Noctua NF-A12x25 PWM Fan

[GlacierNine]

Now it's correct.

And Noctua has clearly been aiming for marketing hype.
What's with experience from NF-F12 and NF-P12.
And SPCR had to revise their original fan testing method, because NF-S12 was basically designed to fool free air impeller anemometer measurements.
While actually having lower airflow if there was slightest impedance/back pressure.
Easy to make quiet per RPM fan, if it really can't actually push that air except to downwind...

That's really not at all what happened. The S12 *does* push more air around. It just has terrible static pressure - and that's absolutely fine for what it is. Remember, you're talking about a fan that came out in 2005.

At that time, even SPCR weren't testing fans for static pressure versus airflow - that's a practice that came later. As did manufacturers quoting a static pressure figure in their fan's literature - look at their first fan roundup - http://www.silentpcreview.com/article695-page2.html#nexus

Not a single static pressure figure to be seen on any spec sheet. (And not on wayback on the manufacturers sites either)

The Noctua S12 did actually perform as advertised when used on case exhausts or non-restrictive intakes. It lost efficiency when placed on heatsinks or cases with restrictive fan grills. That's something we understand pretty well these days, with companies like Noctua producing the F12 and S12 PWM, where the F12 has more than twice the static pressure, but also produces more noise and less airflow (by Noctua's own spec sheets) and Corsair producing the AF120 and SP120, where the same is also true.

If it were a case of "fooling" measurements, these products wouldn't exist. They serve different needs in different ways. It's just that both manufacturers and consumers now know more than they did in 2006 about two things:

1 - How to achieve great performance
2 - How to minimise the degree of compromise required to achieve that performance.

SPCR are/were great people, but at the time in 2005/6 they did not have the breadth of knowledge we have today - in fact they were largely responsible in many cases for bringing that knowledge to us and changing the marketplace. They didn't always get it right - For example, the original Antec P180, a case SPCR helped design, originally shipped with a useless GPU cooling duct, restrictive front intakes with square stamped holes rather than the more efficient hexagonal ones we use today, and the front grilles were also more restrictive than they were on later revisions. (and even those were much more restrictive than the grilles on modern cases).

It's unfair to characterise Noctua here as attempting to "fool" anyone.

Aerocool?

Nope, those fans were made by SilenX. Their big selling point was the smaller hub for more fan blade length/area, and the aerodynamic shape of the hub. They quoted ABSURD noise and CFM figures at the time.

Here's one of their 120mm models - you can see how undersized the hub is, and the little "bladelets" on the hub face - https://content.hwigroup.net/images/products_xl/010430/silenx-ixtrema-pro-120mm-46-cfm.jpg

Unsurprisingly, they did not perform as advertised.

 

[Ubersonic]

Name the brand:

SilenX, amazing fans back in the old days

 

[GlacierNine]

SilenX, amazing fans back in the old days

Congratulations on managing to answer the question after the answer had already been posted.

They were not amazing fans. They massively exaggerated their performance in their spec sheets.

The company also got caught shilling on multiple accounts in the SPCR forums.

 

[VSG]

Hah yeah I remember SilenX. I am thankful they got exposed and mostly disappeared quickly after word spread out.

 

[lexluthermiester]

You must be new to capitalism

No, he's right. When companies get enough feedback from their customers, or potential customers, they make changes based on that input. This is the way the world works and has always worked. It's just now in the age of instant feedback response times, the companies in question can act/react sooner.

The S12 *does* push more air around. It just has terrible static pressure

You kinda just contradicted youself. "Static pressure" = "push more air around".

It's unfair to characterize Noctua here as attempting to "fool" anyone.

Correct. The performance of Noctua fans is testable and provable.

SilenX, amazing fans back in the old days

They were not amazing fans. They massively exaggerated their performance in their spec sheets.

Here we disagree. Personally tried out a few of those fans. They performed well. They unfortunately were, like Noctua's fans, ugly as hell unless you were going with a full metallic look in your case.

 

[RichF]

Frankly I expected more... way more.

