Something is Wrong here - Alphacool Apex Stealth Metal Fan Reviewed

[@JJFX-]

What bothers me most isn't so much that Igor got a crazy test result but that he didn't recognize how absolutely ridiculous that would be. Mistakes happen but he's experienced enough to know this just isn't going to happen unless there's a serious malfunction somewhere. Honestly, I hate to say it but I have to question what motivated publishing such figures. I am noticing a strange marketing push behind what is effectively a decent fan sandwiched between two metal plates.

[@riklaunim]

Isn't he first to jump to conclusions? RTX 3000 capacitors, RTX 4000 power connector melting and some more.

[@Bularistan4o]

The marketing push is called publishing embargo and the product being released just now. Why would he, Jay, Roman, I think Steve too, be publishing around now if not at the same time. As for the motives - who knows.

[@noxious89123]

Igor was involved in the development of these fans with AlphaCool. He's a complete shill at this point.

[@fredEVOIX]

@@riklaunim both were true and still are so I hope he said they were a problem not wrong ? the capacitor thing better quality is always better especially when you know an actual pcb designer like me and he showed you his training on how to lower quality as much possible so it lasts only the warranty...they are taught to do this and the connector is an nvidia design problem, only people who have never seen one would think otherwise the 12v connector is SMALLER than a single pciex8 and we had 3-4 of them also the problem is rtx4000 are wider than prev gens or amds and that connector adds more width wich no case can handle, they all push on the cable and move it, it's not a human error you cannot do miracle it's going to touch the side panel or glass

[@nimbulan2020]

It doesn't surprise me at all. Igor's a clown who often pushes obviously incorrect information or jumps to conclusions while seriously lacking in necessary data. He's just really good at dressing up his BS as thorough technical analysis so he can convince a decent number of people he's right.

[@_kardus]

As an engineer I'm sure you knew from the beginning that both of those claims being true was almost a near impossibility...

[@Spiggle.]

The Airjet coolers need to get bigger/better, then we can forget about regular fans I guess.

[@timshel1429]

The graph themself are shoddy; they all include: "linear interpolated based on measured values at 500, 1000 and max RPM". so only 3 values ... For the alpha cool there is over 20 data points which placement do not look like anything close to the result of a linear interpolation. So they randomly placed points to make the graph look more "real", tells you everything you need to know ...

[@47CryXMA]

Major Hardware has been doing 3D printed fans that his viewer's make, to see if anyone can beat the Noctua A12x25 fan. Very, very few have been able to edge past that fan. For another test to claim 2x better in any comparable regard, has me really questioning their test methods.

[@Cypeq]

@@Spiggle. hoverboards and flying cars will be out before that.

[@wyattroncin941]

Using an integrated duct on the blade tips can reduce noice at the same performance. But with injection molding adding that duct ring means they have to use a much simpler blade geometry. So to me it's possible they did have a better performing fan diring the prototype stage. And then manufacturing told prototyping they can't make the geometry, so performance got cut. But the product was already approved, and marketing was already off to the races, so they had to ship something.

[@MaxNakfoor]

Igor's testing has gotten pretty unreliable over the years, glad to see you doing your due diligence. These are about the results I expected from these, Igor's just seemed ludicrous.

[@noxious89123]

Igor's testing was straight up making fraudulent claims. He was involved in the development of these fans, so he is not impartial.

[@Xfade81]

@@noxious89123 was this mentioned in his video or somewhere ?

[@varmastiko2908]

This was not the first time Igor's been posting bogus test results. The problem is consumers thinking testers are to be trusted. Nobody is trustworthy simply by virtue. Always double check every claim no matter who is making the claim. Same goes for much more important topics than tech testing.

[@stevog893]

Igor also reported higher SP numbers on the 14900k samples than the norm.

[@MasonStormSunny]

His results are exactly what he was paid for.

[@tuomashautamaki2258]

Alphacool has Cybenetics test results on their web page and interestingly CFM and static pressure are both about 33 % lower than Igor's tests at max rpm (or Igor's numbers are 50 % higher). Cybenetics max CFM is 71 and static pressure 4.2, while Igor's max CFM is 117 and static pressure 6.2. I'm pretty sure that Cybenetics results are much more in line than Igor's.

[@tuomashautamaki2258]

Cybenetics also has Noctua NF-A12x25 test results and if we compare both fans at 2000 rpm, Noctua's CFM is 30 % higher, and static pressure is 22 % higher than Alphacool's. So Roman's tests are quite well aligned with these.

[@rammagaming]

Thanks for pointing that out! Cybenetics only have the 3k RPM Apex and no 3k RPM Noctuas to compare, but I doubt there's any major differences between the Apex variants. Still not terrible test results despite not being revolutionary like they touted. It's BARELY quieter than Noctua's 3k RPM fans at full speed (comparing Cybenetics test to Noctua website, 40.1 vs 43.5 dBa). But they lose massively in CFM by 38 and static pressure by 3.4 mmAq. Hopefully someone like Gamers Nexus gets there hands on these soon so they can be put through testing on the same equipment. Probably still far from a game changer at this point, but could be a nice option for a stylish case fan.

