New Noctua NH-D12L Low-height 120mm dual tower cooler

[Dogzilla07]

Summary

 

Perfect combination of Scythe Fuma 2 + Thermalright Silver Soul 135, but more importantly the new fan in it changes everything!

 

Quotes

Quote

 Noctua today presented the all-new NH-D12L low-height dual tower CPU cooler and the corresponding NF-A12x25r round-frame 120 mm fan. At a height of only 145 mm (13 mm lower than Noctua's regular 120 mm models), it fits many 4U enclosures as well as narrower tower cases that have been previously limited to solutions with 92 mm fans. At the same time, its five heatpipe dual-tower design and state-of-the-art NF-A12x25r 120 mm fan allow it to achieve a level of efficiency that surpasses many full-height 120 mm models.

 

My thoughts

 

Here we go, best thing to happen to cooling 25 years. My god I was not expecting LCP fan in this size, this is magnificent! Noctua prefectly copied the round frame 120mm (from Thermalright Silver Soul 135) + the 2 different girth styled towers (from Scythe Fuma 2). This fan will destroy the Thermalright's frame-less/round 120 (being LCP, and benefiting from increased motor quality). Awesome that the fan is sold seperataly as well, from the get go.

The only flaw* of the heatsink itself is it's only 145mm, but the beauty of the new awesome fan will be a tremendous addition to Thermalright Silver Soul 135 so nicely for a 135-140mm height. This brings dozens up dozens of cases to an unprecedented level of of CPU cooling performance (both as a base Noctua product, and for those few extra thin cases where combing the new NF-A12x25r will be glorious.

 

Sources

https://www.techpowerup.com/293614/noctua-presents-nh-d12l-low-height-120mm-dual-tower-cpu-cooler

[Hairless Monkey Boy]

 

"For Broader Case Compatibility"

 

Don't they mean for Narrower case compatibility?

[TrigrH]

Because including a picture was too hard:

 

[darknessblade]

13 minutes ago, Hairless Monkey Boy said:

 

"For Broader Case Compatibility"

 

Don't they mean for Narrower case compatibility?

That is soo cursed, but me likey

[Heliian]

50 minutes ago, darknessblade said:

That is soo cursed, but me likey

There is no psu and the rest of it is photoshopped. 

[Dogzilla07]

Here's Thermalright original frame-less(round) 120mm fans and the new Noctua side by side (pic below). It's very had to say without detailed measurements (if the mounting holes are the same). Regardless for case mounting the new Noctua nh-a12x25r can be mounted with screws on a case in 92mm hole positions but the flaps on the left and right would be up and below and block other fan mounting.

 

As for heatsink mounting for Thermalright Silver Soul 135/Silver Arrow 130 that should be fine possibly with even those cooler's stock mounting metal pieces (but possibly in the wrong orientation in the case of Gintama 135, detailed measurements of the fans would be really useful).

 

[ToboRobot]

Excellent, just in time for new Sliger cases.

[Sarra]

So, it's basically a NH-D15S that they photocopied at 80% resolution, but it glitched slightly?

 

Sounds good to me.

[GhostRoadieBL]

1 hour ago, Heliian said:

There is no psu and the rest of it is photoshopped. 

Psu for that case is behind the Motherboard tray (thus the need for low profile coolers)

 

The photo would have some retouching like ALL product photos but that look is just good lighting and a polarized filter for the lens to cut through glare on the glass. The rest looks like 3-4 lights with a great photographer.

Looks like (guessing) front right key light, left front softbox, front fill and a background light to separate the grey from the case.

[Deadpool2onBlu-Ray]

3 hours ago, Dogzilla07 said:

Summary

 

Perfect combination of Scythe Fuma 2 + Thermalright Silver Soul 135, but more importantly the new fan in it changes everything!

 

Quotes

 

My thoughts

 

Here we go, best thing to happen to cooling 25 years. My god I was not expecting LCP fan in this size, this is magnificent! Noctua prefectly copied the round frame 120mm (from Thermalright Silver Soul 135) + the 2 different girth styled towers (from Scythe Fuma 2). This fan will destroy the Thermalright's frame-less/round 120 (being LCP, and benefiting from increased motor quality). Awesome that the fan is sold seperataly as well, from the get go.

The only flaw* of the heatsink itself is it's only 145mm, but the beauty of the new awesome fan will be a tremendous addition to Thermalright Silver Soul 135 so nicely for a 135-140mm height. This brings dozens up dozens of cases to an unprecedented level of of CPU cooling performance (both as a base Noctua product, and for those few extra thin cases where combing the new NF-A12x25r will be glorious.

