Noctua NH-D15S CPU Cooler Review
We’ve seen numerous CPU Coolers from Austrian giant Noctua here at pcG and over the months that have passed I have found myself appreciating their products more and more. But of course, as some of you know, I’m not a fan of the Brown colour scheme, although I may be warming to it a little! But, to stop me prattling on about it during the review I’m going to mention it once here: (a) I don’t like the colour brown! (b) Brown doesn’t fit with any colour themed Rig build I can think of! Ok, so now I have got that of my chest, let’s move on, and I promise not to mention it again, honest… 😉
This time around my review of this particular cooler is likely to be the biggest of them all, not the review that is, but the cooler itself, as the Noctua NH-D15S is one BIG CPU Cooler, in fact I think it’s the biggest I have ever installed! The cooler itself measures in at a massive 165mm(H) x 150mm(W) x 135mm(D), weighs in at 980g (with fan) and is equipped with a single 140mm NF-A15 PWM controlled fan. This 1500RPM fan sits between two aluminium cooling stacks at the base of which we find Noctua’s SecuFirm2 mounting system. This system supports the following sockets: Intel LGA2011-0 & LGA2011-3 (Square ILM), LGA1156, LGA1155, LGA1151, LGA1150 & AMD AM2, AM2+, AM3, AM3+, FM1, FM2, FM2+ (backplate required).
The Noctua NH-D15S arrived at pcG in the usual Noctua Brown/White box, a large box too, suggesting a pretty large cooler is hiding within. On the front we can see (courtesy of the silver sticker) that the NH-D15S is a High-compatibility version of the original NH-D15. In addition to this Noctua has chosen to highlight the following features:
The top of the box shows very little other than an image of the top of the cooler and its associated heatpipes.
The back of the box goes on to further explain the highlighted features on the front of the box. In addition to this the back of the box also promotes the included NH-H1 thermal compound and the impressive 6 year warranty.
Looking at the left side of the box we see the main product blurb in various languages. On the right side of the box we find the Noctua NH-D15S’s specifications and features (see Specifications/Features below for more detail).
Opening the box we can see that the Noctua NH-D15S is extremely well packaged as we have come to expect from Noctua. The two boxes are also well presented with their individual markings and images, note how the NOCTUA NH-D15S name lines up on both boxes (now that’s what I call attention to detail). The two boxes within are also both further protected with soft cell foam padding.
Within the main box there are two further boxes; one contains the Noctua NH-D15S itself, while the other contains all of the accessories/fittings etc. Note how the contents of the box are actually displayed on the top of the box, clever stuff and further signs of Noctua’s attention to detail and their Quality levels.
At the time of writing, the Noctua NH-D15S is retailing on Amazon for approximately £80 and comes with an impressive 6 year warranty.
courtesy of Noctua
|Socket compatibility||Intel LGA2011-0 & LGA2011-3 (Square ILM), LGA1156, LGA1155, LGA1151, LGA1150 & AMD AM2, AM2+, AM3, AM3+, FM1, FM2, FM2+ (backplate required)|
|Height (without fan)||160 mm|
|Width (without fan)||150 mm|
|Depth (without fan)||135 mm|
|Height (with fan)||165 mm|
|Width (with fan)||150 mm|
|Depth (with fan)||135 mm|
|Weight (without fan)||980 g|
|Weight (with fan)||1150 g|
|Material||Copper (base and heat-pipes), aluminium (cooling fins), soldered joints & nickel plating|
|Max. TDP||see TDP guide|
|Fan compatibility||140x150x25 (with 120mm mounting holes), 140x140x25 (with 120mm mounting holes), 120x120x25|
|Scope of Delivery||
|Model||1x Noctua NF-A15 PWM|
|Max. Rotational Speed (+/- 10%)||1500 RPM|
|Max. Rotational Speed with L.N.A. (+/- 10%)||1200 RPM|
|Min. Rotational Speed (PWM, +/-20%)||300 RPM|
|Max. Airflow||140,2 m³/h|
|Max. Airflow with L.N.A.||115,5 m³/h|
|Max. Acoustical Noise||24,6 dB(A)|
|Max. Acoustical Noise with L.N.A.||19,2 dB(A)|
|Input Power||1,56 W|
|Voltage Range||12 V|
|MTBF||> 150.000 h|
First impressions of the Noctua NH-D15S are wow, now that’s a damn big cooler! No seriously it is, if this thing doesn’t keep your CPU cool than I cant believe another CPU Cooler is going to do much better!? Maybe I shouldn’t have said that, as now I’ve set myself (and maybe the NH-D15S) up for a fall! Seriously though, not only is the NH-D15S a big cooler it’s a good looking cooler too, not so much down to its aesthetics (remember James you promised!) but down to its construction, as this is also one beautifully made cooler too…
Looking at the front of the cooler we can begin to appreciate the sheer size of the thing, and how much heatsink we actually have here. We can also now see the six copper (Nickel-Plated) heatpipes present on both sides of the cooler. The heatsink itself contain 45 (yes, I counted them (twice!)) aluminium fins, with the upper fins being larger than the lower ones.
