• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Air Cooling -- Myths and setup tips for the novice performance / gaming builder

Some pretty heavy copper coolers like the TRUE Copper (1900g *without* fan) worked alright, and as far as I know, no popular air cooler these days is anywhere near that heavy. Even big dual towers like the NH-D15 (1320g with fan) or the Deepcool Assassin III (1464g with fan) don't approach that kind of mass. It can be an issue for shipping, but good enough packing material can mitigate risk there too.
I know, I know. Like I said, it's a personal aversion on my part, not anything based in facts.
 
What most air-cooling aficionados overlook is that liquid cooling allows gives you a lot more control over your entire cooling setup, by virtue of the fact that the radiator can be located far away from the source.
It is not overlooked. It is just not relevant to those who prefer air cooling.

You suggest and imply there is some advantage to having some flexibility in radiator placement. NO WAY! The very fact you must (1) find a place to put the radiator and (2) route the hoses so they minimize impact on case cooling/air flow are HUGE disadvantages to liquid cooling.

I personally prefer liquid simply because I've never been comfortable with hanging big heavy blocks of metal off a motherboard.
A little bit of homework before buying, and after-assembly care mitigates any problems there.

A decently designed motherboard will have multiple mounting points around the socket to distribute those forces. A decently designed case will have multiple corresponding mounting points to absorb those forces. And of course, a decently designed cooler will come with a proper backplate to distribute those forces too.

Then after assembly, the user just needs to take a little care when moving and/or transporting the computer. If shipping/transporting the computer across town or beyond, lay the computer on its side, or simply remove the cooler until you reach your destination.

If lugging the computer outdoors to clean, you don't want to bounce the computer off the floor. But then I would not want to do that with liquid cooling either.
 
You suggest and imply there is some advantage to having some flexibility in radiator placement. NO WAY! The very fact you must (1) find a place to put the radiator and (2) route the hoses so they minimize impact on case cooling/air flow are HUGE disadvantages to liquid cooling.

There actually is an advantage in some cases. Some ITX cases are pretty constrained around the CPU area, but have space for a radiator.

A few examples:



 
You suggest and imply there is some advantage to having some flexibility in radiator placement. NO WAY! The very fact you must (1) find a place to put the radiator and (2) route the hoses so they minimize impact on case cooling/air flow are HUGE disadvantages to liquid cooling.

It's a lot easier to install a custom loop than it is an AIO, in a small case.
Choosing the size of the rad is part of it, but the simple fact that you get to choose where the pump goes and get the hose lengths exactly as you want them...

Even just the ability to turn my 3090 into a single slot GPU that's half the length is a massive plus on the custom water side.
 
What I don't understand about water cooling is the discrepancy of going with AIO for 105-250W CPU while using air on 300-600W GPU.
You see many examples of that sort.
I know that it is much easier to go water with the CPU, but if your much more power hungry GPU (that is stress out more than the CPU in gaming) is fine with air- why going water on the CPU in the first place?
 
What I don't understand about water cooling is the discrepancy of going with AIO for 105-250W CPU while using air on 300-600W GPU.
You see many examples of that sort.
I know that it is much easier to go water with the CPU, but if your much more power hungry GPU (that is stress out more than the CPU in gaming) is fine with air- why going water on the CPU in the first place?
That ones easy - because modern GPU's require stupidly complicated heatsinks with full coverage blocks, where 3+ years ago you were fine with just a GPU block and some passive heatsinks
I'd rather have a water-cooled GPU and an air-cooled CPU (my sons system was set up this way until the AIO on the G12 died)

I think since then, it's the higher CPU wattages that pushed people that way - a 250W CPU does indeed heat up everything inside a case, so venting it directly out was beneficial
 
What I don't understand about water cooling is the discrepancy of going with AIO for 105-250W CPU while using air on 300-600W GPU.
You see many examples of that sort.
I know that it is much easier to go water with the CPU, but if your much more power hungry GPU (that is stress out more than the CPU in gaming) is fine with air- why going water on the CPU in the first place?
Probably a part of it is because tubes look cool. If the popularity of RGB is anything to go by, aesthetics do matter to a lot of people.
 