My fans never get over 850 rpm outside stress testing

At 750 rpm, it delivers 4.2 % more cfm while being being 1.5% louder than the 5 year old design Phanteks
At 1000 rpm, it delivers 4.8 % more cfm while being being 1.0% quieter than the Phanteks
At 1250 rpm, it delivers 6.6 % more cfm while being being 3.2% louder than the Phanteks

While that certainly is a win, it's a win against a 5 year old design. When 1st read about Nocs' stengthened blade / reduced gap design, I was expecting improvements in double digits. But since Phanteks also has an upcoming stengthened blade / reduced gap design, that's the comparison on looking for. However, if I was to go out and switch from the Phanteks in my box now, I'm looking at a $480 investment for these 120s. Assuming the 140mm will cost the same, since (5) 140s came with the case, the (11) 140mm Phanteks fans I added (10 radiator / 6 case fans) cost just $110.

So I don't see folks rushing out to replace their fans .... as for new build, I will certainly have these in mind; I expected more... especially for $30 but now anxiuous to see how Phanteks similar design measures up.

There is only so much that can be done to improve fan performance.

SPCR's tests suggest that fixed speed fans can beat PWM fans by a slight amount — that a fan works best at exactly the speed its design is optimized for. Perhaps things have changed with the newer bearings and such but that was the impression I got from the reviews I've read from them in the past. For instance, an 800 RPM GT outperformed faster variants of the GT that were slowed to 800 RPM.

One thing that I wonder about is whether or not it's possible to strongly outperform the 120mm fan format. From what I've seen, the results have been mixed for 140mm fans. Size changes are going to probably have a bigger impact going forward than more tinkering with internals, dampeners, and such. Thick designs are supposed to have a static pressure advantage but the ones on the market I've seen tend to not do all that well because of the noise they produce. That may be due to the nature of what is being produced, though. It may be that thick profiles and larger formats hold the key to better CFM-to-noise. Another strategy seems to be increasing the physical size of radiators and making the fins wider-spaced. Instead of moving toward the ITX format, quiet enthusiasts might want to move more toward the giant external radiator format.

 

[EsaT]

That's really not at all what happened. The S12 *does* push more air around. It just has terrible static pressure - and that's absolutely fine for what it is. Remember, you're talking about a fan that came out in 2005.

Pushing air always needs producing pressure.
And fan optimized for basically zero back pressure is pretty bad idea in real world.
For all their advertising about that scientific design approach, they should have known it perfectly.
And my bet is on them knowing it precisely.
Because that science hasn't been any new "state secret" in long time.

Nope, those fans were made by SilenX. Their big selling point was the smaller hub for more fan blade length/area, and the aerodynamic shape of the hub. They quoted ABSURD noise and CFM figures at the time.

Here's one of their 120mm models - you can see how undersized the hub is, and the little "bladelets" on the hub face - https://content.hwigroup.net/images/products_xl/010430/silenx-ixtrema-pro-120mm-46-cfm.jpg

Unsurprisingly, they did not perform as advertised.

Biggest problem was bad bling bling fashion material making it worser than normal fans made from good material:
The fancy "metallic" plastic used for the frame may look flashy, but it's terrible for sound. Lightweight and brittle, the material picks up fan vibration easily and audibly, giving the fan a highly resonant character. At slower speeds, the resonance manifests itself as a deep buzz or growl, while at higher speeds it's more like a pure tone. The resonance prevents what is an otherwise low-noise fan from ever becoming truly inaudible, even at 5V.
http://www.silentpcreview.com/article739-page3.html

Standard black plastics using Scythe Slipstream actually performed better in airflow per PRM while having very smooth noise profile.

There is only so much that can be done to improve fan performance.

SPCR's tests suggest that fixed speed fans can beat PWM fans by a slight amount — that a fan works best at exactly the speed its design is optimized for.

More likely there's certain optimal speed range for specific bearing/motor design.
For those tested fans with different speed variants SPCR found lowish/middle speed variants generally having best noise profiles scaling very well at lower speeds.
While high speed variants didn't scale well having various "extra" sounds.

Funny, dude: i have that cooler on my sister's PC, which used to be my old PC. I upgraded that cooler to my current one many years ago (see system specs): still serves me well, even in this new Ryzen system.

Unlike water (pipe) coolers heat pipe coolers just don't have anything which could degrade/wear except fan.
Hence use of heat pipes in satellite thermal control, which also drove their development.

 

[jabbadap]

That isn't how decibels work.

Every 3 decibels increase in noise is a DOUBLING of sound pressure.

The human hearing system percieves this in a logarithmic way, so we perceive sounds to be twice as loud if they are 10db louder than another sound.

This is consistent throughout our range of hearing. A 30db sound compared to a 40db sound is half as loud. A 90 db sound compared to a 100 db sound is ALSO HALF AS LOUD, as we perceive it.

