Best cooler/case/fan combo for high-performance quietness?

[On Holiday]

Looking to cool an i7 9700k and a 2070 Super.

Think I'll go for the Dark Rock Pro 4 but open to alternatives.

What's the current go-to case and fans for "quietness" but still having a good amount of cooling performance?

Advice and experiences appreciated

 

[Quartz]

I have an i7-8700 cooled by a Dark Rock Pro in a Silverstone FT05.

 

[ttaskmaster]

What's the current go-to case and fans for "quietness" but still having a good amount of cooling performance?

The fan fanboys are generally split between Noctua, who have a long standing reputation in this matter, and Arctic, who are almost as good but cheaper and of debatable build quality.
Really though, you'd need to see reviews of all the different fans on this exact cooler, and even then the resonance of your particular case may differ.

Try the fan(s) included with the DRP4 first and see how you get on...

 

[On Holiday]

The fan fanboys are generally split between Noctua, who have a long standing reputation in this matter, and Arctic, who are almost as good but cheaper and of debatable build quality.
Really though, you'd need to see reviews of all the different fans on this exact cooler, and even then the resonance of your particular case may differ.

Try the fan(s) included with the DRP4 first and see how you get on...

Ohh yeah, I'd be going for the DRP4 fans as that's ultimately what it's had research + design based on, but I was looking more for case fan suggestions. Which I think Noctua S12A seem to be the leader for at the moment?

 

[EsaT]

Which I think Noctua S12A seem to be the leader for at the moment?

NF-S12 is "downhill in tail wind" design fan.
Any obstructions like meshes or dust filters will drop its airflow fast.
That's why it's also very quiet per RPM: It just can't actually do much work/push that air.
(though its design is good for fooling impeller anemometer)

For usual case you want something well balanced and capable to moving air regardless of impedance.
Arctic P12 is hard to beat in that especially considering its price.
https://www.overclockers.co.uk/arctic-cooling-p12-pwm-pst-black-fan-120mm-fg-04h-ar.html
https://www.overclockers.co.uk/arctic-cooling-p12-black-case-fan-120mm-fg-04g-ar.html

 

[PieEater]

Just a quick note about "debatable build quality" on the Arctic P series PWM fans you get 4 years more warranty with the Arctic fans than you do with Noctua fans (10 years vs 6) so Arctic have a lot of faith in the design and build qualities of these fans. They are not perfect but on a price to performance level they are undeniably unbeatable.

 

[Something Else]

I've always used akasa Apache black fans in my pc.

The manufacturer stats are hard to beat imo and they're around a tenner per fan.

Currently have 2 140mm intakes and a 120mm exhaust and the only fan I hear is the wraith prism.

They have slightly better stats than the Arctic cooling fans everyone likes but they are more money.

 

[ttaskmaster]

Ohh yeah, I'd be going for the DRP4 fans as that's ultimately what it's had research + design based on, but I was looking more for case fan suggestions. Which I think Noctua S12A seem to be the leader for at the moment?

No, the S12 is a case fan, designed for moving high volumes of unobstructed air.
On heatsinks you need something that will push air through restrictions, so if you're going for Noctua that generally means one of the F12 series: https://noctua.at/en/products/fan

NF-S12 is "downhill in tail wind" design fan.
Any obstructions like meshes or dust filters will drop its airflow fast.

Yawn...
Yes, because it's an airflow-centric case fan. It's not designed for restricted spaces.

Just a quick note about "debatable build quality" on the Arctic P series PWM fans you get 4 years more warranty with the Arctic fans than you do with Noctua fans (10 years vs 6) so Arctic have a lot of faith in the design and build qualities of these fans.

They might well have faith, but I'm looking more at how often I see user reviews where their fans have broken (usually the frame), or developed bearing noises.

 

[PieEater]

They might well have faith, but I'm looking more at how often I see user reviews where their fans have broken (usually the frame), or developed bearing noises.