[@forsgsrekylgevr2113]

if the company is advertising stats lower than what an independent reviewer gets, then theres no way the independent reviewer is right

[@Frozoken]

​@@rammagaminguhhh no u absolutely cannot compare a manufacturer's noise rating to an unbiased test. The t30 demolishes that fan in every way including noise at the same rpm and the t30 makes roughly the noise u claimed for the noctua at 3000rpm. That 3000rpm noctua fan has notably worse noise to performance than the nfa12 even. Alas these new fans just continue to make me appreciate the t30 more and more as it just demolishes new premium fan after new premium fan. Tempting me to buy some lmao.

[@j4ck3t]

@@tuomashautamaki2258 Oh interesting! Cybernetics is very honest if they show those results. That deserves praise for not deceiving the customer.

[@SilkMilkJilk]

Tell Steve to use his new toys. Not the battery powered ones, the big air chamber thing I mean.

[@TheDeeGeeNL]

They bought that thing like what... 5 years ago...

[@davidepannone6021]

I don't understand why he made such a big deal a year ago when he got the chamber and he has yet to come out with a video that test fans in it. Especially after spending so much money on it.

[@prorataxns8425]

I think they may have found that there was no hidden lies in fans.

[@mycosys]

They have to develop standardised test methods, and a comparative/control data set to test against. @@davidepannone6021 - its like designing a scientific experiment, and its time consuming.

[@Smallman647]

​@@KoudZit passed like 2 years...

[@Zordonzig]

Not nearly enough TechTubers cover the Aric P12 Max. Thank you so much for including it!

[@joannaatkins822]

As they say; extraordinary claims require extraordinary evidence. I'm glad you checked this

[@kurgo_]

I can't really imagine a fan coming out that's half as loud and twice as performant as just about *any* other fan, let alone a Noctua one (no fanboying, but their stuff are objectively usually good), while costing less, even. Wouldn't be the first time igorslab messed up (what with the whole 4090 connector debacle) but it'll be interesting to see if GamersNexus or other reviewers try these and what results they'll get.

[@shadow7037932]

GN has their new fan tester setup, so hopefully they will test.

[@SianaGearz]

@@shadow7037932 Aris (Hardware Busters, long time PSU reviewer/expert) also has the same fan testing contraption from Taiwan like GN got themselves, he might be curious to check that as well. Indeed unless you get some kind of horribly unbalanced fan no-name reject, fans of similar construction speed and purpose perform... sort of eerily similar to each other, that's what you expect. Oh btw i have a pet peeve about people calling 10dB "half as loud" or "double as loud" because dB scale is actually perceptually linear; but also you have an internalised experience that two of some uncorrelated noise source aren't twice the dB but just 3dB louder, not all too much louder than one of these noise sources. If the sources are speakers playing the same signal then doubling them is 6dB. This makes comparing loudness really weird and the only sensible statement that can be made about one source that is 10dB louder than another is just that it's 10dB louder, not some ratio louder. But yeah 10dB is a massive crazy difference. If you have a fan that's 10dB less loud at a certain performance level, then 10 of those fans taken together are as loud as one louder fan! Just imagine the implied efficiency difference, that you could 10x your air performance without any extra noise. But for speaker (highly correlated sound sources) 10dB ratio is approximately 3 speakers worth, not 10. 10 would be 20dB. dB business melts your brain, and there just isn't necessarily a better measurement, because audio perception is so weird.

[@thorwaldjohanson2526]

Yeah, these results should have raised an immediate flag for any tester. You would have to very actively try to set up a test that gives these results.

[@RonnieMcNutt666]

Thermal mass is incredibly powerful sound dampener, in fact 4 cents of hot glue can make a 15 or so watt fan have reduced noise and vibrations 80% but these companies are so cheap and can only give tiny rubber pads I guess lol

[@lemagreengreen]

If someone had genuinely come up with a fan design that outperformed the previous best by 2x that would be an engineering/scientific breakthrough and break a lot of understanding of previously well understood things... how that faulty result slipped through to the point of publishing is confusing since that obviously has not happened.

[@DLTX1007]

The package size of the montech means the outer box, thankfully. I own a few of the metal 120, P12 max and they're the same size and dimensions And the P12 max does have a ring as well, that ring is for reinforcement to stop the blades from stretching due to centrifugal forces

[@SianaGearz]

I've been wondering, so the fan blades unfurling themselves is a real problem. You want the tips close to the frame for efficiency, but you can't have them scrape the frame. I think Noctua's solution is LCP resin plus stiffening fillers. But there aren't any PC fans with blades made from magnesium...

[@DLTX1007]

@@SianaGearz LCP resin is indeed the cure. It also resonates less. Why there aren't magnesium? Magnesium would resonate pretty badly compared to polymers. And heavy too

[@gerthddyn]

After watching several seasons of the fan showdown by Major Hardware, where users submit fan blades that he prints, I think I've only seen 1 or 2 beat the baseline Noctua for both sound and airflow, and it is a pretty narrow margin, so I would be doubtful of anyone who says a fan is twice as good and half as loud. I'm pretty sure we are at the point of diminishing returns, and your results are why I just buy Artic P12s. The asymptotic money isn't worth the small amount of improvement for the fan sound level and performance.

