Question AIO front mounting weakens other components

Jun 26, 2022
18
1
10
Hi all, I know I read somewhere that front mounting an AIO shortens the life span of other components because it causes the intake to have warmer air coming into the PC and thus warms the GPU and MoBo more. I can't find any information on it, so I am wondering if I just read someone's bad take and don't remember or if I was making it up in my tiny mind.

Has anyone ever heard this about front mounting AIOs, particularly if you front mount the larger radiators? Trying to ask the experts to see if my memory is failing.

Thanks!
 
  • Like
Reactions: Wolfxmw
Do NOT worry about this. The warming of incoming air by the rad is at worst a very few degrees, and makes almost NO difference in the ability of the case ventilation fans to remove heat from the mobo and other components. That includes the graphics card. About the only real effect IF you try hard to measure it is that the case vent fans may tend to run a tiny bit faster - like, 1050 or 1100 RPM instead of 1000. That is because those fans are controlled according to the actual TEMPERATURE measured by a sensor on the mobo. Likewise, the graphics card takes care of its own cooling by controlling its fan's speed to keep the GPU chip cooled to its proper temperature all the time. The impact of the heating of incoming air by the AIO rad is like (but smaller than) the difference in your room temperature between a warm day and a cool one .
 
  • Like
Reactions: future_wave
I didn't mention a memory failure specifically. Not sure what you mean there.
Trying to ask the experts to see if my memory is failing.
^That, in your first post, is either referring to ram, the memory in a M.2 drive, or the gpu's memory.

~BUUUUT, I believe I've misread this part. Memory was referring to what you remember, and not the memory devices in the PC.
 
  • Like
Reactions: rickypicky5
Do NOT worry about this. The warming of incoming air by the rad is at worst a very few degrees, and makes almost NO difference in the ability of the case ventilation fans to remove heat from the mobo and other components. That includes the graphics card. About the only real effect IF you try hard to measure it is that the case vent fans may tend to run a tiny bit faster - like, 1050 or 1100 RPM instead of 1000. That is because those fans are controlled according to the actual TEMPERATURE measured by a sensor on the mobo. Likewise, the graphics card takes care of its own cooling by controlling its fan's speed to keep the GPU chip cooled to its proper temperature all the time. The impact of the heating of incoming air by the AIO rad is like (but smaller than) the difference in your room temperature between a warm day and a cool one .


Thank you. That was my assessment. I just wanted to verify to make sure since I couldn't find the original information I read.

What do you think about the situation of a 420 mm front-mounted when the case has a power supply shroud? I need to fix the wires in there so it's less of a rat's ness, but I have an HDD in there and I am not sure there is a good way to help move the air coming in to the shroud and hitting my HDD and PSU. Am I overthinking that one?
 
^That, in your first post, is either referring to ram, the memory in a M.2 drive, or the gpu's memory.

~BUUUUT, I believe I've misread this part. Memory was referring to what you remember, and not the memory devices in the PC.

oooh sorry. lol Yes, yes, I did mean my personal memory and not in the PC. Given the context, I can see the confusion. I should have been clearer. I apologize.
 
Its a none issue really, I mean I don't like it mounted in the front only because I run my GPU ram and stuff to the edge as it is, so the cooler the better even if its only 1c difference. But if you well with in spec temp wise its not going to shorten the life other components, at least not in a manner that would matter.
 
Its a none issue really, I mean I don't like it mounted in the front only because I run my GPU ram and stuff to the edge as it is, so the cooler the better even if its only 1c difference. But if you well with in spec temp wise its not going to shorten the life other components, at least not in a manner that would matter.

Yeah, I don't think I am going to try out overclocking things any time soon. And if I did it would be on my next build and I would tinker with a different PC in case I break something. lol I am well within specs. Just testing temps with a 240 AIO front mount my GPU is actually running cooler somehow. Not sure I understand the principle on that one exactly.
 
Yeah, I don't think I am going to try out overclocking things any time soon. And if I did it would be on my next build and I would tinker with a different PC in case I break something. lol I am well within specs. Just testing temps with a 240 AIO front mount my GPU is actually running cooler somehow. Not sure I understand the principle on that one exactly.
Its probably due to how the air flow is in your case, a lot of GPUs mostly them ones with a 2 or 3 fans recirculate hot air at the bottom of the case, your AIO being mounted in the front is probably creating enough air flow with fresh air even though its warm air at the bottom of the your case.
 
