Friday, January 13th 2023

Cooler Master Shows Off Cooling X, a Unique Prebuilt with Whole-body Liquid Cooling

Cooler Master at the 2023 International CES showed off the Cooling X, a patent-pending unique desktop PC architecture by the company where the liquid cooling of the CPU and GPU isn't confined to a couple of closed loop AIOs, but rather a giant liquid-cooling loop that even involves the side-panels of the case itself, that double up as additional heat-dissipation surfaces. Notice how some compact fanless cases use extruded-aluminium body panels that double up as heatsinks, to cool the hot components? The Cooling X re-imagines this, where instead of dissipating heat from a heatpipe, the extruded-aluminium body panels have coolant channels, so some of the heat from the liquid-cooling loop is dissipated. These panels supplement a conventional liquid-cooling radiator that's located along the rear panel, which has active ventilation. The liquid-cooling loop cools the processor and GPU.

For its compact dimensions of just 266 mm x 149 mm x 371 mm (LxWxH), the Cooling X prebuilt packs some mighty high-end hardware—an AMD Ryzen 9 5950X 16-core processor, and a Radeon RX 6800 XT GPU up to 64 GB of dual-channel DDR4-3200 memory, two 2 TB M.2 NVMe SSDs, and a homebrew 850 W 80 Plus Gold SFX power supply. For now Cooling X is just a codename, and Cooler Master intends to take this concept forward with high-performance gaming/creator prebuilt desktops sold under its own marquee.
Add your own comment

56 Comments on Cooler Master Shows Off Cooling X, a Unique Prebuilt with Whole-body Liquid Cooling

#1
FreedomEclipse
~Technological Technocrat~
Would be interesting to have this case in for review.
Posted on Reply
#2
Assimilator
Nothing special about this... you can run that hardware in a NZXT H1. Maybe if CM had used components that are known power and heat hogs like Intel CPU and NVIDIA GPU, I'd be more impressed.
Posted on Reply
#3
HBSound
I would like to see this case pulled apart.
I saw another company with this same concept based on using your own equipment.
Posted on Reply
#4
natr0n
Already thinking about standing fan mod.
Posted on Reply
#5
Unregistered
AssimilatorNothing special about this... you can run that hardware in a NZXT H1. Maybe if CM had used components that are known power and heat hogs like Intel CPU and NVIDIA GPU, I'd be more impressed.
Yeah like the new 13900ks.
#7
bug
It's hard to tell from that picture alone, but it looks like another solution that doesn't care about VRMs at all.
Posted on Reply
#8
TechLurker
This isn't exactly new, Thermaltake did similar with their SwordM and TaiChi cases from the 00s, though they didn't utilize the aluminum case as well as this, but still advertised it as helping the PC to stay cooler by having the body "absorb" some of the ambient heat and radiate it out, while the built-in AIO helped cool the CPU (and the GPU if one did early GPU WB cooling and spliced it into the line). Then there are all those massive "pure passively cooled" cases that are big chunks of extruded aluminum with copper heatpipes connecting to the CPU and GPU.

The main issue for all of those, and this case included, is the lack of cooling for the VRMs and potentially other elements such as NVMes.
Posted on Reply
#9
bug
TechLurkerThis isn't exactly new, Thermaltake did similar with their SwordM and TaiChi cases from the 00s, though they didn't utilize the aluminum case as well as this, but still advertised it as helping the PC to stay cooler by having the body "absorb" some of the ambient heat and radiate it out, while the built-in AIO helped cool the CPU (and the GPU if one did early GPU WB cooling and spliced it into the line). Then there are all those massive "pure passively cooled" cases that are big chunks of extruded aluminum with copper heatpipes connecting to the CPU and GPU.