 

Sources

https://www.techpowerup.com/293614/noctua-presents-nh-d12l-low-height-120mm-dual-tower-cpu-cooler

.

 

Literally makes me want to throw up. I was expecting a dual tower 120mm Chromax all black version with dual a12x25s to replace my u12a. Not this fugly monstrocity. I guess if you don't have a tempered glass window it would be okay. EPIC disappointment. 

[dizmo]

2 hours ago, ToboRobot said:

Excellent, just in time for new Sliger cases.

Oh? What kind of cases? Fingers crossed for another small ATX. 

57 minutes ago, Ryan829 said:

.

 

Literally makes me want to throw up. I was expecting a dual tower 120mm Chromax all black version with dual a12x25s to replace my u12a. Not this fugly monstrocity. I guess if you don't have a tempered glass window it would be okay. EPIC disappointment. 

I find black to be painfully boring. 

[Deadpool2onBlu-Ray]

1 minute ago, dizmo said:

Oh? What kind of cases? Fingers crossed for another small ATX. 

I find black to be painfully boring. 

Boring sure, but it matches with other colors. If you are going puke tan/brown noctua, you gotta full send it and do the whole build that color

[ToboRobot]

24 minutes ago, dizmo said:

Oh? What kind of cases? Fingers crossed for another small ATX. 

 

 

[Dogzilla07]

So the ITX compatibility for LGA 1700 is not as godly as I anticipated (though there might be mistakes in the list for compatiblity):

 

https://ncc.noctua.at/coolers/NH-D12L-69/motherboards/all/LGA 1700

 

And together with that it's not offset away from the GPU like the NH-D15S, and is not 120mm wide (it's 125mm)  There also won't be a Chromax version right away, until they see how this one sells (from their instagram).

 

Either way I can't wait to see how that fan performs in action, aesthetics aside the design is magnificent + the implication of LCP (Liquid Crystal Polymer). I'm drooling at the prospect of going through all the small form factor cases again with the NH-D12L or NF-A12x25r+Thermalright Silver Soul 135 in mind (it's like a xmas gift come early xD, double so as I wasn't expecting to get hyped before the 140mm Phanteks/Noctua LCP versions arrive later this year).

[Deadpool2onBlu-Ray]

30 minutes ago, Dogzilla07 said:

So the ITX compatibility for LGA 1700 is not as godly as I anticipated (though there might be mistakes in the list for compatiblity):

 

https://ncc.noctua.at/coolers/NH-D12L-69/motherboards/all/LGA 1700

 

And together with that it's not offset away from the GPU like the NH-D15S, and is not 120mm wide (it's 125mm)  There also won't be a Chromax version right away, until they see how this one sells (from their instagram).

 

Either way I can't wait to see how that fan performs in action, aesthetics aside the design is magnificent + the implication of LCP (Liquid Crystal Polymer). I'm drooling at the prospect of going through all the small form factor cases again with the NH-D12L or NF-A12x25r+Thermalright Silver Soul 135 in mind (it's like a xmas gift come early xD, double so as I wasn't expecting to get hyped before the 140mm Phanteks/Noctua LCP versions arrive later this year).

It just makes no sense. Why not do the chromax version first and see how that does before coming out with the puke color one 

[cmndr]

19 hours ago, Dogzilla07 said:

Here we go, best thing to happen to cooling 25 years.

That's likely to be the advent of tower coolers. Think Tuniq Tower when it first came out.
A runner up is likely closed loop AIOs since that mostly mitigated the benefits to custom loops.

 

There are benefits to a slightly better shaped and lower profile tower cooler and a smarter take on fans but those didn't really change the underlying dynamics of the system. Heck if you don't need THE BEST you could probably get by with a ThermalRight Ultra 90 and some zip ties. It'll fit in small cases.

[TopWargamer]

Planning a small home server build at the moment and this cooler is looking very hot.

[Dogzilla07]

13 hours ago, Ryan829 said:

It just makes no sense. Why not do the chromax version first and see how that does before coming out with the puke color one 

Money(specifically fiscal responsibility) + Company stability. it's much more expensive (and challenging for the fan color) to do the chromax, basic color costs them peanuts compared to chromax and if they recoup the initial upfront R&D + tooling costs of both these guys (heatsink+fan) and the cooler turns a profit, we'll see chromax. If it doesn't, well we still might see it later on, but it's gonna be dependent on what else they use this fan on, and how good that sells, and how fast the new fan makes financial sense to turn into chromax version overall.