Its probably due to how the air flow is in your case, a lot of GPUs mostly them ones with a 2 or 3 fans recirculate hot air at the bottom of the case, your AIO being mounted in the front is probably creating enough air flow with fresh air even though its warm air at the bottom of the your case.

Maybe. With the AIO I now have the two 120 fans as intake on the radiator and one 120 below it as intake. Before I moved it from the top, I had 3 140MM fans in front as intake. In theory, they should have put in more air than the 120s do. But perhaps since then are intake and running faster because they cool the PC they are putting in more than the 140s were.
 
It's ok. The confusion was cleared up.

What do you think about the situation of a 420 mm front-mounted when the case has a power supply shroud?
Wow, 420 is big for an AIO. I think 280mm is enough if the goal is just to use 140mm fans.
Plus, there's no need to have a fan blowing underneath the psu shroud anyway, since there's the option of setting the psu as bottom intake or top exhaust*.
[Some psus have a fan off mode, or are fanless. To take advantage of this, they should be flipped the other way, but not all cases have perforations in their psu shrouds, so beware.]

Look at how you use your PC, and monitor power use in your day-to-day activities. You can use HWINFO to see how much power your cpu and gpu use for tasks.
Unless you're running Blender or similar on an i9(or heavily OCed i7) for a few or more hours straight, the gpu often is the primary heating element of the space heater that is the PC, and they do so easily.
Plus with newer gpus only looking to get more power hungry... yikes.


[rant]
I swear society started this top cpu AIO mess.
Yes, a little warm air comes out the back of a front cpu AIO - it's overblown in most user's situations - it's still supplying air to the components behind it, and that's better than no air.
What about the open air gpus that dump their waste heat inside PCs, heating up components around and above them? Still overblown for most, but they are using more total power, and thus can heat up components more.
A front cpu AIO warms up components 1-2C more. An open air gpu can warm up components 4C or more, because they're drawing more, there's more energy leakage, that energy becomes heat, and heat has to go somewhere...
You look at how some upper end RTX 30 series can do some 400w of power with ease, and cpus like the 5800X and 12700K are doing like 120w - talking in game power use here - that's a gap of more than 3x.

Tip: If you do top cpu AIO, leave the rear fan slot empty, and put a filter over it. The rear exhaust alongside a top AIO is outdated(?) method. That just takes away a 2nd cool air source that could serve the other half of the AIO.
If you believed the front intake alone fed the entire AIO, you'd be wrong...
With tower air coolers, yeah, leave the rear exhaust fan where it is.

Look at it another way - if we ignore everything else, except the cpu and gpu:
"Front cpu AIO warming up the gpu isn't cool, but the gpu warming up the top cpu AIO is"... :heink:
That's what I see on tech forums most of the time. The whole argument should be situational depending on your hardware and how you use your PC.
[/rant]


What is your case? Case design will have an effect on overall cooling.
Also, more fans or stronger fans isn't always more better:
 
Just a small comment on the issue of comparing 120 and 140mm fan cooling effect. The 140's run slower to achieve the SAME air flow, so fan SPEED does not tell you much. More importantly, you should understand HOW fan speed is determined. The two major automatic cooling control systems (one for the CPU, the other for the case ventilation) operate the same way. Each is really a TEMPERATURE control system that uses a real temperature measured by a relevant sensor. CPU cooling is guided by a sensor built into the CPU chip; case cooling is guided by a sensor on the mobo. For each system, the software control has a temperature target for that sensor (depends on hardware, pre-set in BIOS) and some control loop tuning parameters. Each manipulates the speed of its fans to whatever it takes to achieve the temp target. But speed is not the aim - it is manipulating air flow to get whatever cooling effect is required. In fact, the system does NOT look at or use the speed reading from the fan because it does not care about that. (A fan header DOES look at the fan speed signal to detect possible fan FAILURE as a secondary task, but that's not part o cooling control.)