So sure, on paper a 5db increase from 36.4 to 41.4 is only 13.7%.

But actually, the sound pressure has increased by 333% (a doubling and then 2/3rds of a second doubling) and the perceived volume has increased by 50%.

Thats a huge difference in performance.

It's not 3dBs, it's 6dB. Sound intensity or power is doubling at 3dBs (like voltage(6dB) vs power(3dB)). Thumb of rule: what you can measure is doubling at 6dBs and what you have to calculate is doubling at 3dB.

 

[Ubersonic]

Congratulations on managing to answer the question after the answer had already been posted.

The fact the answer was posted by the poster before me should make it obvious that that post wasn't there when I started mine.

They were not amazing fans. They massively exaggerated their performance in their spec sheets.

The two are not mutually exclusive. Yes they did exaggerate their performance, (and yes they are complete **** by modern standards) however that doesn't detract from the fact that at the time they were amazing fans because you have to remember that they came at a time when the Antec TriCool was considered a good fan.

SilenX completely fudged their dB ratings to a comical degree, but due to the 38mm thickness of the fan and how bad everything else was back then they still offered amazing performance (comparatively).

Thankfully we would later have the definition of amazing fan improved further by the likes of Nidec and Noctua

 

[GlacierNine]

The fact the answer was posted by the poster before me should make it obvious that that post wasn't there when I started mine.



The two are not mutually exclusive. Yes they did exaggerate their performance, (and yes they are complete **** by modern standards) however that doesn't detract from the fact that at the time they were amazing fans because you have to remember that they came at a time when the Antec TriCool was considered a good fan.

SilenX completely fudged their dB ratings to a comical degree, but due to the 38mm thickness of the fan and how bad everything else was back then they still offered amazing performance (comparatively).

Thankfully we would later have the definition of amazing fan improved further by the likes of Nidec and Noctua

See, that first point might have been believable if it weren't for the fact I posted that answer at 8:24 PM on Saturday and you didn't post your reply until 12:58AM on Sunday. I doubt you spent 4 hours typing.

As for "how bad everything else was back then" - No. The SilenX fans were reviewed in a fan roundup by SPCR and utterly failed to impress. Also they weren't 38mm. - http://www.silentpcreview.com/article739-page3.html Right there in the specs. 25mm.

The Antec Tricool fans were considered middle of the pack even then - SPCR said: "But [it's flaws] should not take away from its good points: It is cheap, widely available, and is quiet enough for many casual users. "

Pushing air always needs producing pressure.
And fan optimized for basically zero back pressure is pretty bad idea in real world.
For all their advertising about that scientific design approach, they should have known it perfectly.
And my bet is on them knowing it precisely.
Because that science hasn't been any new "state secret" in long time.

Biggest problem was bad bling bling fashion material making it worser than normal fans made from good material.

Pushing air requires pressure proportional to the expected amount of backpressure or resistance. More pressure does not mean more airflow overall if you're hitting finishing returns over the amount of backpressure you actually need to overcome.

If you optimize for static pressure then you will be able to force air more efficiently through things like radiators and heatsinks.

If you optimize for cfm at the expense of pressure you wont be able to do that, but those fans WILL more efficiently replace the volume of hot air inside a case with cooler air from outside of the case. The ideal circumstance (assuming your case intakes are not highly restrictive) is high cfm-low pressure fans on exhausts and case intakes, and high static pressure-lower cfm fans on heatsinks and radiators.

The current trend of slab-fronted R5 and R6 like cases is making Cass intakes more restrictive, which makes high static pressure fans more useful in more instances these days, but in cases where that isn't a problem like the Silverstone RL06 and the H500p Mesh, the high cfm low pressure approach pays dividends.

This is also important in watercooling - low FPI radiators are radiators designed to present a reduced resistance to airflow so that low pressure fans (ie, fans spinning slowly, or with airflow optimized blade geometry) can be used. This allows quiet operation at the expense of some surface area/heat transfer capacity. High FPI radiators demand static pressure optimized fans to work efficiently, like EK's Vardar range, however Vardars on a low FPI radiator or, say, the intakes of a silverstone RL06, are wasted and noisier than is needed for that application.

The S12 is effectively the opposite of a vardar in terms of the airflow/static pressure compromise and should be used accordingly, but it is not in any respect a bad fan design and it is not marketed or designed to mislead.