 

Coloring the heatsink itself is a minor obstacle (I doubt the electroplating facility needs to change much to deposit the black color), It's the initial R&D costs for both, and the monstrous task of fan coloring.

 

38 minutes ago, cmndr said:

That's likely to be the advent of tower coolers. Think Tuniq Tower when it first came out.
A runner up is likely closed loop AIOs since that mostly mitigated the benefits to custom loops.

 

There are benefits to a slightly better shaped and lower profile tower cooler and a smarter take on fans but those didn't really change the underlying dynamics of the system. Heck if you don't need THE BEST you could probably get by with a ThermalRight Ultra 90 and some zip ties. It'll fit in small cases.

My context for my "25" comment is completely unrelated to the heatsink itself, or any of the points you mentioned above. I meant LCP (Liquid Crystal Polymers) being the most important thing in cooling in the past 25 years. And the proliferation of the LCP technology to this smaller fan, and hopefully to bigger 140mm fans at the end of the year.

[cmndr]

8 minutes ago, Dogzilla07 said:

My context for my "25" comment is completely unrelated to the heatsink itself, or any of the points you mentioned above. I meant LCP (Liquid Crystal Polymers) being the most important thing in cooling in the past 25 years. And the proliferation of the LCP technology to this smaller fan, and hopefully to bigger 140mm fans at the end of the year.

It's very possible it's big in the world of plastics but...

 

fan performance doesn't matter THAT much. Assuming you're not pushing the limits (i.e. no crazy overclocks for questionable gains), the difference between ~900RPM and 3000 RPM on a fan isn't meaningful in terms of temperature improvements. If you hold noise constant... like a lighter set of blades will help but it only goes so far when much of the noise is from the air.

It's possible I'm missing something. I'm not a materials engineer.

[Dogzilla07]

7 minutes ago, cmndr said:

It's very possible it's big in the world of plastics but...

Not the context I had in mind, my context is noise-normalized performance.

 

7 minutes ago, cmndr said:

fan performance doesn't matter THAT much.

Agree to disagree, the difference is incredible, not from a pure performance standpoint ignoring noise, but from a noise-normalized stand-point, it's a tectonic shift. 200-400 RPM higher operating values for the same relative noise. And just the overall performance boost is astounding (and previously unthinkable, that there could be any changes of this amount in the fan space).

 

It's such a huge step-up, raising the base floor, that T30-120 from phanteks is having trouble dominating the A12x25 at low speeds, and 140mm fans are having trouble competing with the a12x25. And if you factor small-form-factor and new thermally dense CPUs, the difference becomes even more important.

[cmndr]

4 minutes ago, Dogzilla07 said:

Not the context I had in mind, my context is noise-normalized performance.

 

Agree to disagree, the difference is incredible, not from a pure performance standpoint ignoring noise, but from a noise-normalized stand-point, it's a tectonic shift. 200-300 RPM higher operating values for the same relative noise. And just the overall performance boost is astounding (and previously unthinkable, that there could be any changes of this amount in the fan space).

I did edit a bit so forgive that.

I still don't see how 200RPM matters more in terms of performance or if that's even achievable when most of the noise comes from the movement of air. Do you have a source for 200RPM? I didn't see that mentioned in the article you linked to. I also have a hard time imagining how less weight would lead to materially higher RPM (though it could reduce motor noise... which again isn't as big as the noise of the air)

Wouldn't better fluid simulations and blade shape be the bigger factor (which HAS improved a fair bit in the last 25 years, much of it in the last 15)?

[Dogzilla07]

14 minutes ago, cmndr said:

I still don't see how 200RPM matters more in terms of performance or if that's even achievable when most of the noise comes from the movement of air.

it's not 200RPM, it's 350-400 RPM (for Noctua, 200 is for some other LCP fans)

 

14 minutes ago, cmndr said:

I still don't see how 200RPM matters more in terms of performance or if that's even achievable when most of the noise comes from the movement of air.

It is, it's been shown beyond a shadow of a doubt, it solved the biggest bottle-neck (harmonic resonance due elasticity of the impeller tips), and having only 5mm room between the tips of the blade and the frame, has indeed achieved not only a reduction in high and low frequency noise, but a reduction in noise coming from movements of air, that's why it's such a tectonic shift. The new material has achieved things that were considered impossible previously, by physics standards (reducing overall movement of air noise).

