This is a thing. Liquid cooling, without any chance of mess..

[lexluthermiester]

The only things you got right there are cost and installation.

No, he was spot on with every point. There are several threads here on TPU that have multiple examples of such.

 

[Vayra86]

The only things you got right there are cost and installation. My loop gets pulled apart every two years, other than that, it gets a drop of biocide every 6 months and dusting just like every other cooler.

They have more points of failure but barbs and clamps are not likely to fail. If you start getting into compression fittings and quick connects then yeah, you could almost say prone. Anyone that puts even basic effort into their loop will have no more problems and maintenance than aios or air.

You pull your loop apart every two years?

I've got air coolers doing just fine after five years of never having entered the case to even dust them. Right now. If something would EVER break, its a 10 dollar fan replacement.

The current one is in service for a year and I didn't open the case yet. I mean really, if you have dust filters and positive pressure, its hard to see dust accumulate and even harder to have it cause problems. And AIOs... they just slowly degenerate into crap and have limited service life. Good luck repairing or replacing parts there to keep something that works as advertised

 

[EarthDog]

I just found this from 2005...https://www.overclockers.com/thermosyphons-better-approach-to-cpu-cooling/

 

[Deleted member 185158]

You pull your loop apart every two years?

I've got air coolers doing just fine after five years of never having entered the case to even dust them. Right now. If something would EVER break, its a 10 dollar fan replacement.

The current one is in service for a year and I didn't open the case yet. I mean really, if you have dust filters and positive pressure, its hard to see dust accumulate and even harder to have it cause problems. And AIOs... they just slowly degenerate into crap and have limited service life. Good luck repairing or replacing parts there to keep something that works as advertised

My water block is modded to fit anything. Is older than a decade. All copper. Was from expandable ThermalTake Big Water SE. Still have the radiator as well. pump and res long gone, but for it's time pretty decent. Got 7 to 8 years from that pump.

You bought several air coolers through all them years. Spent just as much if you where to have bought a custom water loop one time. A lot of water cooling kit manufacturers will send you updated mounting brackets, or you design your own like I did.

There is no comparison between Liquid cooling (custom) to any market air cooler. IDC who you are, how big the cooler is and how many heat pipes it has. Water cooling is just better in a fist full of different ways.

 

[moproblems99]

No, he was spot on with every point. There are several threads here on TPU that have multiple examples of such.

Well, there are several threads about 2080ti's failing prematurely so I guess that means there is a manufacturing defect with 2080tis? The loops failed because most likely they weren't put together correctly. Or how about all the countless threads of people bricking GPUs by screwing up the flash? There are always going to be people that screw things up but when loops are done properly they are not prone to failure.

You pull your loop apart every two years?

I've got air coolers doing just fine after five years of never having entered the case to even dust them. Right now. If something would EVER break, its a 10 dollar fan replacement.

There are people who pull their systems apart to repaste yearly. There are people who upgrade systems more often then two years. I only took mine apart this time because I upgraded.

I've got air coolers doing just fine after five years of never having entered the case to even dust them. Right now. If something would EVER break, its a 10 dollar fan replacement.

I've had the same cooling loop since 2012 with the exception of the block. I could have reused the block if I wanted to by a mounting kit but I chose to upgrade the block. The only thing on my block that could fail in a different manner than your system is a pump or o-ring. If an o-ring fails, it is likely because of a poor installation such as the barb over-torqued. There just isn't a lot of stress on a cooling loop other than heating and cooling cycles and with a properly designed loop, is not a big difference between off and normalized.

And AIOs... they just slowly degenerate into crap and have limited service life. Good luck repairing or replacing parts there to keep something that works as advertised

No disagreements there. AIOs suck unless you have some niche reason to have one like clearance issues.

Edit:

Also, I'm definitely not going to argue that custom loops are for everyone. They definitely aren't. But they also don't have to be a maintenance nightmare or a failure machine. That said, most people don't rtfm or care about doing things right so half my arguments go out the window and become void. BUT, if a loop is designed and implemented correctly, it is not much more maintenance than an air cooler. It will never be as low maintenance as air but for most people system upgrades will happen more regularly than busting the loop open because you need to.