[@KarolusTemplareV]

At one point one wonders why other fans exist when a Nidec Gentle Typhoon of yonder years it's still up there with the best and the Arctic Cooling ones keep embarrasing everything in the price/perfomance category.

[@draco10111b]

While the performance might not change much there can still be a big difference in the quality of the components. With Noctua you're paying mostly for the over engineering.

[@jttech44]

@@KarolusTemplareV Those nidec fans are good case fans, but they're not good on coolers or on radiators. The reason other fans exist is that you can certainly engineer a fan that performs substantially better in a given scenario. Also, Noctua fans are very, very quiet for their performance level, and when they do make noise, it tends to be less annoying than other fans. There are, of course, more performant fans. Hell, the old DeltaV's running at 5000RPM perform way better, and they draw 0.8A from a fan header. SAN-ACE and other server suppliers have similar models in that same vein, all are more performant than noctua's desktop offerings. They're also all incredibly loud and annoying. So it depends on what you're designing for. Is it possible to make a fan more performant while being quiet, and do so in every situation? Of course, it's just a matter of throwing money into R&D, which is what noctua and others are doing. It's not going to be rapid progress, it's going to be slow, iterative steps. That and a fan is only part of a cooling solution, and we've seen significant progress in other areas in the last decade or so, and that's likely to continue for some time. Noctua might make the best desktop fans currently, but, they no longer make the best cooling solutions, at least at the moment, so it'll be interesting to see how all of that develops. Thermalright is doing very good work in that space currently, if you're in the market for an air cooler. Arctic is doing very well in the AIO space, and custom loops are better than they've ever been. It's a good time to be alive.

[@gerthddyn]

​@@jttech44 I don't know where you got that Gentle Typhoons are optimized as case fans. Scythe-branded Gentle Typhoons were pretty much the only fan people used on high FPI radiators as they were the only thing out that could push air through them without sounding like a jet turbine. When the reviews of the A12x25s first came out, they were within 10% of the performance/noise at every point. The biggest difference was the better tone of the noise and the lack of the ball bearing clack. If you stand them next to each other, they look almost identical, but you can tell that the blade design started with the GT and was then optimized to the design on the A12x25. The biggest claim to fame for the A12s when they came out was that they were better for a much longer lifetime due to their new materials. When you coupled the fact that all of the Gentle Typhoon supply had dried up around 2016 (I bought my last 16 Nidec GTs in bulk at that time), it was the perfect storm for taking over that market segment. But they were 2-3 times the price of GTs when you bought them in bulk and I've never got a bulk discount for A12x25s. Artec P12s are about 30% less than the Nidec GTs were when bought in bulk and other than the vibration at about 1050RPM, they are the good enough solution for many things. Being at the point of diminishing returns means that it takes a LOT more effort to get the additional performance. Noctua took 4 years to release a black version of the A12. They haven't released an A14. Their limited research budget isn't being spent on the successor of the A12.

[@KarolusTemplareV]

I stopped reading at the Gentle typhoons not being good for radiators and heatsinks. Back then they where all out of stock because everyone who knew of them buyed as many as possible for their watercooling. In fact GT have very good air pressure, better than Noctua's NF12. @@jttech44

[@Weecka2007]

I'm just here for the cat. Fan testing is just a bonus.

[@Broxerlol]

Appreciate all the effort you put into your tests. Keep up the great work!

[@phero6933]

Glad to see someone actually test these. I doubted the initial claims from igors lab and I wanted the fans based off of that info. Thanks for saving me some money

[@tahustvedt]

I design 3D printable RC jets and EDF fans. It's important to minimize the gap between the blades and shroud to maximize efficiency and PC fans always have a huge gap compared to RC EDFs which bothers me. You can even feel the huge amount of blow back through the gap when they run. I usually 3D print a shroud that slips in between the blades and fan housing on my CPU cooling fan which always ends up reducing the temperature significantly. It's so close that it almost touches and I usually have to sand to get it perfect. Even if there's a ring around those blades there's a pretty big gap outside the ring.

[@ska042]

That's one of the things that differentiates the more expensive fans - more expensive ones have higher quality polymers and a smaller gap. Manufacturers need to allow for some creep, lengthening the blades. Due to centrifugal force and thermal cycling, it's almost unavoidable. The better the material, the less of it, and the smaller you can make the gap. Noctua for example advertises a 0.5mm gap for the A12x25, I'm sure they've gotten that as low as they can without compromising the reliability of the fan - keep in mind they need to allow for many years of use when deciding on that gap... To make serious improvements and maintain long term reliability you'd probably have to go away from injection molded plastic. To metal fans, or maybe carbon fiber reinforced plastics, but both of those would push manufacturing cost waaaay high. Edit: the reason the gap is smaller for EDF fans while they're still plastic is in the geometry. They're typically smaller diameter and/or have straighter blades.

