Picking out new fans for a build is a process you either love or hate. You’ve come to terms with the fact that your PC needs fans – after all, air has to be moved to cool it. For today’s comparison, let’s take RGB out of the equation – assuming all you care about is that your PC gets cooling, nothing more.
We quickly run into the triangle of cooling:
Now, you have a choice to make: do you prioritize cooling performance, cost, or noise? You only get to pick two.
Say you want the best cooling fan: It has to have the absolute best cooling performance, run absolutely quiet, and cost is no object. There is one clear winner there: the Noctua NF-A12x25 G2 — at least by reputation. We’ll only know for sure whether that’s true when we’ve tested a lot more fans. But let’s take it as a given for now.
In comes an option on the other end of the spectrum: You just need some fans to move air, and please do so without adding much more cost to the already-expensive build. Since I’m deciding which fans to spend time on based on their reputation, I’m picking the Arctic P12 Pro as the competitor in today’s head-to-head, in what may be the most unfair fan comparison in history – or at least in PC fan history.
However, there is a reason behind this madness: The Noctua NF-A12x25 G2 is a brilliant fan, but it costs a mighty $35 on Amazon for a single spinner. If you need five fans for your PC, that’s $175, spent on just fans.
On the other side of the spectrum sits Arctic, with the recently released P12 Pro. A single P12 Pro costs $8.50. Want a 5-pack? Sure thing, that’ll be $27.50, please. Do you see where I’m going with this?
For what effectively comes down to a cost difference of 6:1, how much noise and cooling performance are you really sacrificing by using one of the cheapest fans that still has a good reputation, as opposed to the most expensive? The difference might be less than you think. Because if being an enthusiast for decades has taught me one thing, it’s never to underestimate the underdog.
Deeper Dive: Noctua NF-A12x25 G2
The Noctua NF-A12x25 G2 (what a handful, that name) is Noctua’s second-generation A12x25 fan. The A is for all-rounder, at 120mm in size and 25mm thick, and it is the second-generation model.
The all-round aspect of the fan is key here: The point of the NF-A series fans is to offer great performance, both for airflow and static pressure. This means that the NF-A12x25 G2 will function brilliantly as a case fan where its sole purpose is to move heaps of air. But it should also function great as a fan on a radiator, where it encounters much more resistance.
Obviously, building a fan that can move a lot of air through an unrestricted opening but also effectively cool a restrictive radiator comes with its own complexities — especially if you then also want this one-model-fits-all fan to achieve top-end acoustic performance.
So, hats off to Noctua for trying. In the past, the company was only satisfied with these two categories by offering two different pedigrees of fans. But as of 7 years ago, the company firmly stands behind the NF-A12x25, and more recently, the G2 variant.
To achieve its capabilities, the NF-A12x25 G2 features a self-lubricating, self-stabilizing fluid-dynamic bearing, which the company calls the SSO2 bearing, with SSO denoting self-stabilizing oil pressure. I suppose SSOP2 didn’t have the same ring to it. In essence, this is a hydrodynamic bearing with an added magnet that self-stabilizes the rotor axis for a very long and very quiet running lifetime.
From there on out, the NF-A12x25 G2 has a beige frame with vibration dampers on every corner. But at the heart of it all is where the magic happens: the 9-blade, Sterrox LCP fan blade assembly, shaped with a strong forward sweep, steep angles, impeller winglets, a centrifugal turbulator, and flow acceleration channels. It all sounds like marketing talk, but the sum of it all seems to produce a winner.
What I find most noteworthy is the 0.5mm tip clearance from the frame – it’s honestly unreal to see the fan blades so close to the frame – and that, paired with the high blade density, is bound to do good things for the static pressure.
To top it all off, Noctua includes a low-noise adapter, and the hub’s PWM microcontroller is even equipped with programming to ensure that even when pushing through high-impedance cooling devices, the fan retains the RPMs it is meant to achieve and a particular input voltage and PWM duty cycle.
By default, the Austrian goldilocks spinner will rotate at up to a bit above its rated RPM, but the company also includes a low-noise adapter that lowers this figure to give or take 1500 RPM. As an enthusiast feature, the fan will stop when fed a 0% PWM duty – something not all fans do, in order to protect the system.
To add complexity to the matter, Noctua also introduced the NF-A12x25 G2 Sx2-PP, which is a kit with two of the spinners, one of which is programmed to run 25 RPM lower than the standard model, and one 25 RPM higher.
Why? Because the NF-A12x25 is said to be so painfully consistent with its rotational speeds per PWM figure between samples, that if you group multiple fans together on one motherboard header, running off the same input voltage and PWM signal, they become harmonic. That may be beautiful in an orchestra, but not so much so when your fans play sing-along with each other in your rig.
