Tuesday, November 9th 2021

Noctua Introduces NH-L9i Low-profile CPU Coolers for LGA1700 and NA-FD1 Fan Duct

Noctua today introduced new, LGA1700-specific revisions of its award-winning NH-L9i and NH-L9i chromax.black low-profile CPU coolers. The new NH-L9i-17xx and NH-L9i-17xx chromax.black are ideal for building ultra-compact HTPCs and Small Form Factor (SFF) systems using Intel's brand new 12th generation Core processors such as the Core i9-12900K, Core i7-12700K or Core i5-12600K. The new, optional NA-FD1 fan duct makes it possible to further improve the performance of the coolers by bridging the gap between the fan and perforated case panels in order to enable the coolers to draw in fresh air from the outside.

"We're very happy with the performance of the NH-L9i-17xx coolers on Intel's new LGA1700 CPUs", says Roland Mossig (Noctua CEO). "We have managed to dissipate up to around 160 W on the Core i9-12900K, pushing it to over 4.2 GHz, and up to 125 W on the Core i5-12600K running at 4.3 GHz. These are excellent results for such small coolers, making them fantastic options for highly compact Intel Z690 builds that pack a lot of processing power!"
The NH-L9i-17xx and the all-black NH-L9i-17xx chromax.black are the latest, LGA1700-specific revisions of Noctua's award-winning NH-L9i low-profile CPU cooler. At a height of only 37 mm, the NH-L9i is ideal for extremely slim cases and, due to its small footprint, it provides 100% RAM and PCIe compatibility as well as easy access to near-socket connectors even on tightly packed Mini-ITX motherboards. The custom-designed SecuFirm 2 mounting system for the new LGA1700 socket makes installation a breeze and, thanks to the highly optimised NF-A9x14 92 mm fan that supports fully automatic speed control via PWM, the NH-L9i-17xx runs remarkably quietly. Topped off with Noctua's renowned NT-H1 thermal compound, the NH-L9i-17xx combines everything users have come to expect from Noctua's larger coolers into a super-compact, premium-quality package for ITX and HTPC builds with Intel's new 12th generation Core CPUs.

"For Small Form Factor builds, we now highly recommend the optional NA-FD1 fan duct kit, both to customers choosing the new NH-L9i-17xx models and to users of the existing NH-L9i and NH-L9a heatsinks", explains Roland Mossig (Noctua CEO). "It's a simple yet highly effective way of boosting the performance of these coolers in compact cases. Tech-savvy enthusiasts have been creating similar ducts with their 3D printers for quite some time, but not everyone has the capabilities to do this, so we thought it would be great to offer an affordable, flexible duct kit that doesn't require any special equipment or skills to set up."

The new NA-FD1 fan duct kit makes it possible to significantly improve the performance of Noctua NH-L9i and NH-L9a series CPU coolers in Small Form Factor (SFF) environments where there is a gap of 5 mm or more (up to 45 mm) between the cooler and a perforated top or side panel. Bridging this gap, the duct prevents the cooler from taking in hot air from the interior and enables it to draw in fresh air from the outside through the perforated panel, which can, depending on the setup, improve CPU temperatures by as much as 5°C or more. As the duct is assembled from EVA foam spacers of different thickness, it is fully modular and can be adjusted in 1 mm increments from 5 mm to 45 mm height in order to fit various different cases and configurations. In short, the NA-FD1 is a simple yet smart, flexible, and cost-effective way of boosting the efficiency of NH-L9a and NH-L9i series coolers

The manufacturer's suggested retail prices are as follows:
  • NH-L9i-17xx: EUR/USD 44.90
  • NH-L9i-17xx chromax.black: EUR/USD 54.90
  • NA-FD1: EUR/USD 12.90
For more information, visit the product pages of the NH-L9i, NH-L9i chromax.black, and NA-FD1.
Add your own comment

131 Comments on Noctua Introduces NH-L9i Low-profile CPU Coolers for LGA1700 and NA-FD1 Fan Duct

#51
looniam
MetroidYou are losing focus here, please be focused. Anyway, like i said previously that is clear 319 watts, so i will say again, it used more than 300 watts. So I have not lied or made it up, why can't you accept that? It's so hard to see the image and not accept the truth?
speaking of focus
We have managed to dissipate up to around 160 W on the Core i9-12900K, pushing it to over 4.2 GHz, and up to 125 W on the Core i5-12600K running at 4.3 GHz
all those power consumption graphs you're showing show speeds of 5.1Ghz plus for 300+ watts.

now go back to the graph dedi posted and see an i9 @4.2Ghz ~160 watts.