There actually is an advantage in some cases. Some ITX cases
Ummm, I was responding to the issue made by Assimilator being uncomfortable having a big, heavy air cooler hanging off a motherboard, and thus the need to find a place to put a radiator.

And to his specific issue, I made the point of saying doing our homework and carefully selecting the case will mitigate that problem.

You simply illustrated my point by carefully selecting a case that resolves that issue. That said, your ITX example offers just one place for the radiator. I would hardly call that flexible.
 
Probably a part of it is because tubes look cool. If the popularity of RGB is anything to go by, aesthetics do matter to a lot of people.
Yep, my thoughts exactly.
Bigger crowd today expose himself, some unknowingly, to the potential problems of AIO in the name of "It looks cool" without any real thermal need.
To which its own I guess, but as long as you are aware to all pros and cons of your chosen product.
 
Having 1000g of metal plus the fans to go with it hanging off your board isn’t really a big deal.. just don’t drop it :)
 
Coming from a person with 34 years of experience in this field, who has been trained in air flow management, it seems everyone is missing some points in building a rig.

1. Every component is different. Even if they are the same type of component, this includes cases you must take them as being different if you re trying to be as efficient as possible in your airflow cooling management.

I've had to swap out same brand cooling fans because I was getting an unusual result in a certain part of a case.

2. Every build is different. Even buying the same components for multiple rigs can get you different results in cooling. I've seen this happen too often to state otherwise.

3. YOU MUST KNOW all of the strengths and weakness of all of your components that you use in your build. The perfect example is my MSI X570 A-PRO motherboard and how almost every major tuber craped on it because of hot MOSFED issues. Well yes that will happen if you are going to drop in a R9 5950 and overclock it. That is not what this motherboard is made for and I saw it's potential. For $125.00 out the door, you can beat that price. It now has a 5900OEM (before that a 3600) with its ram and Video card undervolted. It runs very cool and for my CPU cooler is... wait for it.... a Hyper T2... Yup a Hyper T2 instead of a big hunk of metal in the case.

I rig this rig 16 hours a day 7 days a week coming on 3 years. I am not beating my rig to death by overheating and saving my money by not buying expensive water cooling setups. These savings will be rolled over to my next setup or an upgrade on my video card.

Now all of my data on this rig has been posted previously, but I just did a quick shot with MWM on my temps as of this posting. Again 2 Artic 140mm fans side case. 1 120mm fan in the back. The PSU has been flipped to provide extra pushing out of air from the case and the Hyper T2 fan has been rotated 90 degrees to help pull air from the outside.

HWM11302022.jpg


4. Again as stated before. To get the best cooling possible you MUST take the time and patience to fine tune your rig to achieve the best results. This means knowing all to components outcomings, knowing which type of fan/case/cooling setup to be used and finally taking the time to to swap out parts to make the rig to run as cool as possible. It takes a long, long, time to fine tune your rig. And yes I do use a small fogger to track my air flow. If I get lazy then it is a"Punk" (a fire starter on a stick is called a punk) and eye balled the smoke with a flashlight. To fine tune this the first time it took me 2.5 hours which is why not many people do this anymore.

They just buy the components, slap it together and call it a day.


Now concerning the Water/Air thing. That's your choice and I can't tell you what you want in your rig.

However from my perspective and the dozen or so water cooling rigs that I had to clean, you will never EVER see me with a water based set up, because IMHO water set ups deteriorate over time... in certain cases of neglect badly. You must do general maintenance every year to get the best performance from water cooling. Air cooling... heh replace a fan if it goes bad in 5 or so years and blow out your case of dust every 3 months, which you have to do with a water based cooling method anyhow.

Again regardless if you go water or air. IMHO,... If you do not take the time to fine tune your air flow/cooling management, you are just asking for problems down the road.
 
But, But, more fans must mean more airflow!
 