Looking at the Noctua NH-D15S from the side (both sides are identical) gives us a better indication of its height, at 165mm with fan the NH-D15S is not going to fit in all Cases, so please be mindful of this. What you can also see from the image below right is the stepped design of the heatsink, this is to improve RAM clearance. We can also see that the 140mm fan is held in place by wire clips, something I’m not normally a fan of (haha, get it!), but these Noctua clips are easy to use and work well! Also note the anti-vibration mounts on the corners of the fan shroud in an attempt to keep vibration/noise to a minimum.
Looking at the heatsink from above we can see the tops of the twin heatsinks, with each side taking on the furrowed brow of the signature Noctua logo, the Owl. In the centre of each one we also find another Owl logo and the Noctua brand name. Here we also find the tops of the six Nickel-plated heatpipes.
Flipping the cooler on to its side allows us to take a look at the base of the Noctua NH-D15S and we get to see that copper/Nickel plated base plate in all of its glory and find out where all of those heatpipes go! We also get so see a little more of Noctua’s SecuFirm2 mounting system, as we can now see the threads that will screw down onto the studs on the mounting brackets.
Doing a comparison with our regular Test heatsink, the Raijintek Themis, we can begin to appreciate how much bigger the Noctua NH-D15S is, in all dimensions too… Wow!
As we have come to expect from Noctua the NH-D15S is one good looking, well made cooler. And, judging by the sheer size of the thing its cooling capacity should be pretty impressive also. Let’s get it installed and find out shall we…
|Case||Cooler Master HAF XB||Power Supply||Corsair Professional Series AX 760i|
|Motherboard||ASRock Fatal1ty Z97X Killer||CPU||Intel Core i5-4690K|
|CPU Cooler||Noctua NH-D15S||RAM||HyperX Savage 2400MHz 8GB Kit|
|Graphics Card||XFX AMD Radeon R9 290X DD Black Edition||SSD||HyperX FURY 120GB|
The first task before installing the Noctua NH-D15S was to locate the parts required for our Socket 1150 install. Thankfully Noctua has made this nice and easy courtesy of a bag labelled Intel! From this bag we will be using all of the parts apart from the studs that are used for an Intel Socket 2011 install. Below you can see all of the parts required for install, and thankfully thanks to Noctua’s SecuFirm2 mounting system, there’s not many parts either. In fact it’s worth noting that our existing test CPU Cooler (Raijintek Themis) uses 18 parts for assembly, the Noctua SecuFirm2 mounting system has just 11… Maybe it’s time for a change James!? 😉
To be honest the fitting of Noctua’s SecuFirm2 mounting system is so easy it hardly needs covering, but for the sake of completeness I will. The first task is to thread the backplate through the back of the motherboard as shown in the image below left (be careful of the backplate’s orientation). Next we add one plastic spacer to each of the uprights that are now poking through the motherboard, see image below centre. The final step is to add the two brackets and then simply secure with the four nuts and you’re good to go. BUT, hold on aren’t those brackets (left & right in the image) on the wrong side? Well actually no, the NH-D15S can be fitted either way round, but I opted to fit is so the central 140mm fan faced the rear of the case, therefore the exhaust fan of the case could further aide in cooling.
With the mounting system in place I added a small pea sized blob of Arctic Cooling MX-4 to the top of our test Intel Core i5-4690K. The cooler was then gently lowered into place with the threads on the cooler aligned with the stand-offs on the brackets. Each screw can now be tightened in turn (using the handy screwdriver supplied), don’t tighten one side first as this will likely tip the cooler to one side resulting in poor cooling performance. The great thing about Noctua’s SecuFirm2 mounting system is that the screws can be tensioned until they stop, no guess work here guys. Good eh! 🙂
The last step was to connect the 140mm Noctua NH-D15S PWM controlled fan to the CPU_FAN1 header on the ASRock Fatal1ty Z97X Killer motherboard. As you can see thanks to the stepped design of the NH-D15S’s heatsink there’s plenty of RAM clearance, even for those pesky taller modules.