Not heard of those issues myself, I have 7 of them and am more than happy to recommend them from personal experience rather than bias. Don't get me wrong if I had an unlimited budget I probably would have gone with Silent Wings 3 but £50 to outfit my case and coolers with the Arctics vs £140 for the BeQuiets for very little if any benefit in performance or noise levels just wasn't justifiable. In the unlikely event the frame randomly decides to crack on its own or the bearings decide to give out at some point in the next 10 years I'll just swap it out for a new one courtesy of the warranty.

Bringing things back on track these are worth watching regarding case choices.

 

[EsaT]

No, the S12 is a case fan, designed for moving high volumes of unobstructed air.
On heatsinks you need something that will push air through restrictions, so if you're going for Noctua that generally means one of the F12 series: https://noctua.at/en/products/fan


Yawn...
Yes, because it's an airflow-centric case fan. It's not designed for restricted spaces.

This is the amount of impedance which brought NF-S12's airflow measurement to same level with other fans:
https://silentpcreview.com/wp-content/uploads/2007/05/hwa_box_setup.jpg
Pretty darn sure any actual case has more impedance.
So wouldn't use them elsewhere than exhausts with some more pressure capable fans as intakes.

Again NF-F12 is overall one of the crappiest fan I've touched in 20+ years of hobby.
Motor is same insane vibrating roller as in preceding NF-P12, but instead of its OK good noise profile, marketing design makes also noise profile restless.
In fact Noctua's own data confirms that restless noise signature!
Acoustically good fans are designed to avoid such frequency spikes.
Claiming anything else is same pseudoscientific BS creationists, oil companies etc use.

 

[ttaskmaster]

This is the amount of impedance which brought NF-S12's airflow measurement to same level with other fans:
https://silentpcreview.com/wp-content/uploads/2007/05/hwa_box_setup.jpg

Hmm, nice picture....
But if you actually read the article it's from, you'll see that every single fan they tested with this second set up (ie, the one you have pictured there) dropped about 10CFM from it's MFR's specs.

Of the Noctua, they also note:
"It does not push significantly more air for a given RPM than other fans. In this regard it is ordinary. But most importantly, it remains a very quiet fan, and comparison with both the Nexus and Scythe SFF21E 120mm fans, the Noctua still seems to have an edge in airflow at lower speed."

Pretty darn sure any actual case has more impedance.

And it will still impact whichever fan you choose, as demonstrated by the article to which your own linked photo shows.
What is your point, here?

So wouldn't use them elsewhere than exhausts with some more pressure capable fans as intakes.

The S12 is specified as being for low-mid impedance application, where low noise is a requirement.
If you want high airflow with better static pressure for medium impedence application, the A series and P series are better options.
Again, what point are you trying to make by saying you wouldn't use fans in ways they're not intended to be used in the first place?

Incidentally, your lovely half-linked article actually shows the Noctua S12 doing rather well for itself in a restricted application, namely a Scythe Ninja, especially at lower speeds!

Again NF-F12 is overall one of the crappiest fan I've touched in 20+ years of hobby.

Yes, yes, we all know you personally don't like Noctuas for whatever reason, MartinsLiquidLab, car engine noise in freefall and all the other half-baked ideas from ill-researched concepts and buzzwords from articles you didn't read properly.... But for the rest of us, Noctuas remain very good options.

In fact Noctua's own data confirms that restless noise signature!
Acoustically good fans are designed to avoid such frequency spikes.

Firstly - From the very link you just posted:
"Moving over the stator guide vanes, the fan blades create pressure pulses. This type of rotor-stator interaction significantly contributes to the fan’s overall noise emission. With a uniform distribution of the stator vanes, the relative geometrical constellation of the rotor blades and the stator guide vanes is identical each time the rotor moves over the next stator vane. This means that each time a stator vane is passed by a rotor blade, the same pressure pulses are created. Acoustically, these periodic pulses lead to noises at distinct frequencies and manifest themselves as frequency spikes in the fan’s acoustic spectrum, giving it a more tonal and distinguishable character. To avoid this phenomenon, the NF-F12’s stator guide vanes are set out in Varying Angular Distance (between 31° and 37°). Due to this measure, the geometrical situation between the rotor and the stator parts is slightly different each time the rotor moves over the next stator vane, which results in aerodynamically different pressure pulses and hence spreads the noise emission over a broader frequency range. Producing a less tonal and more broadband acoustic profile, the VAD stator guide vanes make the fan more agreeable to the human ear and let it blend into the background noise more easily".

"Secondly, Vortex-Control Notches also help to make the fan’s acoustic profile more pleasant to the human ear. Due to the Vortex-Control Notches, trailing edge vortices are split up into several smaller vortices, each having slightly different tonal characteristics. This has a similar effect as the Varying Angular Distance of the stator guide vanes insofar as the noise emission is spread over a wider range of frequencies and frequency spikes are reduced. As a result, the fan’s noise pattern is perceived as much less annoying and more agreeable to the human ear".

In short, every fan has a 'restless noise signature'. Noctua's own data (which is mostly just words and pretty pictures, rather than actual data) actually shows how they reduce it.
So again, what's your point?


Secondly - Doing a search for Noctua and "restless noise" yields quite a few results.... and every one of them is a fairly homogenous forum post on the likes of OCUK and LinusTechTips by some bloke called EssaT, ranting against Noctua like some angry little puppy. A couple of people even asked for substantiation of these assertions, but the poster declined to respond.
No-one else on the internet appears to support or even share this view of yours....

Claiming anything else is same pseudoscientific BS creationists, oil companies etc use.

Ah, so it's a conspiracy theory by the secret Lizard People, to make us buy their evil Noctua fans?
Oh well..... So long as they're quiet!

 

[PieEater]

They might well have faith, but I'm looking more at how often I see user reviews where their fans have broken (usually the frame), or developed bearing noises.

I have looked really long and hard for any negative reviews of the latest P series Arctic fans and honestly cannot find anything that would back up your claims other than the 1 star Amazon customer reviews where interestingly one person said the fan frames were broken (on delivery) and a few complain about noise, however the fans still have a really good overall rating of 4.7 stars out of 5. If trawling Amazon for negative reviews of products is really where we're at here's an interesting one I found on the Noctua S12.

"This fan is good at being quiet and that is the only good thing I can say about this fan.
It is a terrible exhaust fan that has been completely out performed by a £5 arctic static pressure fan.

Results of Realbench CPU stress test with each fan:

NF-S12A @ 100% - average CPU temps - 59C

Arctic cheap-o £5 fan @ 75% - average CPU temps - 54C

I've been using Noctua for years. This is the first of their produts that I cannot recommend, it is too expensive and simply does not perform well when compared to cheaper alternatives."

 

[ttaskmaster]

I have looked really long and hard for any negative reviews of the latest P series Arctic fans and honestly cannot find anything that would back up your claims other than the 1 star Amazon customer reviews where interestingly one person said the fan frames were broken (on delivery) and a few complain about noise, however the fans still have a really good overall rating of 4.7 stars out of 5. If trawling Amazon for negative reviews of products is really where we're at here's an interesting one I found on the Noctua S12.

Google actually, which includes various Amazon sites from different countries (not all reviews show up on .co.uk), Overclock.net, a couple on Reddit, and so on.
I saw several with broken frames, one or two where the blades snapped off while in use, a fair few that were just DOA (as many as five in one delivery) and a couple with duff bearings, in addition to more than just a few complaints about noise.
Enough that I still am not convinced enough to try them out.

 

[PieEater]

Google actually, which includes various Amazon sites from different countries (not all reviews show up on .co.uk), Overclock.net, a couple on Reddit, and so on.
I saw several with broken frames, one or two where the blades snapped off while in use, a fair few that were just DOA (as many as five in one delivery) and a couple with duff bearings, in addition to more than just a few complaints about noise.
Enough that I still am not convinced enough to try them out.

OK I'm happy to leave this here knowing your negative bias towards these fans is based primarily on feedback from retail stores rather then actual trusted technical review sites. You're not stupid, you know full well the typical purchaser leaving feedback on a retail site will be reporting issues including damages in transit (broken or damaged frames / blades), and that they probably won't have the techcical knowledge to identify whether perceived noise is actually a fan or a mounting issue, or even how to control fan speed. You know very well that you can find the exact same issues you are assigning to the Arctic fans reported on any other manufacturers fans if you look hard enough, seriously that review of the Noctua fan I posted was the first negative review I read on Amazon. You have your own reasons for disliking the Arctic fans but I don't think it's fair to warn other people off them when you have no personal or technical reason to do so other than hearsay.