Coming from a person with 34 years of experience in this field, who has been trained in air flow management, it seems everyone is missing some points in building a rig.
Well, I won't hold your obvious youth and minimal experience against you!

I also think you should go back and read what has been said, and follow the conversations because clearly, not everyone.

I will also comment about your claim,
IMHO water set ups deteriorate over time...
That's not an opinion. That's just a fact - though it applies to air cooling setups, and just about everything else too.

One of my problems with some water users is they are very diligent in performing inspections in a timely manner - at first. But after a year or two or three of never finding anything wrong, the time between inspections get longer and longer just when they should become more frequent. When new, fittings typically are nice and tight and hoses are soft and flexible. But over time, fittings can work loose and hoses become brittle.
 
What I don't understand about water cooling is the discrepancy of going with AIO for 105-250W CPU while using air on 300-600W GPU.
You see many examples of that sort.
I know that it is much easier to go water with the CPU, but if your much more power hungry GPU (that is stress out more than the CPU in gaming) is fine with air- why going water on the CPU in the first place?
GPU dies are much larger than CPU dies.

The bottleneck isn't thermal load, it's thermal density.

Air cooled CPUs run a lot hotter than air cooled GPU's for this reason.

Liquid cooled beats both, because you can't beat fast flowing coolant for quickly pulling heat away from a small contact surface.
 
Does anyone have any figures for VRM temperatures with CPU air cooling and then with water cooling? I am concerned that water cooling may result in higher VRM temperatures due to a loss of air flow.

As far as GPUs go, I use an air cooled GT-1030 without dedicated fans, and it seems fine.

One thing that seems risky is to replace a power supply fan with anything but an identical fan.
 

Attachments

  • GT1030.jpeg
    GT1030.jpeg
    95.2 KB · Views: 67
Last edited:
Coming from a person with 34 years of experience in this field, who has been trained in air flow management, it seems everyone is missing some points in building a rig.

1. Every component is different. Even if they are the same type of component, this includes cases you must take them as being different if you re trying to be as efficient as possible in your airflow cooling management.

I've had to swap out same brand cooling fans because I was getting an unusual result in a certain part of a case.

2. Every build is different. Even buying the same components for multiple rigs can get you different results in cooling. I've seen this happen too often to state otherwise.

3. YOU MUST KNOW all of the strengths and weakness of all of your components that you use in your build. The perfect example is my MSI X570 A-PRO motherboard and how almost every major tuber craped on it because of hot MOSFED issues. Well yes that will happen if you are going to drop in a R9 5950 and overclock it. That is not what this motherboard is made for and I saw it's potential. For $125.00 out the door, you can beat that price. It now has a 5900OEM (before that a 3600) with its ram and Video card undervolted. It runs very cool and for my CPU cooler is... wait for it.... a Hyper T2... Yup a Hyper T2 instead of a big hunk of metal in the case.

I rig this rig 16 hours a day 7 days a week coming on 3 years. I am not beating my rig to death by overheating and saving my money by not buying expensive water cooling setups. These savings will be rolled over to my next setup or an upgrade on my video card.

Now all of my data on this rig has been posted previously, but I just did a quick shot with MWM on my temps as of this posting. Again 2 Artic 140mm fans side case. 1 120mm fan in the back. The PSU has been flipped to provide extra pushing out of air from the case and the Hyper T2 fan has been rotated 90 degrees to help pull air from the outside.

View attachment 272296

4. Again as stated before. To get the best cooling possible you MUST take the time and patience to fine tune your rig to achieve the best results. This means knowing all to components outcomings, knowing which type of fan/case/cooling setup to be used and finally taking the time to to swap out parts to make the rig to run as cool as possible. It takes a long, long, time to fine tune your rig. And yes I do use a small fogger to track my air flow. If I get lazy then it is a"Punk" (a fire starter on a stick is called a punk) and eye balled the smoke with a flashlight. To fine tune this the first time it took me 2.5 hours which is why not many people do this anymore.

They just buy the components, slap it together and call it a day.


Now concerning the Water/Air thing. That's your choice and I can't tell you what you want in your rig.

