They say that bigger isn't always better. For the most part I'd have to disagree. As men we know size matters and that bigger is very often better. Whether it be screen size, memory size, SSD capacity, Overclock, or for that matter even the size of our....Ahem. Yes well, what we're trying to say here is that we're pretty much hard wired into a perpetual game of top trumps with all the numbers in our life. Where size does matter though, and where the above rule starts to gain credence, is when size causes a problem, when it causes a conflict, where it gets in the way of other things, or where it's too large to fit. Imagine that, having one so big that it's too big to fit...Get your minds out of the gutter, I am of course talking about heatsinks here, and some being too large to fit in your case, fouling your manly large RAM, obscuring your first PCI slot, or stopping the door from shutting on your super slim case.
Enter then the Noctua NH-U12S at just 158mm tall and a mere 45mm deep without fans there's no way it's going to get in the way of your RAM, and unless you're into uber thin Mid towers you're going to be hard pushed to find a case that's too narrow for it.
The sample we have for review comes packaged with a pair of focussed airflow NF-F12 fans, which should make for some interesting temps. Time and testing will tell.
Height (without fan)
Width (without fan)
Depth (without fan)
Height (with fan)
Width (with fan)
Depth (with fan)
Weight (without fan)
Weight (with fan)
Copper (base and heat-pipes), aluminium (cooling fins), soldered joints & nickel plating
Scope of Delivery
NF-F12 PWM premium fan
Low-Noise Adaptor (L.N.A.)
NT-H1 high-grade thermal compound
SecuFirm2™ Mounting Kit
Anti-vibration pads and fan-clips for second NF-F12
Noctua Metal Case-Badge
Max. Rotational Speed (+/- 10%)
Max. Rotational Speed with L.N.A. (+/- 10%)
Min. Rotational Speed (PWM)
Max. Airflow with L.N.A.
Max. Acoustical Noise
Max. Acoustical Noise with L.N.A.
> 150.000 h
Up Close: Packaging and Contents
If you've never opened up a Noctua product you've missed a real treat. From the understated exterior giving details of specification and features to the plain packaging inside they simply ooze quality.
Don't let the plain brown boxes fool you. Even the card used here feels more like thin ply wood. There's a minimalistic style here that says not just "this is good" but also "we've not wasted your money on glitzy packaging". Noctua have even been thoughtful enough to separate AMD from Intel fittings, with a totally separate box for universal accessories, and we're not talking a few fan clips here, oh no, if you buy Noctua you get additional fan stand offs, additional fan clips, a simple screw driver, a speed reducer cable, Y splitter, Thermal Paste, cuddly toy, and even a case badge. Ok so you don't get the cuddly toy, and that reference might be a little dated but you get the idea.
Noctua also excel when it comes to the instructions. Although they consist only of an A4 fold out sheet, they are none-the-less clear and concise with good sized diagrams. Noctua also provide separate instructions in each box for AMD and INTEL so there's no getting mixed up or wondering which set of bolts or brackets you're meant to be using. The people that did the instructions for the Eisberg could learn a thing or two from this lot.
Up Close: The Cooler
Although later in the build section you can see the U12S with 2 fans attached, so that you can better see the fin structure and pipe distribution we've attached just a single NF-F12 to it for this part of the review. The NF-F12 has a focussed flow design and is specifically intended for use with heatsinks and radiators where a high static pressure is required. The PWM feature allows the user, via motherboard control to alter the speed between 1500RPM and 300RPM, although it has to be said even at full tatt the 22.4dB(A) emmited by the fan isn't exactly loud.
50 Aluminium fins with semi closed sides are utilised to conduct the heat away from the heat pipes and provide a large surface are for maximising heat dissipation.
The NH-U12S is symmetrical with grooves for fan clips on both sides. Additional fan clips come as standard within the accessories box,
The NH-U12 has five heat pipes spread fairly evenly within the fin stack although with a slight bias to the lateral aspects
As with all Noctua products there's real attention to detail and clear evidence of high quality standards. Whether it be the immaculate Nickel plating of the copper contact plate or the perfect soldering of joints
The Noctua NH-U12S utilises the same SecuFirm2 mounting system as the 14S we reviewed earlier, and as such is one of the easiest coolers to fit that we've come across. As Noctua distinguish Intel parts from AMD pats by putting them in separate boxes is very easy to isolate those parts you require. Whichever CPU you're touting the effect is much the same as they say. Having first threaded double ended bolts into the motherboard (backplates are supplied if they are required), brackets are then bolted down and screwed tight using the included screwdriver.
With the Brackets in place the cooler, minus the fan is placed over the CPU not forgetting to apply the included NT-H1 thermal compouind by whichever means you swear by. A cross bracket which is hard mounted to the heat sink is then screwed down in place. The mounting screws have springs to determine the correct tension for you
Noctua boast 100% RAM compatability, with the caveat that certain AMD boards breach the ATX exclusion zone around the CPU, and even then all you have to do is stick to RAM under 40mm
As the Cooler is just 125mm wide it also steers clear of covering up the first PCI slot, which as we know can sometimes be the the only 16x slot on a board. We can see below that even with two fans there's plenty of room around the coole
Fitted into our test case, the Cooler Master "Test Trooper" the diminutive NH-U12S does look a little at sea. Good job the trademark colour scheme helps it stand out. Although it might look a little lost here, at just 158mm in height it should fit into most ATX cases and a good few mATX to boot.
Performance and Testing
The test set up consisted of the following
Intel i7 3960X Stock@ 1.1v (undervolted) 4.0GHz @ 1.25v 4.4GHz @ 1.35v 4.6GHz @ 1.45v Gigabyte X79 UD3 Corsair Vengeance LP Memory Corsair HX850 V2 Corsair Force GT 60GB Coolermaster Storm Trooper.
As usual we'll be testing our coolers at varying levels of overclock and increasing levels of voltage. this in turn of course means increasing levels of heat which the coolers need to dissipate. To begin with we start with the undervolted stock speed. Why undervolted? well if you have things set on "Auto", you may well be using more volts than are actually required to run at the chosen frequency, for example our 3960s will run quite happily at just 1.1volts, solid as a rock, 24/7, and as such we use this as our starting point.
Continuity is very important in testing, and for this reason we keep as many of the potential variables as locked down as possible. We will be using OCCT in Linpack X64, AVX compatible with all logical cores tested and 90% free memory utilised. The test is set up to run automatically with just a few clicks to set it going. A 10 minute idle followed by 30 minutes of testing and a 5 minute cool down is the order of the day and brings the total test time per clock speed to 45 minutes. So as to remove subjectivity in determining whether a CPU has failed, OCCT is set to stop the test and register a fail should the max temp exceed 80 degrees. In testing we noted that if even just one of the cores exceeds 82 degrees OCCT halts the test and a fail is recorded.
As with the 14S we looked at recently we've tested the heat sink in single and dual fan set up. An LNA is supplied, which will, as the name suggests lower the noise. If you choose to use this then you can expect a 2-3 degree increase in temps. We felt though as the set up is so quiet anyway the LNA was largely superfluous and have not tested with it.
The very first thing that sticks out here is that in single fan mode the smaller U12S has actually posted cooler temps than it's bigger brother, the U14S. We're putting this down to the U12S having the NF-F12 focussed flow fans with a higher static pressure of 2.61mH20 as opposed to the NH-U14S's use of the NF-A15 fan with a much lower static pressure of 1.51mH2O. Both heatsinks are clearly able to conduct the little heat generated at this overclock equally well up the fin stack but it's the focussed flow fan that is able to push the heat out of the cooling fins. The second odd thing here is that the cooler actually performs slightly better in single fan set up than dual fan set up at this level of overclock. You'd think that the 1.2 degree discrepancy might be put down to inaccuracies in testing, and to be honest we had our doubts also, and so conducted each of the tests with single and dual fan set up 3 times. We also tried with and without the extra thick rubber spacer supplied for use with the second fan, (this we found to have no effect what-so-ever). The results were the same each time, the single fan set up was consistently slightly better than the dual fan set up.
Turning now to the 4GHz test we up the voltage to 1.25 volts, this is what is deemed normally as stock volts. Something we are always harping on about on the forums is AUTO does not mean stock volts, and normally if you overclocking with "auto" volts the motherboard will be upping the volts much more than needed if you were to do it manually. By whichever means it happens, upping the volts (especially from our 1.1v undervolt) does have a big impact on temps, with an average increase of 10-15 degrees seen in the results.
At the higher levels of heat generated by the increase in voltage required for the 4.0GHz overclock, fan performance, although still a factor becomes less critical, replaced instead by a coolers ability to conduct the heat up the heat pipes and more crucially the total surface area of the fins enabling convection to the atmosphere. The dual fan set up now performs slightly better than the single fan set up, but only by 1.3 degrees, with U12S now 3-5 degrees off the pace of the larger U14S. 3 degrees cooler than the Eisberg 120L on it's lowest fan setting which due to the pump noise was considerably noisier. The single fan configuration was also 4 degrees cooler than the Alpenfohn K2 on its low fan setting. When compared to coolers of it's own size and shape the U12S fares slightly better than the Matterhorn Pure, being some 5 degrees cooler in single fan set up, but not so well against the Megahallem Black which pipped it by 0.5 of a degree, again with both in single fan configuration.