Come to think of it, I should have asked Noctua for two additional NF-A12x25 G2 fans for review. Today I only have a single unit, along with a two-fan NF-A12x25 G2 Sx2-PP kit with the speed offset. I’d like to experience the harmonics they speak of, but that’ll have to wait for another day.
Deeper Dive: Arctic P12 Pro
Diving deeper into the Arctic P12 Pro may not be as exciting as the Noctua, but I still find it intriguing. Why? Because for a $7 fan, it’s surprisingly well-built.
Instead of a 0.5mm tip gap, the Arctic fan has a ring around the fan blades. The reason for this is not aesthetic, but rather, for added strength. Whereas Noctua uses Sterrox LCP with high tensile strength, we can’t expect such exotic materials from a $7 fan. The Arctic fan is made from plain ‘ol plastic.
That’s where this ring design element comes in. It gives the fan blade piece added strength – a strength that the plastic itself does not naturally have. As the P12 Pro’s fan blades connect to the hub on the smaller side of the blade, sweeping forward, and can then spin at a mighty 3000 RPM, you can imagine the destruction they would cause if the blade assembly were not reinforced somehow.
I wonder if Noctua’s fan would survive 3000 RPM. Should I sacrifice one to find out? Would its PWM controller survive the 24 volts needed to achieve this?
I’m trailing off target, but the moral of this story is that there really isn’t anything all that interesting about the Arctic P12 Pro fan when you look at it. It doesn’t use exotic materials. It’s not brown and beige. It doesn’t have a fancy 0.5mm tip clearance, and it uses a ring around the blade assembly to hold itself together. What kind of cheap trick is that?
However, by reputation, it’s a blimey good fan, and from the synthetic testing I’ve seen elsewhere, it trades blows with Noctua’s finest for generating airflow and static pressure. I found this all a little difficult to believe, so I had to see for myself.
Testing Methodology
For testing PC fans, there are two main terms you should be aware of: Airflow and Static Pressure.
Airflow is easy: How much air does a fan move in an unrestricted wind tunnel? Static Pressure is a little more complicated – but essentially it boils down to how well a fan can overcome restrictions such as a radiator or heatsink.
Static pressure can be measured by placing the fan in a wind tunnel, closing the end, and measuring the pressure generated with a sensor, which spits out a value in mm of water. The higher the number, the better a fan should be at overcoming restrictions.
However, I find that testing a fan’s static pressure doesn’t actually tell the whole story, and the reason has to do with acoustics.
Radiators and Intakes Fundamentally Change Acoustics
In lots of fan testing, you’ll see acoustic measurements, followed by airflow and static pressure tests. The problem I have with these tests is that they use the acoustic tests of a fan in an unrestricted environment and use the RPM figures from this test to determine static pressure at certain noise levels.
However, have you ever held your hand closely in front of a PC fan? It gets louder. The same goes for a fan installed on a radiator.
Therefore, I will not be testing static pressure. Instead, I prefer a more real-world test – figuring out the fan’s acoustics not only by itself, but when strapped to a radiator and also behind a mesh intake with an air filter.
Once I have datapoints for the RPM’s at noise levels in all three scenarios, I pop the fan onto the wind tunnel, by itself, on the radiator, and behind the mesh intake, set it to the rotational speeds found in the acoustic testing phase, and measure the resulting windspeed with an anemometer.
This takes quite a bit more time to test for, as I end up with a bit over 80 datapoints per fan. But I feel it paints a more nuanced picture on top of the standard airflow test.
A bit about our test environment
For testing acoustics, we do not have access to a certified anechoic chamber. For that matter, nor do we have access to an uncertified chamber. Personally, I live in an apartment near a never-ending construction site. Not ideal circumstances for testing fans, however, this is not a huge problem – it just means I run the acoustic tests late at night.
My home office is padded with thin acoustic paneling on all walls, a part of the ceiling has thicker absorption foam, and there is a bookshelf that helps with diffusion. Thin paneling doesn’t do much for low or mid frequencies, but in the range where fans operate, it does quite a decent job of absorbing the sound. At night, I have a noise floor of 33.2 dBA, and if you’re familiar with acoustics, you’ll know that this is really quite silent.
Although this means we won’t be able to measure a fan’s acoustic profile low down in its RPM range, chances are slim that you have your PC somewhere quieter. As fascinating as it would be to analyze in the very quiet range from a scientific point of view out of pure interest, given how quiet my office is, we can confidently give buying advice without the use of an anechoic chamber.