 

14 minutes ago, cmndr said:

Wouldn't better fluid simulations and blade shape be the bigger factor (which HAS improved a fair bit in the last 25 years, much of it in the last 15)?

In Theory yes, but in practice no, the supercomputer, machine-learning, big data AI are still not here yet, to advance those things enough. While LCP has a usable, measurable, tangible advantage. This is an exotic meta-material, the kind scientists are always talking about we need to overcome our current law of physics limitation (like room-temperature superconductors, and exotic particles for warp-drive, etc, ...) The successful mass manufacturing is that is that kind of breakthrough, just not in those grander, more important spaces, and well for relative small gains when looking at the big picture xD (But still glorious for small-form factor builds).

 

Look at Phanteks T30-120, new innovative blade shape design, better fluid simulations that the competition (experience from their MP series), 5mm fatter, and still cannot beat Noctua A12x25 at the lowest speeds (arguable not important, but in context of my comparison point). This proves that LCP just lifts the whole floor up, and is the biggest impactor out of all the other factors.

 

[cmndr]

It's possible you meant to link to THIS - https://www.techpowerup.com/review/noctua-nf-a12x25-pwm/5.html

Which is a review of that fan, not a heatsink that comes with it. It looks like arguably the best fan in the roster, though I'm still in "so what?" territory.

It's about .5dB louder than the Phantecs equivalent but also moves a hair more air.
I don't see any "it cuts temperatures by 1C" figures though. My expectation is that

Now it's very possible that a resonance would be annoying above and beyond what is measured in dB. I won't rule that out.


I'll still ask - show that this can hit a low target - think 1C better on a semi-reasonable platform (stock 12900k? stock 5950x?) at similar noise to something else. I see platitudes but no data.

[Dogzilla07]

13 minutes ago, cmndr said:

Which is a review of that fan, not a heatsink that comes with it

Exactly, that's my point, and yes that's the big brother of this fan, and here's the graph with example I meant, that fluid simulations and blade shape are inferior in effect to LCP (as all 3 fans are LCP here):

13 minutes ago, cmndr said:

It's about .5dB louder than the Phantecs equivalent but also moves a hair more air.
I don't see any "it cuts temperatures by 1C" figures though. My expectation is that

Exactly, glad you're finally understanding my point, that there is proof, (quality methodology, good testing), proof of what I'm saying 

 

13 minutes ago, cmndr said:

Now it's very possible that a resonance would be annoying above and beyond what is measured in dB. I won't rule that out.

Exactly, the biggest problem wasn't the movement of air, it was the resonance, before the movement of air noise occludes it, and well the inability to have 0.5mm pitch difference with other materials.

13 minutes ago, cmndr said:

I'll still ask - show that this can hit a low target - think 1C better on a semi-reasonable platform (stock 12900k? stock 5950x?) at similar noise to something else. I see platitudes but no data.

The data you showed there proves most my points. But specifically it cannot 100% prove the point you're making there, it's apples to oranges comparison. Partial proof is in the link you provided, as you can see in the graphs there how extraordinary the LCP Noctua fan is compared to the other fans (that VSG test on techpowerup clearly shows the higher speeds for lower noise (not even counting harmonics), so just dB measured, that the fan does compared to other competition. The Phanteks MP, which you probably are referencing was one of the 2 kings of noise-normalized previously together with Silent Wings 3 from Be Quiet!.

 

As long as you're talking absolute, not noise-normalized performance, and not taking into account high and low noise frequencies which don't show up in regular dB/dBA testing, but will show up in the spectrum analysis GamersNexus will do, you've made room for both of our points to be both right and wrong at the same time, due to the discrepancy in the context.

 

Other testing which shows how good noise profiles of the LCP fans is, Machines&More testing, OptimumTech, TechBuyer's Guru.

Other testing that shows how good the fan overall is is Der8auer's roundup.

And lastly but not least, GamersNexus testing is coming soon!

[cmndr]

I still don't see anything like noise normed performance deltas (measured in temps).

 

Saying "best thing to happen to cooling 25 years" is a pretty big claim.

If you said, best thing to happen to fans in 25 years I'd be less critical.

15 years ago we were at a point where we could slap a big HSF in a case and run one fan at 700RPM or so and it was inaudible below the noise of my own breathing. I can't hear my computer right now but I could back in 2004 so...

I don't see how this enables new use cases. I don't see how it makes a material difference to loudness. At least among the crowd willing to sacrifice 1-2C.