The main issue for all of those, and this case included, is the lack of cooling for the VRMs (moreso the pure passively-cooled cases, while Thermaltake's cases could at least mount a fan or two blowing in and exhausting out the radiator).
Just nitpicking here, but "absorb" is the wrong word. Aluminum isn't that good at absorbing heat, it's good at conducing it. Which is fortuitous in this case (pun), as you don't want the case heating up, you want it getting rid of heat asap.
Posted on Reply
#10
Dammeron
bugAluminum isn't that good at absorbing heat, it's good at conducing it.
That's only 1 of 3 problems here. 2nd is low area of cooling fins (although those only complement the big radiator inside, which brings another question - why just not keep the internal one and ditch the other?!!!) and 3rd and most important - alloy galvanically reacts with copper/brass (even anodized one, just slower) which forces every other piece of the whole loop to be made of alloy too, lowering the performance even more!
Posted on Reply
#11
EatingDirt
AssimilatorNothing special about this... you can run that hardware in a NZXT H1. Maybe if CM had used components that are known power and heat hogs like Intel CPU and NVIDIA GPU, I'd be more impressed.
I mean, it's smaller than the NZXT H1 in both height & width. It's deeper but that's arguably the least important dimension(width takes up the most important desk space).

It's also a prebuilt, so it doesn't really 'compete' with the H1. It's just a compact high-performance prebuilt... so it competes with, whomever is in that space.
Posted on Reply
#12
trsttte
Dammeron3rd and most important - alloy galvanically reacts with copper/brass (even anodized one, just slower) which forces every other piece of the whole loop to be made of alloy too, lowering the performance even more!
With so much cooling capacity, a bit less efficiency from copper vs aluminum doesn't make much of a difference. I think it's interesting, could make for an interesting offering as a custom case instead of a pre built.
Posted on Reply
#13
robot zombie
bugJust nitpicking here, but "absorb" is the wrong word. Aluminum isn't that good at absorbing heat, it's good at conducing it. Which is fortuitous in this case (pun), as you don't want the case heating up, you want it getting rid of heat asap.
Wait, theres a difference between an object absorbing heat and an object conducting it? Are we talking heat flux vs capacity? Im a bit confused by what the difference is supposed to be. Absorbtion, to me, is just a generic term for what physics calls conduction. A material that conducts heat well is not a sponge for heat, but the transfer that happens with surface to surface conduction is superficially analogous. The heat transfers from the hot surface to the cool one, like water to a dry sponge.

Sorry, just pickin nits. I think I know what you're saying. Due to its lower relative density, it has a lower heat capacity in spite of rapidly taking on heat transfer. So unless you are constantly pumping it with lots of heat, it will lose its heat/reach equilibrium faster than steel or copper. A suitably large panel would take and dissipate the heat too quickly to reach a high temperature itself.

Am I correct here, or is there something else Im missing?
Posted on Reply
#14
cvaldes
AssimilatorNothing special about this... you can run that hardware in a NZXT H1. Maybe if CM had used components that are known power and heat hogs like Intel CPU and NVIDIA GPU, I'd be more impressed.
There are several considerable differences between the Cooling X and the NZXT H1 (whether it be V1 or V2).

Only the CPU in the H1 gets the AIO treatment. The H1 AIO is technically a 140mm unit but the pump is located in the middle of the radiator (not in the waterblock head), thus cutting down some of the cooling capacity. Real world performance brings it closer to a conventional 120mm AIO.

More importantly the H1 does not offer liquid cooling for the GPU. The graphics card compartment is separate and made possible by a short 90 degree PCIe extension cable. This compartment is pretty narrow so it cannot accommodate 3-slot graphics cards. The V1 model really could only support 2-slot cards like Turing FE cards. So basically you have a graphics card with a conventional air-cooled stock heatsink blowing around hot air inside the chassis. This does not promote desirable GPU thermals.

The Cooling X proposes a liquid cooling loop for both the CPU and GPU. That's a MAJOR difference that cannot be ignored.