I told you it was big! Now I have to admit that looks like a cooler that’s going to keep your CPU pretty cool, but hold on where’s the rest of my Rig gone!? 😉 Looking inside the case now, the Noctua NH-D15S is pretty much all you can see, it’s certainly the thing that your eyes are drawn too, and I’m not so sure that’s a good thing…
|For CPU Cooler testing, we here at pcGameware run Prime95 for a 15 minute period. During this period the temperature is monitored with CoreTemp and the cooling performance recorded (the max recorded is the average for all cores). Between each stress test we allow a 15 minute cool-down to allow for more accurate results. To adjust the fan speed we simply use the UEFI. A close eye is also kept on the ambient temperature, with the maximum being recorded for each run, this allows us to calculate the Delta temperature (Core – Ambient = Delta). Each run was performed with the Intel Core i5-4690K CPU at the following frequencies: 3.5GHz (Stock) and 4.0GHz (using the ASROCK OC Tweaker, shown in the picture below), all results have been recorded with CPU-Z.|
* Please note: To ascertain the maximum and minimum noise levels produced by our CPU test coolers. The dBA is recorded at a distance of 1 metre from the cooler, with all case fans unplugged to isolate the sound in question.
Processor speed is set using the OC Tweaker tab of the UEFI, “Disabled” for stock speeds (3.5GHz) and “Turbo 4.0GHz” for the minor overclock tests. It should be noted that changing the OC settings resets the Fan Speeds, so these are checked on the next reboot and reset to “Full Speed” (see below).
All the fans installed in the system are set to 100% speed using the displayed settings, this is simple with the ASROCK UEFI with the option of “Full Speed” being available for all system fans.
As stated above, to make our performance tests easier to follow and to get the most accurate recordings, all of the following tests have been carried out with case fans set at 100% and the Noctua’s NH-D15S fan also set at 100%.
- Intel Core i5-4690K – 3.5GHz (stock)
|CPU Cooler||Air/Liquid||Fan Speed||Ambient Temperature||Max CPU Temperature (core average)||Delta Temperature||Noise Level|
|NZXT Kraken X61||Liquid||100%||24.50||40.75||16.25||51dB|
|Deepcool Gamer Storm Maelstrom 240||Liquid||100%||24.00||43.50||19.50||50dB|
|Corsair H110i GT||Liquid||100%||24.50||44.00||19.50||57dB|
|Fractal Design Kelvin T12||Liquid||100%||25.00||46.00||21.00||40dB|
|SilverStone Tundra TD03-E||Liquid||100%||25.50||49.25||23.75||51dB|
|Raijintek Themis Evo||Air||100%||21.50||47.50||26.00||37dB|
|Scythe Mugen Max||Air||100%||22.00||48.50||26.50||36dB|
|Prolimatech Basic 81||Air||100%||22.00||50.75||28.75||43dB|
|SilverStone Argon AR06||Air||100%||23.50||79.50||56.00||28dB|
As you can see in the CPU Cooler chart the Noctua NH-D15S sits somewhat lower in the grid than one would have expected. Considering the sheer size of the cooler the 49.75 Core Average and its associated 26.75 Delta looks pretty poor. And to be fair it is, it can’t seem to outperform its little brother the Noctua NH-U12S. Something’s not right here, I wonder whether the cooler will perform better when we lean on it a little more?
- Intel Core i5-4690K – 4.0GHz (OC Tweaker)
|CPU Cooler||Air/Liquid||Fan Speed||Ambient Temperature||Max CPU Temperature (core average)||Delta Temperature||Noise Level|
|NZXT Kraken X61||Liquid||100%||24.00||48.00||24.00||51dB|
|Fractal Design Kelvin T12||Liquid||100%||26.50||53.75||27.25||40dB|
|Corsair H110i GT||Liquid||100%||24.00||51.75||27.75||57dB|
|Deepcool Gamer Storm Maelstrom 240||Liquid||100%||24.50||52.50||28.00||50dB|
|Scythe Mugen Max||Air||100%||22.00||54.75||32.75||36dB|
|SilverStone Tundra TD03-E||Liquid||100%||26.00||58.75||32.75||51dB|
|Raijintek Themis Evo||Air||100%||21.50||58.00||36.50||37dB|
|Prolimatech Basic 81||Air||100%||22.00||59.50||37.50||43dB|
|SilverStone Argon AR06||Air||100%||23.50||92.50||69.00||28dB|
Now I wasn’t expecting that, even when we lean on the NH-D15S a little more, courtesy of our 4.0GHz overclocked test, the cooler still finds itself bettered by its lesser sibling the NH-U12S!?