 

[EsaT]

Incidentally, your lovely half-linked article actually shows the Noctua S12 doing rather well for itself in a restricted application, namely a Scythe Ninja, especially at lower speeds!

In short, every fan has a 'restless noise signature'. Noctua's own data (which is mostly just words and pretty pictures, rather than actual data) actually shows how they reduce it.
So again, what's your point?

Those first versions of Scythe Ninja are/were exceptionally low impedance heatsinks certainly working with any fan.
They were designed as much for passive cooling as with fanned use and passive use is how many people utilized those.
(later models have tighter fin spacing)
Case with typical filters, stamped meshes and indirect noise escape paths has likely more impedance.
Antec P180-serie cases certainly had lot more impedance without modifying filter covers.
And Fractal Design copied those horribly restrictive filter covers to Define R.
(now case makers are again achieving intake restrictions by sticking solid glass panel in close proximity with front intake fans)


Good fans avoid design creating frequency spikes in the first place.
SPCR started noise spectrum measurements in fan reviews and spiky frequency response corresponds well with perceived sound being worser.
And it's precisely those NF-F12's "stator guide vanes" parallel with blade trailing edge, which create frequency spikes by disturbing airflow at complete length of trailing edge at once!
SPCR covered that design detail years ago in general fan design article.
https://silentpcreview.com/anatomy-of-the-silent-fan/

Now take a note at which angle NF-A12's motor support struts intersect blade trailing edge...
As complete reversal at 90 degree angle instead of being parallel.
If NF-F12's design was such good, why such complete change of fan design?
Literally from every single fan Noctua has had before?

 

[ttaskmaster]

OK I'm happy to leave this here knowing your negative bias towards these fans is based primarily on feedback from retail stores rather then actual trusted technical review sites.

And what does Reddit sell? What does Overclock.net sell? Since when have they been retail sites?
I don't need a technical site full of technical experts to technically explain to me in a very technical manner the technical reasons why stuff arriving broken is technically not good, whichever way you look at it...

As for "negative bias" - I have nothing against the Arctics. Just against the people who are trying to slate Noctuas over some unsubstantiated claims.
In truth, I'd not use either of them for my current build... which is why I don't.
That you would snobbishly dismiss someone's review purely because it's not on a 'trusted technical review site' is more telling of your own biasses...

You're not stupid, you know full well the typical purchaser leaving feedback on a retail site will be reporting issues including damages in transit (broken or damaged frames / blades)

I've not heard of it happening with any other fan, though. I tend to order 12-18 fans at a time and not once had any of them arrive busted or DOA, nor seen it mentioned in any user reviews.
So when I was considering trying them out and saw several different users report this with the Arctics, it stood out.

and that they probably won't have the techcical knowledge to identify whether perceived noise is actually a fan or a mounting issue, or even how to control fan speed.

Again, that's quite a snobbish assumption on your part, given that several technically versed members on here also buy through retail sites... As is, several of the other reviews speak of altering fan speeds in BIOS vs software, some while overclocking, so I would presume they at least have some idea what they're doing.

You know very well that you can find the exact same issues you are assigning to the Arctic fans reported on any other manufacturers fans if you look hard enough

Never heard of any other fans arriving broken or DoA like that...

seriously that review of the Noctua fan I posted was the first negative review I read on Amazon.

Yeah, I saw the page. It seems to be the only negative one, too...

You have your own reasons for disliking the Arctic fans but I don't think it's fair to warn other people off them when you have no personal or technical reason to do so other than hearsay.

I'm not warning people off them.
I'm merely giving my reasons why I'm not sold on them, along with taking the **** out of EssaT's repetitive puppy rants and challenging his unsubstantiated assertions.

Those first versions of Scythe Ninja are/were exceptionally low impedance heatsinks certainly working with any fan.

And when tested against a bunch of other fans, the Noctuas did very well despite your predictions.
What is your point?

Case with typical filters, stamped meshes and indirect noise escape paths has likely more impedance.