The Suspense
For testing, I suspend the fan mid-air. I do this because I found that using a clamp or bracket of some sort to hold the fan up can interfere with the readings. For example, in this round of testing, I initially found that Noctua’s fan produced the exact same figures as Arctic’s. Now, I knew Arctic’s fan was supposed to be good, but this seemed to be too good.
Something clearly wasn’t adding up, and it seemed to be down to a tiny lip of plastic that didn’t get in the way of Arctic’s blades, but because Noctua doesn’t use the ring and has the blades pass much closer to the outside of the frame, it caused a bit of turbulence and made the Noctua run louder. I don’t think the vibration carried into the tripod holding the fan up was helping either, so I decided the best way to eradicate external influences as best as possible was to suspend the fans mid-air by means of a string on each of the four corners.
Arctic P12 Pro Vs Noctua NF-A12x25 Testing Results
Simply pinning the Arctic P12 Pro against the Noctua NF-A12x25 G2 isn’t going to give us a whole lot of sensible data by itself, without anything else to compare to. Therefore, I decided to add two more fans to the mix: Noctua’s previous-gen NF-A12x25, along with Phanteks’ M25 G2 as a midrange option, which usually sells for anything between $7.99 and 11.99.
For the testing results, we are starting with the noise floor findings, followed by the dBA vs. RPM data, and concluding with airflow analysis.
Breaking the Noise Floor
In my testing office, the noise floor is at 33.2 dBA. This is pretty quiet, and representative of most home offices when nobody else is home, with all appliances and computers switched off and no nearby traffic.
Arctic’s P12 Pro, at its minimum speed of 510 RPM, breaks the noise floor at 33.7 dBA, and at this speed, it produces a very slight but audible hum, along with a higher pitch from the motor. This is only audible with sensitive ears in a very quiet environment.
Noctua’s NF-A12x25 G2 has a minimum speed of 370 RPM, and at this speed, it is completely inaudible in this office. It does not break the noise floor until it reaches 600 RPM, when it only produces gentle wind noise. To my ears, there is no mechanical or motor noise to speak of.
dBA vs RPM: Unrestricted
A-weighted decibels (dBA) vs RPM isn’t usually all that useful – the metric that really matters is dBA vs. Airflow, which will come in a bit. However, both the Arctic P12 Pro and the Noctua NF-A12x25 showcased some very interesting phenomena, and since this is an analysis of two flagships, I would like to share this with you.
In the first test, the Arctic P12 Pro comes out looking very strong, managing the highest number of rotations per minute at each given noise level. It is followed closely by Noctua’s previous-generation NF-A12x25, and Noctua’s G2 spinner isn’t looking so great at this point.
Across the test, Arctic’s P12 ran 14.8% faster than Noctua’s fan, and at 41 dBA, where Noctua’s G2 spinner ran at 1756 RPM, nearing its full speed, the P12 Pro spun at a dizzying 2048 RPM – almost 300 RPM faster, while maintaining the same noise level.
dBA vs RPM: Radiator Restricted
Put these fans on a radiator, and something very interesting happens. The P12 Pro remains in the lead for RPMs at each noise level, but I want to look more closely at the NF-A12x25 G2. It suddenly jumped way up, running at almost identical RPMs as the NF-A12x25.
I ran the numbers and found that the P12 Pro suffered a 5% RPM penalty for being mounted on a radiator. The NF-A12x25 also handed in lower RPMs on a radiator, suffering a total 7% RPM loss to maintain equal noise levels.
The NF-A12x25 G2 is quieter when installed on a radiator
However, the new NF-A12x25 G2 achieved something quite miraculous: On a radiator, it was able to spin on average 2% faster across the test range, while maintaining noise parity with the non-radiator test.
dBA vs RPM: Mesh + Filter Intake Simulation
This is the most brutal of all the tests. The mesh intake sits in front of the fans, and all the fans' RPMs need to drop significantly in order to maintain dBA parity. The NF-A12x25 needs to run 12% slower to maintain noise levels, and the P12 Pro gets a 16% hit.
dBA VS. Airflow: Unrestricted
This chart represents the most holistic of the three tests. The fan is mounted onto the wind tunnel, the fan is set to the RPM figures from the acoustic tests. The resulting windflow is measured and calculated to cubic meters per hour.
Leading the race is Noctua’s finest, followed closely by Arctic’s P12 Pro. This means that although Arctic’s fan may be able to run at much higher RPMs for each given noise level, that doesn’t actually mean it moves more air. In fact, it has to spin much faster than Noctua’s G2 fan to move the same amount of air.
dBA VS. Airflow: Radiator Restricted
Pin the three spinners to a radiator, and you’ll find that the gap between the NF-A12x25 G2 and Arctic P12 Pro starts to grow. However, the Arctic fan still pushes more air through the radiator than Noctua’s last-generation NF-A12x25, which is quite a feat, especially for a $7 fan.
dBA VS. Airflow: Mesh + Filter Intake Restricted
Burdening the spinners with a mesh intake, here we start to see Noctua’s superiority. The Arctic P12 Pro is at last defeated by both Noctua fans.