When I first started to record these results I guessed something was wrong, after all how could a cooler half its size perform just as well!? So I set about reinstalling the NH-D15S again, the odd thing was; I got the same result! After further testing I could not better my original result, so I decided to contact Noctua. Thanks to a swift reply from Noctua (suggesting a faulty cooler), I received an brand new NH-D15S only days later. But what I wasn’t prepared for was the surprise that – I got the same results again! Try as I might I could not get the NH-D15S to perform as good as I had hoped and I could not get it to better its lesser sibling the NH-U12S!?
Of course I contacted Noctua again saying that try as I might I couldn’t get the NH-D15S to perform in the way I would expect, and believe you me, I really did try. Alas the final word from me here at pcG and Noctua is that on this system setup (see above) this particular cooler doesn’t seem to want to show its logical potential, and I’m afraid we just don’t know why…
As one has come to expect from Noctua the NH-D15S performs well in the area of acoustics. With its single 140mm 1500RPM NF-A15 PWM fan at 100%, in our Test Rig, with all other fans unplugged or at idle, we recorded a noise level of 39 dBA which is pretty good. Of course thanks to its PWM control the noise/performance ratio can be manipulated within your motherboard’s UEFI, so you can get the perfect balance.
Despite its size we couldn’t seem to get the Noctua NH-D15S to perform as expected on our Test Rig, why we just don’t know!? Thankfully all of the normal attributes of a Noctua product are here though; great design, high build quality and quiet performance.
As we have come to expect from Noctua the NH-D15S arrived at pcG in a smart brown box with all of the components within beautifully packaged and presented. Once out of the box the enormity of the NH-D15S was apparent at 165mm(H) x 150mm(W) x 135mm(D), weighing in at 980g (with fan) this is one big CPU Cooler. What’s also apparent is the high level of build quality too, the NH-D15S is a good looking cooler and also beautifully made.
Installation was also a breeze thanks to Noctua’s SecuFirm2 mounting system. With only a handful of parts (11) this is one of the best mounting systems on the market today. Especially as the mounting system is used across pretty much all of Noctua’s coolers. With the cooler installed and the 140mm 1500RPM NF-A15 PWM controlled fan in situ it was time to see what this behemoth of a cooler could do, and it was here that we ran into a bit of an issue, well for pcG anyway…
As try as we might (and truly I tried!), we couldn’t seem to get the NH-D15S to perform as we would expect. In both of our tests the little brother of the NH-D15S, the NH-U12S, outperformed the bigger NH-D15S. We even contacted Noctua and got a replacement cooler but alas we witnessed the same performance of 49.75 (Delta 26.75) at 3.5GHz and 58.00 (Delta 35.00) at 4.0GHz. While these are not bad results per se, when compared to lesser sized coolers the results just don’t seem to add up! We have discussed this further with Noctua, whilst looking for answers, but it would seem that with our setup (see above) the NH-D15S just doesn’t perform as well as expected. Maybe with a different CPU like a Socket 2011 setup perhaps and a higher heat load the NH-D15S would come into its own…
As one would expect the 140mm 1500RPM NF-A15 PWM controlled fan produced very little noise, even with the fan at 100%. We measured around 39dBA with all case fans unplugged and the GPU at idle. Of course thanks to that PWM control the noise/performance equation can be tweaked to your liking within the motherboard’s UEFI.
Overall then, as expected the Noctua NH-D15S is a BIG, good looking and well made CPU Cooler, that’s also incredibly easy to install. But in our testing that BIG cooler didn’t seem to translate into BIG thermal performance gains. Why we just don’t know! Strange…
Please Share, Like & Comment below, we really value your thoughts and opinions…
Where possible we always use Amazon’s price for Value…
Many thanks to Noctua for providing this sample for review