[@Elvewizzy.]

@@ska042 And its borderline impossible to offer a warranty on those tolerances when you're dealing with the volumes Noctua is working with. As you stated they'd have to completely revamp the manufacturing if they go with a different material and it will skyrocket the cost for maybe a few C improvement.

[@Karibanu]

I do wonder what aluminium blades would actually cost, if you dropped any "omg metal!" premium. Titanium would be ideal, they'd be absurdly thin & light, but that really would also be absurdly expensive ( not only is it costly to obtain but it's a pig to work with ).

[@ionstorm66]

​@@ska042noctua took years to offer black fans, because they had to verify the black plastic didn't have enough creep to close the gap. That's what the industrial fans don't have as close of a gap because it's a different plastic that is more durable, but has more creep.

[@Rentta]

@@Karibanu There has been aluminum bladed fans in history. None of them were anything groundbreaking really (then again they were made more for the looks instead of performance / tolerances)

[@kevinhsu6239]

Never thought the montech fan could be this good.

[@thelith]

I'm glad i didn't wait for this, those early reviews were too good to be true. It was quite suspicious that only one reviewer got a sample before release.

[@Crunkmaster]

wouldn't be surprised if it was a gimmick from alphacool to get a shitload of preorders from people who know a little bit about pcs and will look at testing but not too much and won't wait for multiple reviews

[@thelith]

@@Crunkmaster whats the point of demolishing their rep for some preorders that can be cancelled anyways tho?

[@Crunkmaster]

yea, it doesn't make any sense, but i can't imagine why they would only send out one set, before preorders go live, to a reviewer that's been known to be sloppy / shady with their testing over the past few years either

[@generalxp8850]

Wow great catch i didnt look at igor's graph close enough. This is why its all ways good to have multiple reviewers as you said. This is why i watch your videos soo informative

[@Corle0ne]

I love your cats and how they're such an integral part of your videos. Sure the contents was already interesting to begin with, but they just add such a lovely atmosphere to whatever you might be looking into. Keep up the good work!

[@jeremyg4650]

I didn't disagree with your testing methods or results but I think noise normalized results are equally important for this type of comparison!

[@rogerwennstrom6677]

Really appreciate the test! Good to confirm that the hype was quite exaggerated. But there were a few flaws: @6:52 no, the Montech fan isn't "slightly higher". The 140x123x29 mm is the dimension of the box itself! The fan itself has a normal size. @8:20 no, the fan blade design isn't completely different to the other fans: the Arctic has the same basic design with an outer ring connecting the fan blades. Didn't you look at the fans? :)

[@jarnom85]

Shows how good those Noctua fans are. Yes, expensive but incredibly good.

[@croatiagunner]

Thank you for saving me 90+ euros! I considered taking mine Noctua's from u12a and from the exhaust, putting them on the front intake replacing my 3 x P12 Bionix, and buying 3 Alphacool Apex for u12a and exhaust case fan. This video and review are exactly what I was looking for!

[@Beulena.]

I wonder what led to such a difference in results, because they are so out of sync that something has to have gone completely wrong.

[@shoppingforleaves3162]

Maybe Igor sneezed during his Noctua test?

[@xsvrrx]

Igor didn't measure heat transfer only airflow to sound

[@WyndStryke]

Specific RPM versus noise level. Will give quite different results.

[@MrPruske]

I remember hearing someone say at CES that this might end up being a build your own fan, I would love a fan with multiple parts that could be swapped out and replace later. Too many good RGB fans in my computers have died heat deaths when the bearing goes out but the lights work perfectly fine or vice versa.

[@killer01ws6]

I watch your channel because I CAN trust your testing results and you go so far to explain, how, what and why of the results... I trust your info here more than from someone else.

[@josephrodgersjogabonito4624]

It’s not the db level to air moved/temp ratio, it’s the good noise profile of the ram. It has a very deep tilted signature and has pretty much zero high pitched noise, which would be evident on the t30. Watch sts’ video’s sound tests

[@spaghettilocomotive]

It would have been super interesting to see the Phanteks T30 fans tested as well, they are regarded as very good for radiators/watercooling, but they are thicker too. Maybe you can do this for a follow up Roman?

[@666Necropsy]

shhh dont tell anyone about these secrete fans

[@Frozoken]

Go to hw busters, he's tested both. Spoiler alert, the t30 demolishes every fan including this one in noise to performance tests. Oh and in cfm the apex is 2nd or dead last in every single noise increment without fail 😂. Don't get these as case fans to say the least

[@666Necropsy]

@@Frozoken ya they rock. got 3

[@moondark666]

I would have loved if the Phanteks T30 was included in this test. That said, an amazing video as always.

[@sergeys2505]

Have a speedy recovery, man, thanks for your tests 👍

[@war4peace1979]

@6:54 - Correction needed: You are reading the PACKAGE dimensions from the Montech fan box. That is the box size, not the fan size. The fan itself is 120x120x25mm. The box which the fan comes in is 140x123x29mm.