That being said, it still performs admirably, staying very close to Noctua’s performance, and well above Phanteks’ M25 G2 fan.
The dBA figure doesn’t say everything
Throughout my testing of these fans, I spent countless hours listening to them, and although Arctic’s fan delivers highly impressive figures on the SPL (Sound Pressure Level) meter and anemometer (wind speed meter), there was something about the sound that wasn’t quite as comforting as Noctua’s fan. The NF-A12x25 fans, especially the G2 version, produce a very smooth sound – all you could hear was a gentle airstream.
The P12 Pro, on the other hand, although not terrible by any stretch of the imagination, wasn’t as refined in its acoustic profile. There was a rhythm to the airflow. Upon startup, you’d hear the motor click a couple of times, and even when running at its minimum duty, there was a quiet, but just-audible high-pitched noise coming from its motor.
While handling the fan, I noticed that if it was still spinning and I grabbed it from the front to swap it for another, the Arctic P12 Pro sounded choppy. It was the same choppy sound it normally had, but as soon as something got close in front of the fan, it got much louder. This was not an issue with anything behind the fan, but it did lead to adding the intake tests to the suite, which also clearly shows that not all fans respond equally when presented with the same circumstances.
Notes about intake fan noise
Adding intake restrictions harmed the resulting airflow more than radiators. Of course, it must be noted that the acoustic penalty for strapping a fan to a radiator is much smaller than putting it behind a mesh intake. As a result, the airflow penalty is higher on intake fans, not because the intake is more restrictive than a radiator – it isn’t – but mainly because the fan simply needs to run much slower to maintain noise parity, and the intake penalty gets worse the closer the fan gets to the intake grille.
This is where the choppy sound of the P12 becomes problematic, and I recommend buyers ensure there is more than 30mm (1.18 inches) clearance between the fan and front mesh if using a P12 Pro as an intake fan. It’s worth noting that in some cases, this can be problematic: without a radiator, you could opt to mount the fans behind the mounting shroud instead of between it and the front mesh, which would give you plenty of space – but if combined as intake and radiator at the same time, the assembly may become too thick and get in the way of long GPUs on small cases if built this way – but we’re getting into minute details here – just keep it in mind when deciding which setup you’re pairing it with.
In Conclusion
When these two fans arrived in the mail, I held them next to each other, and I had a right laugh – there was no way this cheapo-plastic fan with a ring around its blades was going to beat the beige-and-brown beast. The NF-A12x25 G2 was going to run circles around the P12 Pro. I had invited the circus to my house, and I was the audience.
But it turns out, I was the clown – even with the Noctua as my show partner, the P12 Pro laughed in our faces.
To say that the performance testing surprised me would be putting it lightly. Noctua’s NF-A12x25 G2 is a brilliant fan. Overengineered beyond all measure. And yet, in performance testing, Arctic’s P12 Pro stayed at its heels, desperately clawing for first place, and occasionally coming very close.
The Arctic P12 Pro costs $27.50 when purchased in a 5-pack, whereas the Noctua NF-A12x25 G2 has a street price of $35 each. This means you can buy six P12s for the price of one Noctua G2 fan.
The P12 Pro gets you most of Noctua’s performance for a fraction of the cost. The sacrifices in performance you make are a slightly higher minimum speed, slightly more turbulent wind noise, a bit of mechanical noise from the motor when it’s otherwise very quiet in the room, and you have to be careful about not placing it too close to a front intake mesh.
But I have to emphasize again, we’re comparing to the crème de la crème here, and that’s saying something and picking the nittiest of nits. In exchange for pledging loyalty to Arctic, the P12 pro gives you almost Noctua-level performance at medium speeds, obscene power at its full duty with over 3000 RPM to play with, and it comes in an RGB flavor, which we’ll be testing soon, too. All for a fraction of the price.
The P12 Pro is ideal for gamers who, as soon as their PC boots up, fire up a game and get cracking. If, however, you use your PC for productivity and it spends a lot of time idling, and you know you’re sensitive to these kinds of noises, it’s hard to argue with Noctua’s NF-A12x25 G2. It offers a level of refinement that makes you want to address it formally – the way it deals with ultra-low RPM ranges is so beautiful that I’d almost pay money to see what it’s capable of in an anechoic chamber.
Both of these fans will likely be earning a highly-deserved spot on our upcoming Best PC Fans page, along with our Editor’s Choice award.
