 1156, lga 1155, lga 1150). Many heatsinks out there can mount to all of these processors and the processors themselves are physically different sizes and heat outputs (especially true of lga 2011 and lga 1366) but you don't see [H] reviewing them on those platforms either.

Rest assured if your OEM heatsink can keep your AMD chip under the recommended temps, then so can this beast.
 
It looks to have the same almost no clearance with the first PCIe slot that the DH-14 has.

I actually put electrical tape on the fan clips because I had about 1/32" clearance between the clip and my video card. One bump without it being insulated and you are just asking to fry something.

Also, you should do a review of this cooler with a 4930k or 4960x. I am curious if it is up to the task of cooling an overclocked hex-core.

The NH-D15 is the same width as the NH-U14S, both of which are actually 10mm more than the NH-D14. The NH-U14S is noted to interfere with the first PCI-E slot on motherboards so the NH-D15 should be treated with the same concern. This was actually brought up in the [H] review of the NH-U14S. The photos in the review are of an Asus mATX board that has it's PCI-E slot sitting one down from the usual position, so if you are using a board that doesn't do this, or an mITX board in general, be aware of this.
 
The NH-D15 is the same width as the NH-U14S, both of which are actually 10mm more than the NH-D14. The NH-U14S is noted to interfere with the first PCI-E slot on motherboards so the NH-D15 should be treated with the same concern. This was actually brought up in the [H] review of the NH-U14S. The photos in the review are of an Asus mATX board that has it's PCI-E slot sitting one down from the usual position, so if you are using a board that doesn't do this, or an mITX board in general, be aware of this.

most of the bigger HS that use 140mm will be very close to first PCI-e slot if your MB uses the first top for vid.
Even though the D14 is little bit less wide as it doesn't use the 150mm wide 140mm fan you could still run into issues there ,especially if vid card as bigger backing plate IMO .

Luckly most MB don't use first top slot .
 
Mugen 4 still wins price/performance. 212 Evo would be nice to see on the list too. I love my Mugen!
 
Completey agree pricing makes this not even a bronze award. WIll be interesting if they adjust there prices as way better alternatives out there.

I never really understood the criteria for these awards... they say there is better, for cheaper, then give it an award anyways :D
 
At the $100.00 price point I'd rather look toward water cooling solutions.

It's fully looped W/Cing or go home. AIO are a waste of time. They fail and become a heat trap, air coolers on the other don't as they still act as a "heatsink" if a fan fails. AIO also go loud under stressed load where a D14 for example is still pretty much inaudible. Not to mention the great cooling performance good air coolers offer for the price, better then the AIO's currently out there for the price. The only advantage of an AIO is smaller footprint and no thermal heat buildup's with the circulating airflow inside your case.
 
The thing to consider is not all people want water cooling in their system. So there might be water coolers out there that perform better for the same price but lots of people are simply afraid of water. This cooler is still a good choice for them at the high end.
 
The NH-D15 gives great RAM clearance, but is problematic in it's extreme width and front fan size. If the base was offset to the left like Thermalright is doing on their big coolers the PCIe slot clearance issue would be mute. But 140mm front fan on a fin pack that is 96mm is rather strange. The good news is using a single fan give good performance. ;)
 
Several of the "new" coolers released in the last year are more cosmetic changes than cooling improvement.

Thermalright's offset bases offer much better PCIe and/or RAM socket clearance.
Silver Arrow IB-E & Silver Arrow IB-E Extreme
Archon IB-E X2
TRUE Spirit 140 BW rev.A,
TRUE Spirit 120i
HR-22

Macho rev.B / Macho Zero to a lessor degree with a new base is a great and quiet cooler most know little or nothing about.

TRUE Spirit 140 Power with 8mm pipes instead of normal 6mm pipes is a fantastic performance value.​

New companies like
*Cryorig introduced significant changes is design and improvements with their R1 Ultimate & R1 Universal &andH5 Universal.

Raijintek are mostly copies of other coolers.