So, when you change fans, the TEMP target does not change, and the system will alter the fan speeds to whatever it wants to achieve the SAME temp target within a few degrees.
 
It's ok. The confusion was cleared up.


Wow, 420 is big for an AIO. I think 280mm is enough if the goal is just to use 140mm fans.
Plus, there's no need to have a fan blowing underneath the psu shroud anyway, since there's the option of setting the psu as bottom intake or top exhaust*.
[Some psus have a fan off mode, or are fanless. To take advantage of this, they should be flipped the other way, but not all cases have perforations in their psu shrouds, so beware.]

Yes, 420 IS big. And I know that. I have issues lol I really do. Let me try and explain and this is the short version of the long story (Unless you want to hear what got me to this point in the first place.)

I had a Thermaltake 240 ARGB AIO that I was going to use for a different project. I bought it because there was a sale on Amazon for 70 dollars. At the time I didn't want to pass it up. I haven't gotten around to the other project and the AUO has been sitting in my closet since October 2021. I decided to put it in my PC and I was pleasantly surprised by the performance of an AIO (not just the model specifically). If I was going to get an AIO then I had several types in mind. I was shopping for the Arctic Liquid Freezer series. They don't always have the best shipping times/options on Amazon. I saw they charged for shipping for certain AIOs even for Prime members. I sent Arctic an email and asked them why. As I was shopping I saw the 420 pop up for sale and it had free shipping. In order to have it arrive by a certain date, I had to purchase it within like 45 min or something similar. I was like, hell, why not pull the trigger and get it? I wasn't paying attention to the seller or other details I guess.

If I want to return it, I have to pay for shipping and take a 10% hit on a restocking fee. I calculated the shipping. That plus the 10% is a big hit. So, I can try to sell it locally or I can just install it in the damn system. I was actually going to put it in last night, but I didn't unpack it. I had my PC apart and I was thinking, "Man, this is way more cooler than I actually need." It's a big boy. I like tinkering on my PC anyway even if it always takes me longer to do things than I anticipated. e.g. trying to have good cabling. Which I have not finished yet as I plan on eventually opening it back up to replace fans and probably a different AIO.

I get torn - the 420 is too much, but I may be stuck with it or take a big loss. In anger, I purchased a 360 AIO in anger when I received the official return policy for the 420 - but I checked to make sure returns were also free this time. I could use a 360, but from my understanding in many instances, they perform similarly to 280mm AIOs. I suppose everything is situational and it will come down to people's set ups and brands.

So, right now I am not sure what to do. I just know that every moment I don't have the PC "set up" it grates on me. That is what causes me to make rash decisions because I want to get it "done."

Look at how you use your PC, and monitor power use in your day-to-day activities. You can use HWINFO to see how much power your cpu and gpu use for tasks.
Unless you're running Blender or similar on an i9(or heavily OCed i7) for a few or more hours straight, the gpu often is the primary heating element of the space heater that is the PC, and they do so easily.
Plus with newer gpus only looking to get more power hungry... yikes.
[/rant]

I don't use a lot of power. I sometimes use 3d rendering programs, but not a lot. The most intensive things I might do are video and video FX rendering I suppose. I haven't monitored my power use. But I assume I am fine. I know I saw my total use in general, but I can't recall off hand what it was, sadly.

[rant]
I swear society started this top cpu AIO mess.
Yes, a little warm air comes out the back of a front cpu AIO - it's overblown in most user's situations - it's still supplying air to the components behind it, and that's better than no air.
What about the open air gpus that dump their waste heat inside PCs, heating up components around and above them? Still overblown for most, but they are using more total power, and thus can heat up components more.
A front cpu AIO warms up components 1-2C more. An open air gpu can warm up components 4C or more, because they're drawing more, there's more energy leakage, that energy becomes heat, and heat has to go somewhere...
You look at how some upper end RTX 30 series can do some 400w of power with ease, and cpus like the 5800X and 12700K are doing like 120w - talking in game power use here - that's a gap of more than 3x.
There is a lot of misinformation out there about many configurations. For someone like me who tries to research or learn, then it takes a while to synthesize the actual information from the multitude of BS out there. I have friends who build PCs who recently told me they prefer "top" AIO because they don't want hot air blowing on components. But I wonder if they actually monitored the temps with it front-mounted and used a thermometer to take various temps within their case. I don't think people bother to test their assumptions.