[@c.sec73]

I've got 10 of these pre-ordered to replace my NoiseBlocker's.... Thank you! Much appreciated, you've saved me ~ 300 EUR :)

[@jakedaynes-lmg]

6:40 - 10dB is double perceived sound, but 10-fold the actual sound pressure. Awesome stuff mate!

[@HanmaHeiro]

Thank you for the review! I've been waiting for months!

[@kofeyh]

Temperature is in Celsius. For a second I thought the video was referring to the temperature in Kelvin (not the difference in K) so that threw me for a moment. Other than that.. fascinating. Another instance of a metric aberration and people going "huh" vs "something is very very wrong". Noctua have spent years chasing fractional differences. Anyone claiming "twice as good" is essentially breaking our understanding of physics given how fractonal the differences have gotten.

[@MrFilip121]

I believe the non max Arctic P12 uses a different bearing which should make it quieter at lower speeds.

[@fantabuloussnuffaluffagus]

After some personal experience with Arctic fans, I'm a huge fan of Noctua. Arctic fans are cheap, and it shows.

[@253637zero2]

@@fantabuloussnuffaluffagus 90% the performance of noctua for 1/3 the price. Yeah I'll go with Arctic because I don't the kind of money to just buy the best of everything.

[@fantabuloussnuffaluffagus]

@@253637zero2 I watched the reviews, bought the Arctic product, 2 of 6 fans were broken on arrival, then I bought one of each of the top contenders for comparison, then I bought Noctua. I wish I'd done that in the first place. IMO Noctua is value for money. Especially the 3,000 PRM Industrial fans, they move enormous amounts of air when spooled up. If you find a fan that moves even 2/3 the air of a NF-A14 3,000 rpm industrial for 1/3 the price, I'd like to know what it is. The P-14 moves less than 1/2 of the the air moved by the NF-A14 3,000 rpm industrial 72 vs 158 cfm. Yes, the big Noctua is loud running flat out, but if my CPU is overheating, I don't care, I just want it cooled, and 99% of the time, they are turning slowly and are inaudible.

[@linuxsux41]

How unfortunate, I was hoping that they had done some dev to make their fan bearings quieter while running a faster proprietary motor or something that made their claims possible. I'm still interested to see rebuttal from this to show how those magic numbers actually happened, if it's like Nvidia cherry picking settings to show a 4x framerate boost when the actual uplift is nowhere near that it would make sense but more than likely just never happened. Thanks for the review!

[@justinbouchard]

yay! i love my arctic fans with my arctic liquid freezer so i'm hoping they perform well in your testing lolol oh!! also! do you have any opinion on blowing through a radiator as opposed to pulling through a radiator?

[@PyroVulpes]

Thank you for this well needed sanity check. Some of the claims surrounding the new Alphacool fans were a bit hard to believe. Glad I didn't jump straight into placing a pre-order for them.

[@infango]

when look at price to performance Arctic P12 is a beast with such a slim differences, after looking at Igor's review APEX at 2000 RMP has same noise level as NF-A12x25 at 400 RPM like Mulder "i want to believe" ;)

[@rasmuspedersen2001]

Can it be a difference between the power and non power variant of the apex? wouldnt expect it but thats the only discrepancy i can see, even tho igorslab states it should only be the rpm that is different.

[@duraiden]

Thanks for the review. I was ready to replace my Noctuas based on the hype but I’m happy that it won’t be necessary.

[@deanwhiteside6339]

To define it as a "metal fan" i would expect the fan to be made of metal not just the surround. Its a plastic fan with a metal surround

[@Mickulty]

Thanks for this, the results from igorslab seemed a bit too literally incredible. Hopefully GN can get one and try it in their fancy hemianechoic chamber.

[@SaperPl1]

I feel like this is similar situation to thermaltake engine 17/27 - the prototype/concepts for those were really interesting, but achieving the gaps required for connection between spinning metal fan and the base was unrealistic for mass production, and it looks like in final production the metal fan is just a selling point. If that's the case here with those fans that the prototypes again show there is a possibility to do something like that, but it's not viable to manufacture those at scale, then it's a bummer...

[@hectororestes2752]

I still think about those as the holy grail of low profile coolers. I remember, as you said, they did achieve what they were trying to achieve but only in lab (sandia lab), the mass produced product didnt achieve the tolerances to take advantage of the effect. I wonder if the same thing happened here, like some samples do achieve the numbers but it cant be mass produced at reasonable cost.

[@dextrodemon]

interesting, perhaps there's some oddity caused by the max speed of the stealth being 3000, like some of igor's numbers were scaled wrong, tho he did test at set rpms and everything. very odd.

[@MrDrTheJniac]

Igor's Lab uses a testing machine that simulates a radiator by using pressure. From my understanding, it basically blows air against the airflow of the fan, which, while perfect for measuring airflow, fails to account for the fact that most of the noise of a radiator fan comes not from the fan but from the airflow through the fins of the radiator.