Posted on Reply
#52
Dredi
MetroidWith delta fans they can even do 500 watts ehhe, jokes aside, very high static pressure is king, so yes it depends on things but radiator wise a 240mm and normal fan is up to 150w. Anyway this is off topic.
You were never on topic, tbf.

In the strix lc 3080 ti review, a single 240mm rad with typical quiet-ish asus fans cools down even 400 Watts quite easily.
Posted on Reply
#53
Metroid
looniamspeaking of focus

all those power consumption graphs you're showing show speeds of 5.1Ghz plus for 300+ watts.

now go back to the graph dedi posted and see an i9 @4.2Ghz ~160 watts.

Is not full package, so yes using few cores that is how it is, 12900k single thread power consumption is way better than even my ryzen 5900x. The question is are you buying a 12900k and lock on single thread? I mean, I was thinking few days ago why intel does not release a 12900k single thread version only hehe, i mean 1 performance core only. That will be amazing for power saving ehhe, this low profile cooler will do wonders on that 12900k single thread hehe

DrediYou were never on topic, tbf.

In the strix lc 3080 ti review, a single 240mm rad with typical quiet-ish asus fans cools down even 400 Watts quite easily.
Like i said before, you can't change the law of physics. You put a better fan and yes it does a better job, given normal common fans then 99% will be the same.
Posted on Reply
#54
Dredi
MetroidIs not full package
It’s not? Link pls. Hehe
Posted on Reply
#55
Metroid
DrediIt’s not? Link pls. Hehe
Full package is showing there as per anandtech review, 272w, default clocks, not overlocked.

Posted on Reply
#56
Dredi
MetroidFull package is showing there as per anandtech review, 272w, default clocks, not overlocked.

still waiting for the 12600k > 300W Full package results here.

Hehe
Posted on Reply
#57
looniam
MetroidIs not full package, so yes using few cores that is how it is, 12900k single thread is way better than even my ryzen 5900x. The question is are you buying a 12900k and lock on single thread? I mean, I was thinking few days ago why intel does not release a 12900k single thread version only hehe, i mean 1 performance core only. That will be amazing for power saving ehhe, this low profile cooler will do wonder son that 12900k single thread hehe
just stop with the deflections the stated marketing is perfectly feasible. who in blue blazes is going to throw a 300 watt chip in a SFF build [as this cooler is marketed for!] and air cool it?

you just don't get it - now your using my edited graph w/o realizing where you're wrong. again the clock speeds are limited regardless of the load . . :shadedshu:
Posted on Reply
#58
Metroid
Dredistill waiting for the 12600k > 300W Full package results here.
I already showed the image as per the techpower review, now if you believe it or not that is for you to decide ehhe, not going to bother again, if we keep this up, this will be loop of yes and no, better we stop here, well I will.
Posted on Reply
#59
Dredi
MetroidI already showed the image as per the techpower review, now if you believe it or not that is for you to decide ehhe, not going to bother again, if we keep this up, this will be loop of yes and no, better we stop here, well I will.
PACKAGE POWER. Hehe. I guess you can’t. Hehe.

In the TPU review the gaming load is even higher, HEHE.
Posted on Reply
#60
Valantar
Metroidb: I never said they claimed, I said this is not enough for that, yes they cited the 12900k in their tests which in my opinion was a mistake. This low profile cooler is not for the 12900k and yet they included in the tests to brag.
If you didn't say they claimed that, what or who are you arguing against here? I don't get it. "Nobody said this would be possible, but there's no way this is possible" sounds like you're getting confused to me.
MetroidMy ryzen 5900x uses a 150w aio 240mm cooler and yet it cant cool down a 140 watts cpu. So reason I laughed noctua came up with a low profile cooler using a 12900k hehe, not even their top air cooler can do the job well on that 12900k.
What does "can't cool down" mean? What are your acceptable temperatures/clocks for that? Remember, this is an SFF cooler. It might be able to keep a 12900K running at 160W power draw, but it'll likely be running up against tJmax while doing so. Which is fine - the CPU can take it. If that's not acceptable to you, but you for example insist on thermals in the 60s or 70s, then you obviously have to lower your expectations for clocks and power draws.
napataIt's running unlocked here, which isn't stock.
Nope. All stock settings. The graph is purposely labeled peak power though - they make no claims to this being sustained over any period of time.
rares495I've used the NH-L9a in a SFF Ryzen 3400G build and it barely kept up there. I think this will work well with the LGA1700 celerons, pentiums and maybe up to the cheapest i3 but anything more than that is a no.