And YES, the fan-style GPUs blow far more heat and often hotter heat out into the case in comparison to the temps of the actual air over the radiator.

Again, people have poor critical thinking skills and many assumptions. I have taught rhetoric for going on 18 years now. It's hard to break through a lifetime of bad thinking habits.

Tip: If you do top cpu AIO, leave the rear fan slot empty, and put a filter over it. The rear exhaust alongside a top AIO is outdated(?) method. That just takes away a 2nd cool air source that could serve the other half of the AIO.
If you believed the front intake alone fed the entire AIO, you'd be wrong...
With tower air coolers, yeah, leave the rear exhaust fan where it is.

Not sure I follow you on this one. Which fan slot should I leave empty? Do you mean the rear exhaust?

Look at it another way - if we ignore everything else, except the cpu and gpu:
"Front cpu AIO warming up the gpu isn't cool, but the gpu warming up the top cpu AIO is"... :heink:
That's what I see on tech forums most of the time. The whole argument should be situational depending on your hardware and how you use your PC.
[/rant]


What is your case? Case design will have an effect on overall cooling.
Also, more fans or stronger fans isn't always more better:

My case is a Phanteks P600S. I have a Ryzen 3900X on an MSI X570 ACE with a 2070 Super. I plan on upgrading the GPU eventually, and maybe even the CPU before AMD switches to AM5. Depends on timing and money - the last one always gets us lol

Not entirely sure what you mean by more fans or stronger fans are not always "better." I know that you want to pay attention to the fans and the positive vs negative pressure. In that respect, no, having many fans that give you the wrong pressure will not always make things better. Not sure about the stronger part since most people set fan curves and the temperature and curve dictates the fan speed. Unless I am just being daft. I was up late working on the PC, so forgive me. lol

Thank you for your help! Much appreciated!
 
Just a small comment on the issue of comparing 120 and 140mm fan cooling effect. The 140's run slower to achieve the SAME air flow, so fan SPEED does not tell you much. More importantly, you should understand HOW fan speed is determined. The two major automatic cooling control systems (one for the CPU, the other for the case ventilation) operate the same way. Each is really a TEMPERATURE control system that uses a real temperature measured by a relevant sensor. CPU cooling is guided by a sensor built into the CPU chip; case cooling is guided by a sensor on the mobo. For each system, the software control has a temperature target for that sensor (depends on hardware, pre-set in BIOS) and some control loop tuning parameters. Each manipulates the speed of its fans to whatever it takes to achieve the temp target. But speed is not the aim - it is manipulating air flow to get whatever cooling effect is required. In fact, the system does NOT look at or use the speed reading from the fan because it does not care about that. (A fan header DOES look at the fan speed signal to detect possible fan FAILURE as a secondary task, but that's not part o cooling control.)

So, when you change fans, the TEMP target does not change, and the system will alter the fan speeds to whatever it wants to achieve the SAME temp target within a few degrees.


Yeah, I might pick a bit on the "same" airflow. The bios and fan curves do not know the CFM and static pressure of my fans. It is trying to achieve the same temperatures no matter the fan size.

Thank you for the information on temperature control. Yes, I went into my bios and set curves and chose what each fan header was going to monitor for temperatures. And, of course, I kept the bulk of it looking at CPU temperatures.

Given what we are discussing and what you are saying, then I am even more confused as to why my GPU temps seem to have dropped with a front-mounted AIO. I check the temps when gaming, but I have not run a benchmark and monitored the temps throughout - yet. I do plan on doing that.
 
I like tinkering on my PC anyway even if it always takes me longer to do things than I anticipated.
I know that feeling all too well. Self-proclaimed that I have PC-OCD. I can't keep my hands out of my PC for even a few months...