[@kevinerbs2778]

That view at 8:10-8:20 are just like Bitfenix 200mm fans are put together they were really nice high quality well balanced between CFM & air pressure for 200mm fans. They also had the best possible mounting holes as they have about 3-4 different sets on each corner.

[@tns6862]

Saw that article and basically went "Too good to be true, probably is."

[@Jeffrey_Wong]

Disappointing, but expected. Appreciate you taking the time to review these! Would've been ready to pay out the nose for these if it turned out the claims were true.

[@ChrispyNut]

The different geometry of single-tower heatsink vs 240mm radiator will lead to a significant difference in noise, potentially having greater impact than number of fans. Heatsinks are [typically] quieter by design, due to the larger gap between fins, relative to radiators, with additional variances coming from the shapes of the fins in the stack.

[@jepulis6674]

How you test also matters quite a lot. Put the tower inside a case, measure 1m away and the difference between 1500rpm and 1000rpm in dB is a lot smaller than doing a silly test on a table 15cm away from fans. Would be nice to see full system test perf per euro too. 5 pack of p12 max costs about the same as one of those premium fans. Put 4 case fans and 1 tower, vs one noctua fan for fair comparison :D

[@ChrispyNut]

@@jepulis6674 But he covers. When I thought he'd mentioned the geometry, he instead went for number of fans, but that's less important (especially at low speeds) and immediately obvious to people, whereas the importance of airflow disturbance on noise is less obvious and less intuitive.

[@mikezappulla4092]

@@jepulis6674you want him to test all the different types of cases? They all have different airflow and static pressure.

[@rusTORK]

Most of us probably saw Noctua booth when they show new fans and heatsinks with their roadmap. Noctua invest HUGE amount of money in R&D of a FAN. Materials, design, prototypes, tests of prototypes, even shut down some dead end paths! If someone beat Noctua FAN then he eather: VERY rich or VERY lucky.

[@m0r73n]

Finally we will see some more tests on these fans.

[@cyphre]

One of those things where samples could have very small differences that would make a large difference in how quiet they can be. Interesting test!

[@fracturedlife1393]

Will keep an eye for follow ups from Igor. Usually on there a few times per year for RBE/MPT tips. I seen Jayz reviewed the fan too, will have to get a look. Hopefully hwbusters and GN also have ones.

[@tuomashautamaki2258]

Jayz didn't do any testing. That video is just unboxing and twiddling the fan in the hands.

[@muhdkamilmohdbaki7054]

Chiq was very involved during the testing and even vocalizing its "concerns" at least a couple times😹. Anyway, informative good contents as per usual.

[@jjlw2378]

This video was a GIANT PROMOTION for Montech Metal Fans! 30mm fan with Noctua performance at Artic Fan prices. Seems like a solid product.

[@mateuszkwietowicz2470]

What about max rpm? Alphacool fan is rated at 3000 RPM, the Artic P12 Max is rated at 3300 RPM? Couldn't you compare the max rpm efficiency vs volume?

[@ruikazane5123]

Curious if the fan is tested with them pulling the air from the radiator, instead of push. Would it even make a difference?

[@SrApathy33]

From what I've seen it barely makes a difference, Jayz2cents and Linus also say that. Pulling air through instead of pushing makes for much easier radiator cleaning. Just wipe off the radiator instead of toothbrushing all the fan blades.

[@WyndStryke]

Testing for a specific RPM might be the difference possibly? Different fan blades would have a different angle of attack (airflow versus static pressure). The blades with the highest angle of attack would move more air at a specific RPM, but would probably also be noisier and have a lower max RPM. It might be helpful to test against either a specific input wattage, or to test to a specific noise level.

[@SpoiledBadgerMilk]

Your the best man. Love your content it never disappoints.

[@royalrepublican3480]

yes the loudness is similar but at what frequenzy? is it soft humming or a annoying mosquito?

[@WatchMysh]

According to STS the Alphacools emit a low hum and almost no mids

Like always great effort and great test. However I can't get along without saying that these are not Kelvin values :) These are Celisus God forbid

[@OTechnology]

I knew it was an impossible that making a fan frame out of metal would improve the noise to airflow performance lol how does that even make sense when the component that interacts the most with the airflow and noise are the blades.

[@reedlaww]

As an aside - why verbally reference temps in Kelvin when it’s the same scale as Celsius and it’s even labeled as C on the diagnostic tool lol

[@Petch85]

It would be nice with a link in the discribtion to the article/video you mention to make it easy for us to find it and be sure it is the same.

[@RetroBerner]

just google Alphacool Apex Stealth Metal Igor's lab

[@stevewatson6839]

A job for the GN crew, methinks. There will always be rogue results and here I lean to Igor's Lab being a rogue sample. This'll happen to every reviewer if they are in the game long enough. There are no 100% perfect products all the time. Let's wait for a dozen more samples. I think Alphacool have done well with the silencing, and kudos to them for that. If the performance were there or there abouts I'd at at least sample them myself; even not caring for that sort of pigtail daisychaining at all.