ID-Cooling's vapor hybrid vapor chamber & heat pipe coolers look interesting, but have yet to reach the western market.

Captherm's development of MP-2230 looks promising.​
 
So is this highest preforming air cooler then? I thought new models were coming out from some European company with 3 sets of fins, or at least some other new model since 2011. This is really hardly any better than the D14 and the alponfohn were.
 
I have tested and compared NH-D15, Cryorig R1 Ultimate Silver Arrow IB-E and PH-TC14PE. All perform about the same with fans at same rpm. NH-D15 has 1500rpm fans compared to 1300 for the others.

At 1500rpm NH-D15 is 1-3c better, slow the fans to match the others and all are +/-1c.

Don't have an NH-D14 to test, but know a guy who tested NH-14 & TRUE Spirit 140 Power at 1200 and full speed with TY-143 2500rpm fans. with 8c lower temps at full speed.

I'm planning to test several of the top coolers using TY-143 as reference fan at idle, 800, 1000, 1200, 1300, 1500, 1800, 2100rpm and full speed to figure out how different airflow levels effect their cooling performance. I use the air temp of cooler intake, not room ambient as baseline .. They are not always the same temp, even when testing in an open bench test system .. and definitely different when testing in a case. ;)

The biggest difference is at higher speed because most cases cannot remove all the cfm of heated air coming out of cooler .. and it mixes with case air / cooler intake air and raised intake air temp.

Run NH-D15 at 1300rpm and it's performance is similar.
 
It was the Alpenföhn Everest I was thinking of, was prototype news in 2012, haven't heard anything since. Did they scrap it?
 
Ah yes! Everest was big, but would still fit into standard cases.
EKL-Alpenfoehn-CPU-Cooler-Update-04.jpg


Not as big as AF41 Cryorig announced the first of April last year. Would have to cut out back of case and let it stick out. :D
04_top_sh_af41.jpg
 
Anyone know why they scrapped this? I'm guessing the performance wasn't working as well they were hoping, manufacturing cost was hard it to find a place in the market. Kind of a shame though, would interested to see how 3+ set would do. I'm thinking at this point though you're reaching the limit of what this type of design has to offer, can't get enough cool air into the final set of fins. I think I finally see now why the did the heat pipes in that seemingly inefficient distribution, just came down to manufacturing and minimum curvature needed on heat being heat pipes.
 
The Cryorig AF41? The cooler's model number and date of news release are a giveaway. :D
As for the Everest, it probably didn't improve performance and obviously blocks RAM.
 
I chickened out on running an H100i, (fear of leakage), and needed something that would clear some Dominator Platinum's.

It seems to work well, but if I select any of the Q-FAN profiles in the bios or Fan Xpert, the fans cycle on and off unless there is a significant load. The fans are on CPU_FAN and CPU_OPT. I can create a manual profile to maintain a minimum 30-40% duty cycle, so the fans keep turning.

It's actually pretty neat to watch it run passively when sitting at the desktop, I just don't know if the fans are really designed to spool up and down repeatedly. I've put the question to Noctua, but haven't received an answer. Nor have I gotten a response on the ROG forum about the bios part of the equation, (version 4901.)

Prior to the H100i I was running a Silver Arrow with PWN 140's and didn't see this behavior. However, that was a much earlier bios revision, (and much lower memory.)
 
@deees: I've heard that it can block the first pci-e slot on the r5e which means it's useless to me. It's hard to get definitive info on this though, last time I checked Noctua it read "may" have compatibility problems. I've seen a couple of posts that mention this blockage, but you don't. Unfortunately the x99 boards seem very tight in this area. I'd like to avoid water myself, my Noctua i4 single tower server cooler works well at stock, hoping it will do good enough to get me a tad over 4.

Oh and my i4 fans run pretty steady at idle and never shut down completely, what temps happen when they shut down? If the proc is underclocked at the time it makes sense passive would work on a big sink.
 
Back
Top