However from my perspective and the dozen or so water cooling rigs that I had to clean, you will never EVER see me with a water based set up, because IMHO water set ups deteriorate over time... in certain cases of neglect badly. You must do general maintenance every year to get the best performance from water cooling. Air cooling... heh replace a fan if it goes bad in 5 or so years and blow out your case of dust every 3 months, which you have to do with a water based cooling method anyhow.

Again regardless if you go water or air. IMHO,... If you do not take the time to fine tune your air flow/cooling management, you are just asking for problems down the road.
You should be using HWinfo64 if you are using Ryzen. 5900X is very easy to cool when it is at stock settings.
 
GPU dies are much larger than CPU dies.

The bottleneck isn't thermal load, it's v.

Air cooled CPUs run a lot hotter than air cooled GPU's for this reason.

Liquid cooled beats both, because you can't beat fast flowing coolant for quickly pulling heat away from a small contact surface.
But good air cooler can achieve about the same (to slightly higher) temps as average AIO (say 240-280mm) and beat the weak ones on CPU`s.
So it is true that the temps on CPU is higher than GPU (and I take your word that it`s the thermal density doing, sound resendable), but that`s effects both water and air the about same and offer no real insight as to why favoring AIO.
Moreover, in the places where average AIO do get a few degrees lower over good air it usually doesn't translate to any meaningful (if any) performance difference, especially if gaming is the main use.
 
A good CPU cooler can only be good if it has a steady supply of fresh air. Just like an AIO, they benefit greatly from fresh air.
 
But what if the water CPU cooler blocks of the air supply to the VRMs?
 
Last edited:
Does anyone have any figures for VRM temperatures with CPU air cooling and then with water cooling? I am concerned that water cooling may result in higher VRM temperatures due to a loss of air flow.
Get one of the Arctic AIOs with a fan on the CPU block.

But what if the water CPU cooler blocks of the air supply to the VRMs?
How would it do that?
 
Just going to throw this out there, my Vetroo V5 with two fans in push/pull has seriously impressed me. Keep in mind I am using LGA 1700 Contact Frame, which probably helps 5-10 celsius.

I was going to go Noctua, this Vetroo was only meant to be a backup cooler... but honestly not sure I will upgrade it. I get 60 celsius in Witcher 3 at 1440p max settings 165 fps 165hz....

I mean, don't really need to upgrade that... I am sure some games are more demanding on CPU, but even then I expect 70-75 max... synthetic testing hits mid 80's, but its synthetic for a reason, I won't ever see that in a game.
 
Just going to throw this out there, my Vetroo V5 with two fans in push/pull has seriously impressed me. Keep in mind I am using LGA 1700 Contact Frame, which probably helps 5-10 celsius.

I was going to go Noctua, this Vetroo was only meant to be a backup cooler... but honestly not sure I will upgrade it. I get 60 celsius in Witcher 3 at 1440p max settings 165 fps 165hz....

I mean, don't really need to upgrade that... I am sure some games are more demanding on CPU, but even then I expect 70-75 max... synthetic testing hits mid 80's, but its synthetic for a reason, I won't ever see that in a game.
That's the deal!

Is the cooler good for you or good for your CPU??
Pretty sure that's what stress testing is about.
Well halfway though stress testing is also a good indicator of stability which thermal productivity has a major influence upon.

7 pages of rubbish. I've yet to find a place to fit in.

ezgif-4-7265a72910.gif
 
That's the deal!

Is the cooler good for you or good for your CPU??
Pretty sure that's what stress testing is about.
Well halfway though stress testing is also a good indicator of stability which thermal productivity has a major influence upon.

7 pages of rubbish. I've yet to find a place to fit in.

View attachment 272327

It's good for a 13600kf. I wouldn't use it on a higher chip though. Under most brutal prime95 load I could throw at it, it hit 97 celsius... that was its max... I have it set to 102 celsius limit in bios.

That's extreme though... I mean I won't ever see more than 60 celsius in most games I play. so eh
 
Back
Top