[@SianaGearz]

GN didn't work out fan test methodology yet. They have this big tank from Taiwan but they really haven't been testing fans.

[@Ouss-2077]

I think you should've included a sound demo as readings along don't give the full picture. There's a big difference between db readings and how it actually SOUNDS as there are multiple frequencies that may or may not be present even between two identical readings. The resutls published by Igor and even Alphacool are far from reality though.

[@maozedowner5915]

The Metal Fan's fan frame is many times thicker than other fans in the video, it reduces the fan area by a ton. While every high end fan manufacturers are doing their best to minimize the frame and have the fan blade sit as close to the frame as possible, the Metal Fan wasn't. That should be a good sign that the claims are bs.

[@zoltancsibykilyen6030]

Hi. Thanks for the intresting test, just want to remark it is not Kelvin it is Celsius. 0 Celsius is 273 Kelvin. Nice test though. Wish you the best.

[@cannesahs]

Thank you for the test. Goes in line with Igor's reputation

[@jacksun_jack]

what about the phanteks t30 fan?

[@25MHzisbest]

Is it all about the 3000RPM? The P12 MAX can do 3300, worth comparing them?

[@TheSkiddywinks]

10dB is a tenfold increase, no? 3dB is a doubling. EDIT: Just looked it up, I'm right and wrong; the actual air pressure measured is as I posted, but your subjective experience is different, and a 10dB increase is indeed a doubling of perceived "loudness".

[@FreakyAngelus]

I was in doubt for a minute as well.... then realized this is the only comment section with people searching things up and helping! Thanks for adding the information!

[@DuyLeNguyen]

3dB is a doubling in sound pressure level (SPL), but dBA (dB, A-weighted) takes into account average human 'perceived' loudness which does not scale directly to sound pressure level. I think in general 10dBA is about a doubling in perceived loud ess

[@SianaGearz]

Any talk of "doubling" in terms of a fixed dB difference is bogus. You simply can't compare noises like that. The only thing that could be reasonably said about a noise that's 10dB louder is that it's 10dB louder. Unhelpful but true. Indeed you need 10 fans to make up a 10dB difference.

[@Del_UK]

Interesting. When I saw the "metal" fan, I was thinking along the lines of the chassis and blades being made from metal. Really disappointing to find out that it is only the chassis. Noctua have a new fan coming in 2024, will that beat their old fan?

[@eliadbu]

Per their road map, it's only 140mm fan. And probably it will be big upgrade over their existing 140mm fan which is quite old in design and needs a replacement.

[@tactical-bucket]

Thanks for the honesty. I love hyped up cash grab products.

[@puppet_master]

When possible, I prefer a noise normalized benchmark, but it's still an interesting test. Thank you.

[@rustler08]

Noise-normalized benchmarks can really only be taken seriously if you have an anechoic chamber, or at least an attempt at replicating one.

[@puppet_master]

@@rustler08 depends how close the mic is, but if doing repeat tests over multiple days I agree

[@SianaGearz]

@@rustler08 Why? Same conditions, same noise level is same noise level.

[@RBMDragon]

Thank you so much for the results.

[@lussekatt6425]

Damn, good testing. Have to wait and see other tests too. Shame, because they did look very nice and would be sweet if they did end up performing greater and still be very quiet.

[@Filippo-su2xv]

if I understood correctly, given my limited English, only by increasing the fans up to 2000rpm you lower the water temperature by ~9° c?

[@tetryds]

Since it is quieter, have you tried normalizing for noise? Afterall it's what matters the most for an end user and if it spins faster you might get better results. Unlikely to be similar to igor's, but I would still like to see this test.

[@ashliehiggins]

I'm not happy they aren't performing as expected, but I'm glad they are as I've just bought some T30s.

[@wjajr84]

wonder about golden sample of the fan in igors lab? that could also bode badly for the fan. Like wide variance in quality.

[@TheWolfLoki]

A small correction in an otherwise great exploration of Alphacool (and Igor)'s claims. @6:51 Dimensions for Montech Metal Fan are the SAME as regular 120mm fan, they are 25mm thick, packaging just states that the PACKAGING is 29mm thick, you can verify this by packaging also being 140mmx123mm, definitely not the dimensions of a fan which is square, and certainly the dimensions of the cardboard box. They perform better (in some metrics) than even Noctua which is surprising but it is NOT due to them taking the route of thicker fans like Phanteks T30/D30 (30mm), EK Meltemi (38mm), LianLi Unifan V2 (28mm), or the plethora of 26/27mm fans that trade compatibility for performance, this makes the Montech's an insanely good contender, especially given the price point.

[@agoogleuser7899]

Fan decoupling isn't a new thing. There have been rubber isolater sold for fans. I have tried separating the fan from the case and have not noticed any benefit; although I did not measure the sound.

[@forbiddenera]

You should do noise normalized testing - bring them all to the same dB level and see which cools the best at thatdB regardless of RPM - realistically I'd guess most tune their fans to run as fast as they fan without being too loud, at least I do

[@forbiddenera]

IMHO comparing fans at the same RPM is a flawed testing method. Each fan is going to flow differently at different RPMs and will have different efficiency and flow curves versus RPM and static pressure.