And so other such fans will also drop in accordance with that impedence, as proven by your article.
So what is your point?
So far, all we've got is that Noctua fans are the worst ever, because they behave like every other fan does... and several articles to prove so.

Good fans avoid design creating frequency spikes in the first place

The spikes are created by having struts to hold the fan propeller. Every fan does this. It's the same reason why helicopter rotors make noise.
This is your cue to post some fans which do not create any frequency spikes...

SPCR started noise spectrum measurements in fan reviews and spiky frequency response corresponds well with perceived sound being worser

Which is exactly what your Noctuer link said, whereupon they then explainer how they reducer this.

And it's precisely those NF-F12's "stator guide vanes" parallel with blade trailing edge, which create frequency spikes by disturbing airflow at complete length of trailing edge at once!

But they're not parallel... That's the whole idea of variable angle vanes as an acoustic control method.
This, along with Noctua's vortex shedding methods, is even supported by the SPCR article you linked to... The very things you're accusing Noctua of are utterly disproven by the supporting links you post!

Now take a note at which angle NF-A12's motor support struts intersect blade trailing edge...
As complete reversal at 90 degree angle instead of being parallel.
If NF-F12's design was such good, why such complete change of fan design?

NF-F12 is designed to reduce turbulent airflow and is purely about forcing air through a space as straight as possible.
The A12 is more of an all-rounder, using different designs for a different application.

I'm starting to wonder if you ever actually read the articles you post... and whether you'd actually read something if I posted it for you:
https://noctua.at/en/nf-a12x25-interview-lars-stromback

If you do ever read that, I look forward to hearing what you misinterpret from it...!

 

[EsaT]

The spikes are created by having struts to hold the fan propeller. Every fan does this. It's the same reason why helicopter rotors make noise.
This is your cue to post some fans which do not create any frequency spikes...


But they're not parallel... That's the whole idea of variable angle vanes as an acoustic control method.
This, along with Noctua's vortex shedding methods, is even supported by the SPCR article you linked to... The very things you're accusing Noctua of are utterly disproven by the supporting links you post!


I'm starting to wonder if you ever actually read the articles you post... and whether you'd actually read something if I posted it for you:
https://noctua.at/en/nf-a12x25-interview-lars-stromback

If you do ever read that, I look forward to hearing what you misinterpret from it...!

No fan has completely flat noise spectrum.
(that Noctua's graph is really too flat overal)
But to minimize creation of unnecessary noise spikes in the first place they usually make truts intersect blade trailing edge in big angle to disrupt airflow only on small part at time.
While NF-F12's "stator guide vanes" disrupt airflow from big part of trailing edge at once.
That disruption not happening at even intervals during rotation doesn't change the fact.


I first read that some year's ago when it came out...
And first you bash MartinLiquid Lab's P-Q explanation (which I haven't even brought up in this thread) of fans operating in mid area between static pressure and free airflow in real world as false.
But now when Noctua's marketing has it, it's suddenly some word of god?

 

[ttaskmaster]

No fan has completely flat noise spectrum.

Hence my statement above...

(that Noctua's graph is really too flat overal)

Oh?
What should it look like, then? Are you going to supply some graphs of your own?

But to minimize creation of unnecessary noise spikes in the first place they usually make truts intersect blade trailing edge in big angle to disrupt airflow only on small part at time.

Usually, yes... but not always.
In this instance, Noctua has actually done not one, not two or even three, but four different things on the NF-F12 to minimise the noises:

1. The trailing edge of the blades is curved enough that they do not intersect the vanes parallel along the entire length.
2. The leading edge of the vanes is also curved, with the complete opposite arc to the fan blades, further increasing the non-parallel angle. If you look, it's like this: () Perhaps not as pronounced as some other fans, but fairly typical of high SP fans. The angle of intersection is more like a pair of scissors.
3. They use flat vanes instead of thick, squared struts, to minimise the contact surface that the airflow hits. This also has other uses, such as in controlling the flow
4. The vanes are at varying angles, to spread the existing noise across the spectrum and minimise the regular noise that results in the typical fan resonance.​

NOTE: All these things can be quite clearly seen with the naked eye, by the way. I still have a pair of NF-F12s by my desk, from an older build.
So once again, your assertions regarding the Noctuas are unfounded, and actually opposed by the very sources you have posted.

And first you bash MartinLiquid Lab's P-Q explanation (which I haven't even brought up in this thread) of fans operating in mid area between static pressure and free airflow in real world as false.

No, I bash your wide misinterpretation of Martin's article, and your assumption that it is somehow a definitive dismissal of my and others' use of Mfr's stats as points of comparison, particularly when other factors are the focus anyway.

We've already covered this in another thread - Martin and I both say that manufacturer's stats are not the complete picture, but that they offer very useful and ready starting points of comparison between fans.
Martin said that the P-Q is also not a complete picture, but gives some insight into some fans (if the Mfr has provided the data), and then points out that many fans do have relatively linear PQ curves.
My only real point of contention with Martin's work is the choice of 1500rpm as his idea of a 'real world' fan speed. One of the reasons I swapped to ML120s from a Noctua setup was that, at the same speeds, the MLs were moving more air through my particular low-FPI rads and at audibly lower noise levels than the Noctuas, while offering a far higher top end for when performance becomes my primary requirement... Even now I'm running them at 1800rpm. But that is mostly influenced by airflow spread vs rad fin spacing, and has nothing to do with any of the other factors thus far mentioned here.

But now when Noctua's marketing has it, it's suddenly some word of god?

As above, really.
Many Mfrs do not even supply PQ curves, at which point it becomes a useless measure.

 

[doyll]

Ohh yeah, I'd be going for the DRP4 fans as that's ultimately what it's had research + design based on, but I was looking more for case fan suggestions. Which I think Noctua S12A seem to be the leader for at the moment?

NF-S12A are not even a half decent case fan. They are barely able to overcome resistance of a low restriction vent grill, and cannot move needed case airflow through fancy grill with filter as used on most case intakes.

Silverstonetek did a study of how grills restricted 30-71%, and that is typical rear and back grill patterns. It's much higher for front intake vents. Adding filter easily cuts airflow by another 50%. First link below shows their results as well as how even round wire grill creates disruption to airflow and thus lowers flowrate. Second link is filter restrictions.
https://www.silverstonetek.com/techtalk_cont.php?tid=wh_chessis&area=usa
https://www.silverstonetek.com/techtalk_cont.php?tid=wh12_008&area=usa

Fan discussion your thread seems to have evolved into makes lots of good points, but is way more complicated than needed to select decent fans.

Arctic P series seem to be decent fans, especially considering their pricing is about 1/2-1/3 what most decent fans cost. Some owners had bad fans out of box, some of first buyers have fans now starting to make noises they shouldn't (1-1.5 years old) and to Arctic's credit have been sent replacements. But honestly long warranty is more advertising than anything else .. buyers seem to think long warranty means long life, and that may or may not be true. Warranty is only as good as company making it. If they choose to quite honoring it after 3, 5, 7 years there is little to no legal recourse for fan owner. Then there's possibly needing to return fan at owners expense, etc. to prove it's defective which can cost more even if in country, and if to foreign country cost more than fan did. Combine that with how many people keep purchase receipts more than a few years, would even bother trying to get replacement after a few years, etc. and hopefully you can see what I mean about warranty being at least as much an advertising gimmick as functional.

Keep in mind, some cases come with decent case fans, but most do not. For a fan to be able move needed airflow through a case it need to have a static pressure rating of 1.5mm H2O or higher at 1300-1500rpm (static pressure specs on fans are at rated full speed unless otherwise stated).

Noctua NF-A12x25 are one of the best.
Arguably the best fans are the be quiet! Silent Wings 3 line.
Arctic P are good, but quite new and with such a low price it's hard to know if they will last normal 5-7 years of use.

 

[Grandmaster Fap!]

I have a P600S with Arctic P14s and P12s, i was using an ThermalRight Macho Rev B on a 2700X and the cooling was fantastic. Felt the upgrade itch for something different an got an Liquid Freezer II 280 and the cooling is even better.

Without being an Arctic fanboi im really impressed and doubt i'll change case and fans for a very long time.