I get torn - the 420 is too much, but I may be stuck with it or take a big loss. In anger, I purchased a 360 AIO in anger when I received the official return policy for the 420 - but I checked to make sure returns were also free this time. I could use a 360, but from my understanding in many instances, they perform similarly to 280mm AIOs. I suppose everything is situational and it will come down to people's set ups and brands.
Quoted For Truth.
AIOs have certain weaknesses that at a certain size(280mm), the bigger ones don't offer much more; you're in diminishing returns territory at that point.
1)The flow rates in comparison to custom loop, are low - crippled from the get go.
They'll be able to transfer heat to radiator only so fast. Not all AIOs have their flow rates advertised, but those that do are a little higher than what users may actually get; bends, twists, and turns will reduce flow rate further.

2)High radiator FPI(Fins Per Inch). Basically the air resistance presented by the radiator.
This one puzzles me. Higher FPI means faster, and thus LOUDER fans need to be used; in other words, they're performance is optimized around high(er) fan curves. No one likes too much noise, so they're going to run the fans lower.
Lower FPI rads are optimal with low(er) rpm fan curves, and lower means less possible noise. Faster fans doesn't change much for them.
Custom liquid radiators have high - low FPI models, so why don't more AIOs include lower FPI radiators?
My Alphacool Eisbaer Aurora 360 rad is 15FPI... AIOs from other brands use 21FPI. The heck?

3)Us/Our ears; individual sensitivity to sounds.
This goes in hand with #2. Like, how many people run their fans at 100% while sitting next to them? Almost none; they become too audible before that happens.
Everyone's going to run the fans within their own comfort zones, which may or may not be optimal for the AIO's performance. If more AIOs ran with lower FPI radiators, this would be less of an issue.
 
I know that feeling all too well. Self-proclaimed that I have PC-OCD. I can't keep my hands out of my PC for even a few months...



Quoted For Truth.
AIOs have certain weaknesses that at a certain size(280mm), the bigger ones don't offer much more; you're in diminishing returns territory at that point.
1)The flow rates in comparison to custom loop, are low - crippled from the get go.
They'll be able to transfer heat to radiator only so fast. Not all AIOs have their flow rates advertised, but those that do are a little higher than what users may actually get; bends, twists, and turns will reduce flow rate further.

2)High radiator FPI(Fins Per Inch). Basically the air resistance presented by the radiator.
This one puzzles me. Higher FPI means faster, and thus LOUDER fans need to be used; in other words, they're performance is optimized around high(er) fan curves. No one likes too much noise, so they're going to run the fans lower.
Lower FPI rads are optimal with low(er) rpm fan curves, and lower means less possible noise. Faster fans doesn't change much for them.
Custom liquid radiators have high - low FPI models, so why don't more AIOs include lower FPI radiators?
My Alphacool Eisbaer Aurora 360 rad is 15FPI... AIOs from other brands use 21FPI. The heck?

3)Us/Our ears; individual sensitivity to sounds.
This goes in hand with #2. Like, how many people run their fans at 100% while sitting next to them? Almost none; they become too audible before that happens.
Everyone's going to run the fans within their own comfort zones, which may or may not be optimal for the AIO's performance. If more AIOs ran with lower FPI radiators, this would be less of an issue.

In essence, you are saying that 280 is about the optimal size for fitting most cases and cooling since larger rads do not offer more cooling.

And yeah, the FPI on the arctic is supposed to be higher I believe and that should mean it is louder. Not sure I have seen that in reviews.
 
Oh hey, I looked around and found that the Liquid Freezer IIs are 15FPI too - though the thicker rad does offset this a little bit. So I guess it's closer to 17?


In essence, you are saying that 280 is about the optimal size for fitting most cases and cooling since larger rads do not offer more cooling.
They're like the best overall for AIOs. Many times, the fans are what set larger and smaller models apart; some models just win by brute forcing through the use of really strong fans. That'll win performance benches, but from a more practical view, that's not really a winner.
An older series of AIO, EVGA's CLCs, are still one of the best around for raw performance, but the fans will make it sound as if your PC is taking off... set the fans to a more tolerable level, and it's really on par with a lot of other units.
I guess if one needs the extra performance, it's there, but I wouldn't wanna be in the same room while that's going on...