 

[lexluthermiester]

See, that first point might have been believable if it weren't for the fact I posted that answer at 8:24 PM on Saturday and you didn't post your reply until 12:58AM on Sunday. I doubt you spent 4 hours typing.

There are times when a browser has been open and not refreshed. New posts are made and missed by a logged in user. This has happened to me a few times. Not unusual. Quit attacking them over it.

As for "how bad everything else was back then" - No. The SilenX fans were reviewed in a fan roundup by SPCR and utterly failed to impress. Also they weren't 38mm. - http://www.silentpcreview.com/article739-page3.html Right there in the specs. 25mm.

As I said above, I tried a few of those fans out and while they were noisy, they weren't much louder than most of the fans in that same price range and they performed well. Additionally, if you had actually read the review you cited, you would have discovered that the SilentX fan performed similarly to much of the competition.

So let's get back on topic and be nice.

 

[Ubersonic]

See, that first point might have been believable if it weren't for the fact I posted that answer at 8:24 PM on Saturday and you didn't post your reply until 12:58AM on Sunday. I doubt you spent 4 hours typing.

No I was driving, admittedly in hindsight I should probably have checked the thread hadn't been updated before finishing my post, but as I didn't want to lose what I had written (and it wouldn't have changed what I wrote) it's of little consequence

As for "how bad everything else was back then" - No. The SilenX fans were reviewed in a fan roundup by SPCR and utterly failed to impress. Also they weren't 38mm. - http://www.silentpcreview.com/article739-page3.html Right there in the specs. 25mm.

I think maybe you're remembering things wrong and Googling random info to try and back that up. The original SilenX fans that were popular with us water coolers at the time were indeed 38mm hence why they performed so much better than most of the stuff on the market back then, it was their trump card, the review you have linked is for a much later revision of the fans that conform to the 25mm form factor thus losing their main performance advantage. Also lots of newer fans had arrived on the scene by the time that was written which didn't have fudged performance numbers, this makes the downgraded SilenX fans look just as bad as they were, however it doesn't mean the original SilenX fans weren't awesome back in the day

 

[GlacierNine]

No I was driving, admittedly in hindsight I should probably have checked the thread hadn't been updated before finishing my post, but as I didn't want to lose what I had written (and it wouldn't have changed what I wrote) it's of little consequence



I think maybe you're remembering things wrong and Googling random info to try and back that up. The original SilenX fans that were popular with us water coolers at the time were indeed 38mm hence why they performed so much better than most of the stuff on the market back then, it was their trump card, the review you have linked is for a much later revision of the fans that conform to the 25mm form factor thus losing their main performance advantage. Also lots of newer fans had arrived on the scene by the time that was written which didn't have fudged performance numbers, this makes the downgraded SilenX fans look just as bad as they were, however it doesn't mean the original SilenX fans weren't awesome back in the day

I'd love a source on that, since the wayback machine for their site doesn't even show them advertising the fan in the SPCR review.

 

[SK-1]

Thank you for the review. Do they make a 140mm version?

 

[danbert2000]

Anecdotally, I replaced an NF-F12 with this new NF-A12 on a Be Quiet Shadow Rock LP top flow cooler in my small shoebox PC, and I went from max temps of 85 degrees on my i7-5775c at 3.9 GHz to 72 degrees. I was then able to tweak my overclock up to 4.0 GHz and remain between 75 and 78 degrees. And the whole thing is quieter. I didn't even max the fan out because I was getting the temps I wanted.

So, the new fan beat the "static pressure optimized" fan by quite a bit. This fan is a godsend to people with small cases and limited CPU cooler options.

Thank you for the review. Do they make a 140mm version?

They have adapters to use 140 mm mounts. No 140 mm yet but Noctua says this fan will beat most 140 mm fans on radiators, and from this review I bet it might...

 

[John Naylor]

Some interesting comments but far too many still rely on outdated results. And I agree, quality cost money but I am not going to pay 3 times as much for 4.2% more cfm while getting 1.5% more sound.