Their claim of 160W is ridiculous at best.
What does "barely kept up" mean? What thermals? In what case? With what airflow? Context is crucial for SFF cooling (and all cooling, really). I've seen plenty of builds cooled using this cooler with much, much more powerful CPUs than a 3400G.
MetroidIs not full package, so yes using few cores that is how it is, 12900k single thread power consumption is way better than even my ryzen 5900x. The question is are you buying a 12900k and lock on single thread? I mean, I was thinking few days ago why intel does not release a 12900k single thread version only hehe, i mean 1 performance core only. That will be amazing for power saving ehhe, this low profile cooler will do wonders on that 12900k single thread hehe
Or they could just buy a 12900K (or 5950X for something similarly overkill) with the purpose of maximizing performance in a given case+thermal envelope through selective tuning, power limiting and airflow optimizations. You'd still keep your ST performance through all of that unless you start going all-core underclocking with turbo disabled.
Posted on Reply
#61
Metroid
looniamjust stop with the deflections the stated marketing is perfectly feasible. who in blue blazes is going to throw a 300 watt chip in a SFF build [as this cooler is marketed for!] and air cool it?

you just don't get it - now your using my edited graph w/o realizing where you're wrong. again the clock speeds are limited regardless of the load . . :shadedshu:
Wrong in? sorry i did not understand what you meant. Your image is spot on, not saying is false, you need to understand that is using pov ray, 160w is perfectly normal for that application, is not full package, dont know why you tried to say otherwise and attacking what i have not said at all. I agreed with your image but pointed to facts shown there.
ValantarIf you didn't say they claimed that, what or who are you arguing against here? I don't get it. "Nobody said this would be possible, but there's no way this is possible" sounds like you're getting confused to me.

What does "can't cool down" mean? What are your acceptable temperatures/clocks for that? Remember, this is an SFF cooler. It might be able to keep a 12900K running at 160W power draw, but it'll likely be running up against tJmax while doing so. Which is fine - the CPU can take it. If that's not acceptable to you, but you for example insist on thermals in the 60s or 70s, then you obviously have to lower your expectations for clocks and power draws.

Nope. All stock settings. The graph is purposely labeled peak power though - they make no claims to this being sustained over any period of time.

What does "barely kept up" mean? What thermals? In what case? With what airflow? Context is crucial for SFF cooling (and all cooling, really). I've seen plenty of builds cooled using this cooler with much, much more powerful CPUs than a 3400G.

Or they could just buy a 12900K (or 5950X for something similarly overkill) with the purpose of maximizing performance in a given case+thermal envelope through selective tuning, power limiting and airflow optimizations. You'd still keep your ST performance through all of that unless you start going all-core underclocking with turbo disabled.
The heat builds up more than the cooler itself cools it down, that is when you know that cooler is not up for set clocks or said tdp if default clocks.
Posted on Reply
#62
Dredi
MetroidThe heat builds up more than the cooler itself cools it down, that is when you know that cooler is not up for set clocks.
But the cooler performance increases as it heats up.