A larger radiator/heatsink does have capacity for higher heat loads, but a number of factors act as roadblocks to keep this from working properly. It's only really going to be seen on custom liquid.
-flow rates.
-fan speeds.
-more is less. The more layers between the silicon and the cooler cold plate, the less efficient thermal transfer is. Desktop cpu cooling isn't as efficient at transferring heat than gpu, because there are currently 2 extra layers: Integrated Heat Spreader + solder, or IHS + paste.
-cooler mounting pressure.
-thermal density of the silicon. Smaller is harder to cool. Take TSMC's 12nm finFET(Ryzen 2000) Vs their 7nm finFET process.

Got an certain 12th gen i9 pushing into the realm of 320w? A 360mm AIO should do it, with a little bit of headroom to spare, right?
Yes, but actually no: it's already hitting 100C at the 230w mark...

Even though I have my 7820X on direct die, using Conductonaut LM, Eisbaer Aurora 360(T30-120 fans)... OC all cores to 4.5ghz, adaptive mode set to 1.237v with a -0.05 offset...
Run it through OCCT small data set, steady load, using SSE instructions... this thing's cores are getting up to 90C, even though HWINFO tells me the cpu is doing 200w max... 😵
 
Oh hey, I looked around and found that the Liquid Freezer IIs are 15FPI too - though the thicker rad does offset this a little bit. So I guess it's closer to 17?



They're like the best overall for AIOs. Many times, the fans are what set larger and smaller models apart; some models just win by brute forcing through the use of really strong fans. That'll win performance benches, but from a more practical view, that's not really a winner.
An older series of AIO, EVGA's CLCs, are still one of the best around for raw performance, but the fans will make it sound as if your PC is taking off... set the fans to a more tolerable level, and it's really on par with a lot of other units.
I guess if one needs the extra performance, it's there, but I wouldn't wanna be in the same room while that's going on...


A larger radiator/heatsink does have capacity for higher heat loads, but a number of factors act as roadblocks to keep this from working properly. It's only really going to be seen on custom liquid.
-flow rates.
-fan speeds.
-more is less. The more layers between the silicon and the cooler cold plate, the less efficient thermal transfer is. Desktop cpu cooling isn't as efficient at transferring heat than gpu, because there are currently 2 extra layers: Integrated Heat Spreader + solder, or IHS + paste.
-cooler mounting pressure.
-thermal density of the silicon. Smaller is harder to cool. Take TSMC's 12nm finFET(Ryzen 2000) Vs their 7nm finFET process.

Got an certain 12th gen i9 pushing into the realm of 320w? A 360mm AIO should do it, with a little bit of headroom to spare, right?
Yes, but actually no: it's already hitting 100C at the 230w mark...

Even though I have my 7820X on direct die, using Conductonaut LM, Eisbaer Aurora 360(T30-120 fans)... OC all cores to 4.5ghz, adaptive mode set to 1.237v with a -0.05 offset...
Run it through OCCT small data set, steady load, using SSE instructions... this thing's cores are getting up to 90C, even though HWINFO tells me the cpu is doing 200w max... 😵

So what is the best 280mm AIO currently in your opinion?
 
Thanks. Those were my general picks as well for the most part. I haven't looked into Alphacool much.
The cons I'm not a fan of with those coolers:
LF II:
Very short fan cables if the user doesn't want to utilize the splitter underneath the sleeved tubing.
That VRM fan is a gimmick and another point of failure; few are going to care to return this unit if this part breaks.

EK-AIO:
Rainbows everywhere.
Rainbows means more cables to manage; it can add up quickly with a lot of ARGB fans.

Eisbaer Aurora:
Even though the radiator's FPI is on the lower side, they went with some rather weak fans; hinders max performance. Took a step forward, and stepped right back... /facepalm
Rainbows means more cables to manage.
The tubing is a little on the rigid side;
 
The cons I'm not a fan of with those coolers:
LF II:
Very short fan cables if the user doesn't want to utilize the splitter underneath the sleeved tubing.
That VRM fan is a gimmick and another point of failure; few are going to care to return this unit if this part breaks.

Yeah, it probably is a bit of a gimmick. Despite that they have a good reputation and product. I see on the 420 that the tubes seem short. I haven't looked at the splitter, but I assume it's the same as other splitters and uses a master/slave. 4 and 3 pin thing.