I also agree decibles aren't everything. Silentpcreview,com does a great job of addressing this subject as well as the fact that ya can't judge a fan by it's brand name. Phanteks 1250 rpm fan is great, their 1500 rpm fan is terrible.

http://www.silentpcreview.com/article1345-page7.html

"The Phanteks PH-F140HP/TS is the clear winner in every respect. It edged out the new Noctuas every step of the way, delivering the best overall results of any fan we've tested thus far. To top it off, it had cleanest, smoothest sound of all the new fans in this roundup. If we had to start from scratch, this might be our new reference model. "

http://www.silentpcreview.com/article1480-page5.html

"Given our great experience with previous Phanteks fans, the PH-F120MP is an utter disappointment. It performed poorly on both our cooling and airflow tests and it has an absolutely atrocious sound. As it's marketed as a radiator fan, it may perform better on a watercooling setup but I remain skeptical considering our previous tests with AIO coolers. "

Also another reference to what i said earlier about the unexplicable focus on "high SP" in this day and age.

"Our reference fans, which have traditional designs usually perform similarly to the stock fans provided despite lacking the high static pressure ***widely believed*** to be optimal for pushing air through tightly packed spaces.

The fin spacing on today's heat sinks and fpi on today's rads just do not require a high SP fan

So, the new fan beat the "static pressure optimized" fan by quite a bit. This fan is a godsend to people with small cases and limited CPU cooler options.

That didn't hold true last time ....

The 140s are in yellow... and took 4 outta the top 5 spots. The top 3 were tightly packed in the table and here the difference was more about sound quality then the small dbA dfferene... but the kicker for me was ...

a) The case came with (5) Phanteks fans, when making a choice as what to add, it didn't hurt that they took 1st place, but had then been in 3rd, it was close enough that I wouldn't replace the (5) free ones.

b) I purchased the components over a period of (3) months ... the Nocs hovered consistently at $30 .... the Phanteks $10 - $15. I bought the (11) extra fans for $110 ... Using Nocs wuda been $480. So, to my eyes, (pun intended) besides the "polarizing color issue", Noc needs to address the cost issue, especially if Phanteks own strengthened blade / reduced frame - blade gasp design comes in at half the price again.

 

[Blueberries]

I also agree decibles aren't everything. Silentpcreview,com does a great job of addressing this subject as well as the fact that ya can't judge a fan by it's brand name. Phanteks 1250 rpm fan is great, their 1500 rpm fan is terrible.

CFM/dB is the only metric that matters as far as I'm concerned. I'm going to run my fans at the highest inaudbile RPM for as long as possible before the heat threshold where I absolutely need to go higher. The better the performance of the fan, the longer they will stay at an inaudible speed.

 

[Ubersonic]

I'd love a source on that

Which part?

 

[Vlada011]

Tomorrow arrive 3 x NF-A12xPWM.
I found three for price of two and couldn't resist. Only that decide.
Noctua told they prepare black version maybe in 2019.

But if you ask me they make BIG MISTAKE because didn't launch in th.
Can you imagine NF-A12x25 in this color. AMAZING, AMAZING.
How they missed that chance, because this color scheme fit with everything, litterary every hardware color.

 

[Vlada011]

Guys if you want o buy this fan, Noctua NF-A12x25 PWM, you have my official support.
Rob your parents, wives and family without remorse.
Only grey color scheme could make me sad now.
Everything is better quality than GT, blade, frame, accessories, package, cable, color is not as you want but you could use Phanteks Halo RGB i saw one video clip look nice on Noctua fans,

You can use these for RGB

http://phanteks.com/Halos.html

I can't remember when I hurry so much to buy some fans, I mean after launch date.
3 fans + 2 140mm adapters 70 euro. Not bad. 140mm Adapter Noctua will serve if one day smaller case have 280mm radiator.

 

[HTC]

Guys if you want o buy this fan, Noctua NF-A12x25 PWM, you have my official support.
Rob your parents, wives and family without remorse.
Only grey color scheme could make me sad now.
Everything is better quality than GT, blade, frame, accessories, package, cable, color is not as you want but you could use Phanteks Halo RGB i saw one video clip look nice on Noctua fans,
View attachment 104492View attachment 104493

You can use these for RGB

http://phanteks.com/Halos.html

I can't remember when I hurry so much to buy some fans, I mean after launch date.
3 fans + 2 140mm adapters 70 euro. Not bad. 140mm Adapter Noctua will serve if one day smaller case have 280mm radiator.

Never heard of this before: nice to know there's this kind of option out there.

If i'm ever affected by the "RGB craze", i'll get some of those.

 

[danbert2000]

That didn't hold true last time .....

Hey man, I was surprised too. I was expecting maybe 5 degrees, getting 12 to 15 was like Christmas. And it is much more tonally pleasing than the old NF-F12's at full bore, there isn't as much fan drone. Definitely worth the $29 upgrade for my situation. All I can do is speak to my own experience with the fan.