That unfortunately is where the testing stopped, with the U12S unable to make the grade at the 4.4GHz and franky ridiculous 4.6GHz tests. We don't see this as a failing really, remember this is only a 45mm thick tower cooler. We've had AIOs fail at this level.
Unboxing a Noctua product is one of the truly great treats in the reviewers world. There's a real sense of quality and class from the outset, and although the inner boxing is brown card, even its thin sturdiness belies quality and simplicity rather than cheap and cheerful. Delve deeper and the individually boxes reveal that they are there to separate the fittings required for Intel and AMD CPUs. A third long thin box contains the accessories, an LNA lead, a Y splitter, an additional set of fan clips, a set of thicker rubber fan spacers, a tube of actual NT-H1 TIM, a screw driver and even a metal case badge. All that's missing is the cuddly toy.
Having cursed and sworn at many a CPU mounting system over the years, the SecuFirm2 mounting system coupled with clear and concise instructions make fitting the NH-U12S a simple, perhaps even joyful task, and with a 50 fin stack measuring just 45mm thick, even with a 25mm fan attached Noctua guarantee you will not have any RAM encroachment issues. Which is nice.
Producing just 22.4dB(A) of noise at is full 1500RPM the NF-F12 fan that comes with the U12S makes for a very quiet set up. An LNA is supplied which can be used to further reduce the noise to 18.6dB(A) but we found this to be wholly unnecessary as it's quiet enough as it is especially when you consider the 300RPM drop in fan speed will result in a 2-3 degree increase in your temps.
Just because we've already made our thoughts clear on the Noctua signature colour scheme in our last review doesn't mean we're going to slack off on them now. Yes, we know the colour belies the quality, but would you buy a Porsche if all the only colour you could get them in was poo brown and false arm beige? Come on Noctua give us some style as well as substance.
The performance of the U12S was a bit of a mixed bag, especially when you consider its beating the U14S at 3.2GHz and it having a lower single fan temp than dual fan temp at the same 3.2GHz overclock. We explain the reasons for this in detail in the performance and testing section, suffice to say that at low levels of overclock with low volts, the performance of a cooler is more to do with the fans abilty to remove what little heat there is from the fin stack than the fin stacks ability to conduct the heat away from the contact plate, and as the U12S is fitted with focussed flow High pressure NF-F12s as opposed o the standard NF-A15s on the U14S we think this explains the variances.
In comparative terms the U12S performed well against some bigger coolers, in particular the colossal K2 and the ear bleeding Eisberg (at low fan settings at least). Matched against coolers of its own size the U12S fared better than the Matterhorn Pure by a few degrees but was pipped by half a degree by the Megahalem Black. Let's remember though that the Megahalem black is priced identically to the U12S but doesn't come with even a single fan as standard, so in effect is in the region of £15-£20 more. The Matterhorn Pure however at just £30 is still a very good buy.
But of course it isn't only air coolers that the U12S has to compete with. With AIO water coolers getting cheaper and more user friendly its biggest competition comes from the likes of the Cooler Master Seidon 120 and similarly the Corsair H60. Cost wise the U12S stacks up quite well initially, but adding an additional NF-F12 fan will add an extra £20 to the cost of the set up, and will only reduce your temps by a degree or so, and as it does not open the door to higher overclocks we would have to say hand on heart that it's not worth the extra cash
To summarise, the UI2S is a good cooler, especially if you're looking for something that will keep things cool while not getting in the of comedy tall RAM. It's also one of the quietest coolers we've ever tested. The quality is there in spades, even if the looks aren't to everyone's taste, (and especially not ours, did we mention that?), still there's no denying that to the trained eye a look inside your case window is bound to raise an "oh, you've got a Noctua, Coooooool" The problem with the U12S though isn't anything to do with the U12S, it's to do with the falling cost of AIOs. If we look at this as an air cooler in isolation then yes it's a cracking piece of kit, and in performance per pound measures is a good buy. If however we throw the AIOs into the fray then it becomes a very different ball game.
Thanks to Noctua for the NH-U12S on review today, you can discuss your thoughts in the OC3D Forums.