EK-AIO:
Rainbows everywhere.
Rainbows means more cables to manage; it can add up quickly with a lot of ARGB fans.

Yeah on the elite a push/pull with 6 fans on 360 seems a bit much. I don't think the push pull does much for thermals, but I would have to test it. And yes, RGB cables kinda suck. I don't like that it's D-RGB, but that's my hang up. F that noise.

Eisbaer Aurora:
Even though the radiator's FPI is on the lower side, they went with some rather weak fans; hinders max performance. Took a step forward, and stepped right back... /facepalm
Rainbows means more cables to manage.
The tubing is a little on the rigid side;

On the plus side you can replace the fans. Tubing can be an issue.

They all have plusses and minuses and many things will depend on the user's home and setup. Right now I am stuck staring at case fans. I like symmetry so arctic fans are good - they have a Bionic model but the RGB is a ring around the fan. BeQuiet came out with their first RGB fans, and the specs and reviews are damn good. Granted, those fans are pricey. Not a huge fan of the RGB on those and debating if the cost is worth the tiny increase in specs and decrease in aesthetics.
 
The cons I'm not a fan of with those coolers:
LF II:
Very short fan cables if the user doesn't want to utilize the splitter underneath the sleeved tubing.
That VRM fan is a gimmick and another point of failure; few are going to care to return this unit if this part breaks.

EK-AIO:
Rainbows everywhere.
Rainbows means more cables to manage; it can add up quickly with a lot of ARGB fans.

Eisbaer Aurora:
Even though the radiator's FPI is on the lower side, they went with some rather weak fans; hinders max performance. Took a step forward, and stepped right back... /facepalm
Rainbows means more cables to manage.
The tubing is a little on the rigid side;

A thought came to mind as I was driving, and I wanted to ask your opinion. I started doing research on the diminishing returns of larger AIOs. Given the temperature differences I am seeing in tests, I wonder why manufacturers bother making 360 and 420 sizes at all. Are the benchmark tests online reviewers doing not indicative of the kinds of overclocking users do and that is where the larger AIOs come in handy? It seems to me that the tiny increases in performance do not even warrant the costs of producing and marketing the larger AIOs. I mean, why make them at all?
 
I see on the 420 that the tubes seem short.
They aren't..? The entire LF II line uses longer tubing than most other AIOs, at 450mm long. The usual length is 400mm. Less often seen are 380mm.
Now, depending on the case and some of the other hardware inside, that extra length can be a convenience, or an inconvenience.

Given the temperature differences I am seeing in tests, I wonder why manufacturers bother making 360 and 420 sizes at all. Are the benchmark tests online reviewers doing not indicative of the kinds of overclocking users do and that is where the larger AIOs come in handy? It seems to me that the tiny increases in performance do not even warrant the costs of producing and marketing the larger AIOs. I mean, why make them at all?
Money, and there's enough demand for them to turn a profit. They especially charge more cash for the very niche 420mm. [I was interested in them at first, because 140mm fans, but the limited case options was a major con.]
The larger sizes don't offer much beyond fans, and more fans or bigger fans isn't always more better.

Some review benchmarks be way off the mark. One of the biggest mistakes I've seen with them, IMO, is the cpu choice(s), or not utilizing a dummy heater. I'd think most want to see what the cooler can really do, so the reviewer should use a dummy heater or a 'big' cpu and stress test that. [Maybe they don't have access to such samples - I could cut 'em some slack for that.]
Intel's X-series is one of the best samples to use if a dummy heater wasn't available.

Others... I've seen quad core i5s - i7s(the Intel TIM wasn't helping), and later AMD Ryzen cpus, which don't really pull that much power, but are harder to cool due to the multi chip design and higher thermal density of said chips.
You can't really see what these really big coolers are/aren't capable of with chips like those. Users can't saturate those radiators with chips like those - they're going to run into other limits long before that.

Trying not to make the post too long, but more cooler reviews need to be done with i9s and dummy heaters... and I guess Ryzen shouldn't be left completely out of the loop.
Then viewers need to make their own judgements whether they 'need/want' such a cooler.

I do kinda regret getting 2 360mm AIOs, but my curiosity had gotten the better of me at the time. [I do have one 280mm on the side as a backup.]