[@forbiddenera]

I'm almost disappointed 😢 you're usually great and on the ball for this stuff but missed it here.. I mean that's also why some fans are good for radiators but not case and vice versa, fan blade area, count, blade thickness and pitch..it may very well need more rpm to perform better yet still have the same or lower noise level..

[@AndrewB23]

​@@forbiddenerahow is it flawed, if a fan pushes more air even at low RPMs it's a better fan. Your method for setting it based off db would be worse because you could literally buy a 3 pack of the same fan with one being louder than the other two even at the same RPM

[@DavidFregoli]

finally we know why he left Caseking, they wouldn't let him grow a beard

[@BSEUNHIR]

Finally! The actual reviews! :D Back when both alphacool and igor very proudly announced the testing, I knew there was clownery afoot. This is not a simple mistake imo, it's straight up fraud imo. Stupid maneuver from both parties. Why did they do it? To sell some preorders? Are they that desperate for money? Wonder if there's gonna be legal repercussions. Also, I just can't stand igor and this affirms my position on him. He acts as if he'd "eaten wisdom with a spoon", but this kind of showing amongst other things disqualifies his outlet for any serious considerations. Lastly, a bit sad these aren't magic. Would have loved to replace some fans in my system. Back to waiting for a next gen 140mm white fan from noctua. Just another few quarters.... right?

[@3PurpleSquirrels]

One thing to note the NF-A12 max fan speed is 2000rpm and the Alphacool Apex Stealth Metal 3000rpm so that would handicap the Apex Stealth. But Igor's testing still looks odd.

[@benielishackove2686]

Do you think that the metal case holds cooler air (over time)? because I think the metal should be cooler than the surrounding air because of the air flow...

[@andydbedford]

I think Steve@ GN needs to get hold of these and test them too, I dont doubt your testing, so someone with a very good pedigree in testing hardware who can confirm your testing will shine a light on what the heck is going on over at Igor's Labs, because something is very clearly off here.

[@MrFilip121]

Google cybenetics. They tested this fan in $62k chamber that Steve has. The results are more or less as in here.

[@UNVIRUSLETALE]

Arctic as always great value and good performance, you can do 2 240 rads for the price of a single noctua

[@71janas]

We need GN to take a look at those new metal case fans!

[@lina4469]

now, could a different blade design help the cooling performance?

[@XX-121]

yeah i think the only reason it was testing better at all was from the blades being connected to that outer ring, where as normal fans have a gap which let's some pressure escape. but i see you did mention that in the video. no way a metal frame is going to make that much difference.

[@flandrble]

Noise Blocker fans have done that for the last decade and are actually comparable to A12X25 and Nidec Gentle Typhoons.

[@retrosean199]

Is it possible that Igor's test actually was correct in terms of movement of air but somehow the pressure the fan exerts against the radiator just does not work out as well as the Noctua fan? It's very quiet, but maybe something about the air movement pattern is less effective.

[@Nienormalny]

Hey. Would be nice to add a test with constant noise. For example 40db and 45gb, then to that value set the rpm's accordingly. That might be quite useful.

[@PainterVierax]

even when using dbA instead of db, the noise level is still quite subjective as our perception is really frequency dependent and it changes a lot depending on individuals and circumstances.

[@Nienormalny]

@@PainterVierax 100% agree. Yet it provides fair efficiency comparison between all the of the fans maintaining constant loudness at the same time. Might be informative.

[@Wormenstain]

Might be good thing to check RPM with external tool and not with motherboard rpm sensor. Some fans just report wrong values. Arctic bionix F series example gives wrong values for my MB, it hass max RPM 1800 but MB sensors thinks it spinning 3000RPM.

[@Olimorveu]

It's also in dB(A) right, which is logarithmic in scale? So the difference in Pascals that he would have had to measure from both fans is enormous.

[@lrmcatspaw1]

When it comes to audio levels, Maybe I can see the different pitch of the two fans dissipate faster or slower at different distances? But the airflow part looks weird. I remember when the scythe gentle typhoons came around and had very good CFM due to the blade design (made sense based on how the blades would force air more evenly downwards) so a bit of extra performance was expected. Same as the Noctua fans with the same design philosophy. Same with the arctic fans (and damn these are good value especially the 140mm version). But almost 50% more cfm and 20 db down?! It does sound like an error got somewhere in there. EDIT: There is one thing that could make sense here: part of the air in all the other fans except the Alphacool is blowing to the sides (still cooling everything around it mind you) where the alpha cool pushes the air more concentrated thru it's ring increasing air flow, but not increasing heat transfer between the air and the radiator as it is already removing as much heat as it can). Think of it as the other fans cooling also the outer areas of the radiator where the alphacool is going thru more the middle of the radiator. Depending where you would have the CFM testing device, the air might escape to the sides and not register (as well as lower air pressure) while still cooling more overall area. Does that make any sense? :D