 

[Ruru]

I pull my loop apart for about every six months and I don't find that as a pain in the ass? The maintenance isn't that much of a deal.

 

[moproblems99]

I pull my loop apart for about every six months and I don't find that as a pain in the ass? The maintenance isn't that much of a deal.

My old loop I had setup so I could pull it all out in one piece to work on it. I could even leak test it outside the case and install the loop in once piece. I pretty much had an AIO without shitty parts. I didn't have the GPU in the loop which is why it was possible.

 

[Zach_01]

You are trying to convince one another for things that are pure preferable choises.
I like 280/360mm AIOs because:

1. They are less noisy than alot tower air coolers (Im running H110i with low fan rpm and low pump rpm)
2. I dont have to do anything complex to install it like a custom loop
3. I only have to maintain the rad from dust. Dont want to mess with disassembly of a custom loop and I surely dont need one.
4. I dont have a case so I dont have to consider ambient case temps or intakes/exhausts (max rad performance without heating other components)
5. I dont have to consider the surrounding components to the CPU socket and if the fit (like ram sticks) or heating up under the tower air cooler

Does these reasons make AIOs likable or suitable for everyone? ...NO!
Does it cost a lot for what it does? ...Yes!

It suits me for what I want...
Its nice to have a lot of options to choose from and everyone can do the same weighting the facts, needs, costs... etc... etc...

 

[lexluthermiester]

I pull my loop apart for about every six months and I don't find that as a pain in the ass? The maintenance isn't that much of a deal.

Yeah, but you enjoy that kind of thing as do I, not everyone does.

 

[moproblems99]

You are trying to convince one another for things that are pure preferable choises.
I like 280/360mm AIOs because:

1. They are less noisy than alot tower air coolers (Im running H110i with low fan rpm and low pump rpm)
2. I dont have to do anything complex to install it like a custom loop
3. I only have to maintain the rad from dust. Dont want to mess with disassembly of a custom loop and I surely dont need one.
4. I dont have a case so I dont have to consider ambient case temps or intakes/exhausts (max rad performance without heating other components)
5. I dont have to consider the surrounding components to the CPU socket and if the fit (like ram sticks) or heating up under the tower air cooler

Does these reasons make AIOs likable or suitable for everyone? ...NO!
Does it cost a lot for what it does? ...Yes!

It suits me for what I want...
Its nice to have a lot of options to choose from and everyone can do the same weighting the facts, needs, costs... etc... etc...

Also, I am not trying to convince anyone of anything other than custom loops do not need to be nightmare as they are being portrayed here.

 

[Zach_01]

...custom loops do not need to be nightmare as they are being portrayed here.

But they are up to a point, and the cost start from at least double of nice AIOs price. There are things to consider... setup and assembly, leaks, growing stuff, cleaning outside/inside and cost and others maybe...
Its not for everyone.

And I didnt mention anyone personally. I meant a lot of the posters here trying to convince one another

 

[lexluthermiester]

Also, I am not trying to convince anyone of anything other than custom loops do not need to be nightmare as they are being portrayed here.

I don't think that anyone said or implied that they're a nightmare, just that using them is an involved effort. Whereas solutions like this prototype have the potential to be effortless and very effective at the same time. See what I mean?

 

[Ruru]

My old loop I had setup so I could pull it all out in one piece to work on it. I could even leak test it outside the case and install the loop in once piece. I pretty much had an AIO without shitty parts. I didn't have the GPU in the loop which is why it was possible.

I also leak test outside and install it after I've tested it. Also much easier to fill and get all the possible air out from the loop. I'll never be using hard tubes..

 

[moproblems99]

Since we are using Linus, we may as well look at Tom's as well.

Toms: https://www.tomshardware.com/reviews/icegiant-prototype-thermosiphon-cooler/2

Tom's was using a 2990WX and was using single tower coolers as comparison (either incompetence or telling of how poor the cooling is of the IG) a Noctua U14S and the IG barely beat the single tower cooler (4C). That is REALLY poor cooling mass vs mass. Not impressive at all.

EDIT: And the IceGiant was 20dB louder. Over 2kg, 20 extra dB, $120 versus a .770kg single tower for half the price ($63) and you get an extra 4C of cooling performace. Wow! Revolutionary! Put 2 more fans on it and you can gain 1C. Although you probably lose that advantage buy putting an extra fan on the U14S.

 

[Zach_01]

Should I mention again that this is a prototype? That it will be reduced in size, weight and maybe increase performance from changing materials and/or by further optimized construction? Should we stay opedminded and just treat it with plain scepticism? Should dismissed it and deny it completely before we even see a final and retail product? if ever see one...

Im not saying that this will be a 100% success... Just dont get ahead of our selves...

 

[Jism]

Yeah, that's gonna be a lot of weight hanging on the board for sure.

The threadripper has a enforced socket and all that. That weight is nothing to what it can handle. This is no AM3 or so with just 4 mountingholes.

Impressive results. But i never had a AIO pump failing on me. If that happens the CPU Is hardware protected anyway that it would not burn out. It simply lower clocks or shuts down.

 

[moproblems99]

Should I mention again that this is a prototype? That it will be reduced in size, weight and maybe increase performance from changing materials and/or by further optimized construction? Should we stay opedminded and just treat it with plain scepticism? Should dismissed it and deny it completely before we even see a final and retail product? if ever see one...

Im not saying that this will be a 100% success... Just dont get ahead of our selves...

What are they going to do? Change it to all copper? Aluminum only has 60% of the of thermal conductivity so by changing to all copper they should be able to cut down on the size right? Well, aluminum is also 70% lighter than copper. So, if the change this to all copper, the weight is actually going to go up by 10% or so. 5 lbs of copper! 5 LBS! And guess what, copper is also 3 times more expensive than aluminum. So what will the final price be? $250? That is only double. For $250 you can get a starter custom loop package and smack the IceGiant around. Or for $65 you can buy an air cooler that is pretty much just as good. This thing is this big because it isn't efficient. And I don't see a market for it.

 

[Deleted member 185158]

What are they going to do? Change it to all copper? Aluminum only has 60% of the of thermal conductivity so by changing to all copper they should be able to cut down on the size right? Well, aluminum is also 70% lighter than copper. So, if the change this to all copper, the weight is actually going to go up by 10% or so. 5 lbs of copper! 5 LBS! And guess what, copper is also 3 times more expensive than aluminum. So what will the final price be? $250? That is only double. For $250 you can get a starter custom loop package and smack the IceGiant around. Or for $65 you can buy an air cooler that is pretty much just as good. This thing is this big because it isn't efficient. And I don't see a market for it.

Nah. Theyd just slap some little copper cold plate on the bottom and say its 10% better.
Then give the teen the cooler so he can claim water cooling is dead.

 

[EarthDog]

The threadripper has a enforced socket and all that. That weight is nothing to what it can handle. This is no AM3 or so with just 4 mountingholes.

But it supports all other sockets, some which are simply 4 holes for all intents and purposes.

 

[moproblems99]

But it supports all other sockets, some which are simply 4 holes for all intents and purposes.

I like to see the comparison to a 4.5 pound piece of solid copper or aluminum. Good ole passive cooling. I bet the performance isn't that far off.

 

[Zach_01]

What are they going to do? Change it to all copper? Aluminum only has 60% of the of thermal conductivity so by changing to all copper they should be able to cut down on the size right? Well, aluminum is also 70% lighter than copper. So, if the change this to all copper, the weight is actually going to go up by 10% or so. 5 lbs of copper! 5 LBS! And guess what, copper is also 3 times more expensive than aluminum. So what will the final price be? $250? That is only double. For $250 you can get a starter custom loop package and smack the IceGiant around. Or for $65 you can buy an air cooler that is pretty much just as good. This thing is this big because it isn't efficient. And I don't see a market for it.