In water cooling, going from 40C to 60C liquid temp in a 20C room doubles your coolers performance in Watts at the same fan speeds.
Posted on Reply
#63
Valantar
MetroidThe heat builds up more than the cooler itself cools it down, that is when you know that cooler is not up for set clocks.
Uh ... what? You reach steady-state at some point. If it gets too hot, the CPU will throttle. What is your steady-state level? At what combination of fan speeds, pump speeds, CPU power draw, CPU temperature? That's what matters, not that ... things get hotter (until they don't).
Posted on Reply
#64
Metroid
DrediBut the cooler performance increases as it heats up.
Only if fans still have more room, if fans are at 100% then gameover.
Posted on Reply
#65
Dredi
MetroidOnly if fans still have more room, if fans are at 100% then gameover.
You are 100% wrong here. Thermal transfer efficiency scales linearly with delta t.
Posted on Reply
#66
Caring1
If you want a performance system, give this cooler a miss.
Posted on Reply
#67
Valantar
MetroidOnly if fans still have more room, if fans are at 100% then gameover.
Actually not - as the temperature delta between your radiator and the ambient air increases (i.e. the radiator heats up) heat transfer efficiency will increase, leading to a higher-than-linear increase in thermal transfer as the radiator gets hotter (assuming steady airflow).
Posted on Reply
#68
Metroid
ValantarUh ... what? You reach steady-state at some point. If it gets too hot, the CPU will throttle. What is your steady-state level? At what combination of fan speeds, pump speeds, CPU power draw, CPU temperature? That's what matters, not that ... things get hotter (until they don't).
Yes, that is dependent to manufactures throttle temperature value, not sure when it will hit on alder lake, probably 105c or so, usually is around that.
Posted on Reply
#69
Valantar
Caring1If you want a performance system, give this cooler a miss.
Only if your definition of "performance system" necessarily means "performance with no regard to size", as this cooler will let you build one hell of a portable workstation. (Of course there are also lots of cases supporting slightly larger coolers like the NH-L12s wich will obviously perform even better, but the case will be less portable.)
MetroidYes, that is dependent to manufactures throttle temperature value, not sure when it will hit on alder lake, probably 105c or so, usually is around that.
I was asking about your CPU and your cooler. You said your cooler was insufficient because
MetroidThe heat builds up more than the cooler itself cools it down
which doesn't address the core question of what makes the cooler insufficient - for it to be so it must fail in some tangible way, i.e. too high steady-state temperatures, thermal throttling, excessive noise, etc. I was asking about your specific tolerances and limits for these questions, not manufacturer-specified limits.
Posted on Reply
#70
rares495
ValantarWhat does "barely kept up" mean? What thermals? In what case? With what airflow? Context is crucial for SFF cooling (and all cooling, really). I've seen plenty of builds cooled using this cooler with much, much more powerful CPUs than a 3400G.
It was a Fractal Design Node 202 case so airflow was good since the cooler was right underneath the perforated part of the side panel.

The CPU reached 80 degrees or something like that. I don't remember the exact tests I performed but I do remember thinking to myself "Wow, so this is the limit of such a low profile cooler. Good thing I didn't go for a higher TDP CPU".
Posted on Reply
#71
Metroid
ValantarActually not - as the temperature delta between your radiator and the ambient air increases (i.e. the radiator heats up) heat transfer efficiency will increase, leading to a higher-than-linear increase in thermal transfer as the radiator gets hotter (assuming steady airflow).
So your theory is if fans are 100% always and the radiator keeps heating up then it will increase performance in heat transfer but what about temperature? I have done a similar test here few days ago, steady 100% fans and then temperature going up and up and then it hit the thermal limit, yes probably you right on that but the temperature kept increasing and boom, system turned off.

So my statement holds true, if fans are 100% and temperature keeps rising them gameover, maybe some people think a temperature will be hit and will not rise anymore than that, I thought that too and I was wrong. It kept rising and rising, steady, clearly it needed better fans or better heat transfer to keep from rising.
Posted on Reply
#72
Dredi
MetroidSo your theory is if fans are 100% always and the radiator keeps heating up then it will increase performance in heat transfer but what about temperature? I have done a similar test here few days ago, steady 100% fans and then temperature going up and up and then it hit the thermal limit, yes probably you right on that but the temperature kept increasing and boom, system turned off.

So my statement holds true, if fans are 100% and temperature keeps rising them gameover.
It’s not a theory, but a fact. The temperature will eventually stop climbing as the coolers performance increases with the higher delta t between water and air.

In your case the cpu just got too warm to function properly.

Also, how the hell did you manage to shut your system down with a 240mm rad and fans going full blast? Poor contact with the processor? Silly overclock? Modern processors simply throttle down, and don’t shut down.
Posted on Reply
#73
looniam
MetroidWrong in? sorry i did not understand what you meant. Your image is spot on, not saying is false, you need to understand that is using pov ray, 160w is perfectly normal for that application, is not full package, dont know why you tried to say otherwise and attacking what i have not said at all. I agreed with your image but pointed to facts shown there.
that POV is better than anything you provided. again this is for a SFF build for a HTPC - what load(s) and cherry picked power consumption graphs you are trying to use are not relevant.

focus; anything above that red line i edited is irrelevant in the marketing.
Posted on Reply
#74
Metroid
looniamthat POV is better than anything you provided. again this is for a SFF build for a HTPC - what load(s) and cherry picked power consumption graphs you are trying to use are not relevant.
How is cherry picked? is from techpower review and I also showed the anandtech, why cant you accept the fact what showed on the image you provided? 272w full package.
Posted on Reply
#75
Dredi
MetroidHow is cherry picked? is from techpower review and I also showed the anandtech, why cant you accept the fact what showed on the image you provided? 272w full package.
TPU was not measuring the processor power.
Posted on Reply
Add your own comment