You can hack up GPU liquid cooling for the H1 if you move most of the cooling elements (radiator, fans, pump) outside the case. This is actually achievable because the back panel grill on the H1 has perforations that conveniently accommodate the 105mm screw spacing for a 240mm radiator.
Posted on Reply
#15
Upgrayedd
@Xex360 Or any unlocked zen4 chip. Literally anyone of them. Cause y'kno.. they're built with 95C in mind but 115C as the ceiling.
Posted on Reply
#16
JAB Creations
How many hard drive bays? Not all of us sit around only playing video games all day and we also all don't want a NAS.
Posted on Reply
#17
claes
Itx is not for you my friend
Posted on Reply
#18
Veseleil
JAB CreationsHow many hard drive bays? Not all of us sit around only playing video games all day and we also all don't want a NAS.
Prepare to be boarded...
Posted on Reply
#19
bug
robot zombieWait, theres a difference between an object absorbing heat and an object conducting it? Are we talking heat flux vs capacity? Im a bit confused by what the difference is supposed to be. Absorbtion, to me, is just a generic term for what physics calls conduction. A material that conducts heat well is not a sponge for heat, but the transfer that happens with surface to surface conduction is superficially analogous. The heat transfers from the hot surface to the cool one, like water to a dry sponge.

Sorry, just pickin nits. I think I know what you're saying. Due to its lower relative density, it has a lower heat capacity in spite of rapidly taking on heat transfer. So unless you are constantly pumping it with lots of heat, it will lose its heat/reach equilibrium faster than steel or copper. A suitably large panel would take and dissipate the heat too quickly to reach a high temperature itself.

Am I correct here, or is there something else Im missing?
There is a difference. Each material has a capacity to store heat and a capacity to move heat from hot towards cold. Copper, for example, can store more heat than aluminum can, but aluminum will move heat away faster.

There are different measure units for both of them:
en.wikipedia.org/wiki/Thermal_conductivity
en.wikipedia.org/wiki/Heat_capacity
Posted on Reply
#20
trsttte
bugCopper, for example, can store more heat than aluminum can, but aluminum will move heat away faster.
I believe you got those reversed there
Posted on Reply
#21
robot zombie
The way I understand it, aluminum has a much higher absolute heat capacity, meaning it takes more heat (pure joules/watts/whatever) to raise the temperature. Take two heatsinks, one of each material, taking up the same mass, cooling the same part over the same stretch of time. The copper one will be much hotter, because it has a significantly lower heat capacity and thus will start to reach higher temperatures sooner. Copper also has MUCH better thermal conductivity, also contributing to it heating up faster. Because of this, even though it is hotter, it's also still transferring more heat overall.

The reason the aluminum seems not to hold onto as much heat is because it's considerably less dense and in a PC, sizes are limited. So an aluminum cooling part might actually seem to lose its heat faster when it is actually just transferring less heat for its size. A copper cooler of the same size would serve as a much larger reservoir, to the point where it might actually heat up slower, even though it is transferring heat faster, because it has that much more mass to spread it across. It's like, which heats and cools faster between a steel thimble, and a steel girder? The aluminum still has an overall lower heat flux between it and the source of heat, compared to a copper part with the same contact surface area and mass. However, that is to say that if you heat two chunks of copper and aluminum that are equal in mass to the same temperature, the copper should indeed cool down quicker, because it's holding onto less heat energy than the aluminum at that temperature, and transfers heat quicker.

I always figured that was why people say copper is always the more ideal material for cooling. It's just also more expensive, and impractical for some parts, such as case panels. Compared to steel, aluminum is still miles better. A computer case made out of all copper sheeting would be bitchin, though.
Posted on Reply
#22
A Computer Guy
Why can't they just make an entire side panel a rad instead and strap on 4 RGB 200mm fans?
One of the 4 dead spots (where a fan center sits) can be where the pump can be embedded into the rad.
One of the 4 dead spots (where a fan center sits) can be the reservoir.
One of the 4 dead spots (where a fan center sits) can be a flow indicator.
One of the 4 dead spots (where a fan center sits) can be temp/flow display.
Posted on Reply
#23
thesmokingman
A Computer GuyWhy can't they just make an entire side panel a rad instead?
That would be too logical.
Posted on Reply
#24
claes
It’s too small for 200mm fans friends. The side panels act as reservoirs.
For its compact dimensions of just 266 mm x 149 mm x 371 mm (LxWxH)
Posted on Reply
#25
robot zombie
Forget the logical stuff. I'm stuck on solid copper rear side panels now.