You like to see (at least this thing) and speak of it with black and white, like there is no bilions of bilions grays in the middle of them... I cant help you much with that.
You are making way too many assumptions and drawing conclusions like its something you personally worked on and tested on, and tested different materials and configurations, and change its setup, and thought of it... and finaly decided and pridestinate its fate... Like the super expert on cooling and thermodynamics.

I cannot even begin to understand this kind of thinking (dont want either) and I'm totally opposed to it. I personally believe, if everyone was thinking like you there will be no innovation to anything...
And innovation comes with successes and failures too... when maybe someone else sees the causes of a failure and tries to fix them.

 

[Deleted member 185158]

Zach, it's not always what it's cracked up to be.

First, as noted, it's all aluminum. The heat pipes may help make up for some of that, along with the amount of mass.
But Aluminum heat soaks much quicker than copper from lack of density. And it just won't do well once that happens.
Otherwise, you'd see stock coolers built from all aluminum, (we do low end only) where there is a small BTU to dissipate.

If someone posted a DH that had a silver cold-plate upgrade, them maybe we'd be getting some interesting cooling results....
But this I fear would indeed be tested by average users with poor results depending on climate location most importantly.

 

[moproblems99]

You like to see (at least this thing) and speak of it with black and white, like there is no bilions of bilions grays in the middle of them... I cant help you much with that.
You are making way too many assumptions and drawing conclusions like its something you personally worked on and tested on, and tested different materials and configurations, and change its setup, and thought of it... and finaly decided and pridestinate its fate... Like the super expert on cooling and thermodynamics.

I cannot even begin to understand this kind of thinking (dont want either) and I'm totally opposed to it. I personally believe, if everyone was thinking like you there will be no innovation to anything...
And innovation comes with successes and failures too... when maybe someone else sees the causes of a failure and tries to fix them.

You seem to think that somehow having an opinion on something means that I am oblivious to everything else and will take that opinion to the grave. I am simply looking at what it is in front of me, the materials we have, thermal dynamics, and how absolutely far it has to go to be Wow! Products don't generally go out to review until they are pretty far along in development. This is not prototype 1. They think that it is good enough to send out to people for early review. They did a nice job with the frame and design and aren't likely going to trash a lot of that. And it isn't like this technology was invented yesterday since dog referenced an article from 2005. And again, they need to cut the weight by 2.5 just to get it inline with the Nh-D15 which is already considered massive.

Could this thing surprise me? Sure, and I hope it does. But as it stands now, it is not impressive. Just because I have an opinion doesn't mean that it can't change. There is just no good evidence to change it right now.

 

[Deleted member 185158]

You seem to think that somehow having an opinion on something means that I am oblivious to everything else and will take that opinion to the grave. I am simply looking at what it is in front of me, the materials we have, thermal dynamics, and how absolutely far it has to go to be Wow! Products don't generally go out to review until they are pretty far along in development. This is not prototype 1. They think that it is good enough to send out to people for early review. They did a nice job with the frame and design and aren't likely going to trash a lot of that. And it isn't like this technology was invented yesterday since dog referenced an article from 2005. And again, they need to cut the weight by 2.5 just to get it inline with the Nh-D15 which is already considered massive.

Could this thing surprise me? Sure, and I hope it does. But as it stands now, it is not impressive. Just because I have an opinion doesn't mean that it can't change. There is just no good evidence to change it right now.

Well your opinion is based off some pretty good facts. I'm in agreement with your comment.

Here's some facts about metals thermal conductivity.

Mind you fellas with nickel plated copper heat sinks or waterblocks. Lap the nickel off.

 

[Vayra86]

I don't think that anyone said or implied that they're a nightmare, just that using them is an involved effort. Whereas solutions like this prototype have the potential to be effortless and very effective at the same time. See what I mean?

Thank you! That was the point.

I get it, there are reasons to go custom water and they're good ones for those who do it. In the same way, I prefer the advantages that air brings over all the other solutions and this 'invention' could be a step up from that.