 

[Vlada011]

Sleeve is very nice done on them. I didn't knew thy will sleeve them so nice.
Only one is flaw and that's no two color options. Redux scheme would fit perfect, better than black.
I'm totaly not interest for RGB and LED fans.

You can make RGB from these, with Phantels Halo RGB.
This is image of Noctua with Phanteks RGB Frame and White EK Vardar.

Video clip to see different color, frame is black...

One thing is sure, forgot to pain them. You will destroy performance, feeling of plastic, best alloy on market and blades will weight more, probably will be out of balance as well and cause more vibration. For painting fans is only for color scheme and primer, good pain, everything need a lot of skill. That's it as you see them.

But they fir perfect in some Gunmetal Military RIG dedicate to USCM example.
Some case labs S8S Limited Edition with military look switches and controllers.

I still think how much I luck I had to find them, 6 Noctua fans arrived for testing in 4K resolution and they are tested.
I bought 3, other 3 are sent to other reviews and guy will keep them for self. No availability in stores in my country.
You have luck to cost 29$, I paid 70 euro 3 fans and 2 adapters 120-140mm and free shiping but in country nearest to us they are 35 euro.
But in my country will not be below 37-40 euro single fan. I would not have so much for 3 of them.

 

[terroralpha]

Everything is better quality than GT, blade, frame, accessories, package, cable, color is not as you want but you could use Phanteks Halo RGB i saw one video clip look nice on Noctua fans

my gentle typhoons are 6 years old and went through 4 builds. they still spin up to 1850RPM and they did not get any louder than what they were like the day i bought them. i also paid half of what Noctua is charging. i tried "upgrading" to EK Vardar EVO fans because i wanted an all white build but my temps actually went up about 5*C. so I went back to GTs. in this review, the Vardars beat the noctua fans and my GTs are better than my Vardar EVOs. so i'm going to call BS on the "better than GT" claim.

im curious, what makes you think the frame or blades are better on the noctua fans? i understand that the vomit tan and diarrhea brown color scheme just screams quality at first glace, but is there a factual basis for your claims?

PS: if anyone wants white 120mm EK Vardar fans, i'm still trying to sell my last 3. PM if interested

 

[Vlada011]

Noctua fans are best at the moment, NF-A12x25 PWM.

They are best because they are most silent on 1500-1800-2000 RPM and only few fans push more air just for little.
Off course silence and absence of noise on very low speed is weak spot of many fans, but this model of Noctua is good and there.

EK Vardar and GT push 2-2.5 CFM more air but they are more than 6-7 db louder on same speed.
That make Noctua so good and wanted. If your fan is 8 dBA more silent on 1200 RPM, that directly give you option to increase fan speed speed on 1300-1350RPM and noise will be same as competition but your fan will than give better CFM because spin faster.

Most people will keep them on 1000-1200 RPM, but I think they are best 1500-1700 RPM.
I have one GT 1850 model, he is good fan if you have them and if they are still in good condition, but Noctua on full speed 2000 RPM I doubt it's louder than them.
If you buy fan now no any reason to buy others because you get
good fan,
good warranty,
rubber gasket for radiator,
sleeved adapter for noise reduction 2000 >1700
8 rubber in corners,
rubber to hold fan instead screws,
sleeved Y PWM cable ,
normal screws
sleeved extender cable

First week when I bought Noctua fans I thought did I made mistake. Only because color... Nothing else.
I didn't used them and I didn't worry because I could sell and back all money because I found them cheaper, 3 for price of 2.
But than I install them... I was amazed first during installation fan and special when start to work... My friend who don't know much about computers asked me... What is this, this spin or not and want to touch with fingers... I told him it's fan move your hand, and he laugh and say They looks like not work. You see big motor in middle and no noise and for someone who don't use computers at all look weird.
PERFECT BALANCE, PERFECT WEIGHT OF QUALITY PRODUCT, PACKAGE, ACCESSORIES, WARRANTY... and I start to like their color.
My first intention is to buy Noiseblocker Black Silent PRO, full black classic fans, German, but they are not competition to Noctua and now I like Noctua military color scheme.

Now I sorry only because one thing, guy who sold me didn't want to sell all six for 150 euro.
And I would really really like to have 3 more. Better to say I need them for second rad. But 110 euro is best I can find 3 of them.