I want a coolant-dispersing manifold integrated into the inner face of the rear side panel. Pill-shape contour for the tubing would be better for spread, though far worse for flow. Squish round tubing semi-flat to get the picture. That's snaking as tightly as physically possible up and down the whole inner face of the panel. It can be modular, with scattered ports for super short in/out runs to different components across the mobo. Tubing pops right out of the back at the closest possible point to each block right from the passive radiator reservoir manifold that is the whole back panel. Full copper blocks for your vrms and ram. You know the tubing is copper and the connectors are copper-plated. Better just go for the matte black silicon and let all of the copper sing the rest.

Basically what you then have is a full mobo block flipped inside out, passively cooling your liquid via the case itself with nothing extra covering up the components inside. And probably the best coolant temperatures ever.

The idea is that the coolant contacts as much surface on the panel as possible. That's also your res. Imagine how much coolant a side-panel's worth of tubing holds, right? How bout that? Might need some extra push to get the coolant to go through it. Lets stack impellers where a full size res *would* be in a normal loop to be safe. Case height hand-engraved copper monolith for the housing on the pump stack with an internal gap to make room for some sound insulation. And then maybe it goes to a wall of rads on the front side panel to handle whatever heat is left by the panel.

How's that for logical? It's like if a laptop cooler made a pact with The God Hand, became the next Apostle, and took the corporeal form of a whole-ass liquid-cooled hybrid passive/active radiator case made out of fine polished copper.

Lets go further and try to eliminate the rads completely, have an unobstructed view of that clean layout. Can we make the partition that the mobo mounts onto a 'coarse' copper mesh? And can we then mount two oversized fans stretching the entire span of that partition, blowing from front compartment to rear? And can we get them *just* far back enough from the rear side panel to send the air across the whole manifold and then directly out through a generous mesh strip running along the top, bottom, front and back of the rear compartment this forms? Can we maybe get fins on the tubing of the integrated manifold to assist transfer?

Somewhere in the case, there will need to be a sizeable passive air intake for the fans to draw from. It's kinda okay if it's bottlenecking the fans just a little bit - potentially more stable flow when the max resistance is at entry. Only problem would be potential resonance through the intake. Could always mesh the top out as well if we need more intake. It's intrinsically negative pressure, but we'll go the extra mile to seal it and filter that intake. All of the removable panels clamp into place to seat some nice fat seals into matching grooves. Maybe the whole bottom of the visible compartment is your intake mesh. Maybe have some blockage across the bottom couple inches of the meshed mobo partition to linearize the flow path a bit more. Or do solid trim around the edges of the mobo partition to separate all of the outlet-side compartment's exhaust. The mobo will have to stand further off to let the air through the mesh, and you'd want taller feet too. But nobody said this thing has to be practical. Honestly, if we're making it out of expensive copper sheeting, we can afford to make it as big as it needs to be. Lian Li style partitioning with the spacious back panel hiding integrated manifold, fans, and tubing with plenty of space for airflow. Like, maybe even wide enough in the back to stand an ATX psu on its side.

I mean... at this point, make the OUTSIDE of that back panel a full passive radiator with a full-span fin array. Because why not?


Can I maybe convince a youtuber to try this for educational purposes? Hrmmm...
claesIt’s too small for 200mm fans friends. The side panels act as reservoirs.
This is why I like my idea better. Using those panels as reservoirs is a good plan. But in my plan 200mm fans aren't half big enough! Therefor, it is better.
Posted on Reply
Add your own comment
Dec 21st, 2024 11:40 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts