Review IceGiant Prototype Thermosiphon Cooler Review: Frosty New Tech

[Admin]

Sporting thermosiphon tech and a massive towering shell, the IceGiant prototype holds promise for performance cooling.

IceGiant Prototype Thermosiphon Cooler Review: Frosty New Tech : Read more

 

[nofanneeded]

Motherboard specs should change ... the mounting holes for the CPU cooler should be mounted on the case itself and not o the motherboard . If they do this you can add a huge heatsink eveen upto 2KG without the fear of breaking the motherboard.

 

[redgarl]

Motherboard specs should change ... the mounting holes for the CPU cooler should be mounted on the case itself and not o the motherboard . If they do this you can add a huge heatsink eveen upto 2KG without the fear of breaking the motherboard.

It is impossible since all the case are different and doesn't follow standards.

 

[nofanneeded]

It is impossible since all the case are different and doesn't follow standards.

It is possible , the motherboard mounting screws are followed standard already ,and it is just 4 mounting holes more on the case it self where you mount the heat sink screws.

However this will fix the CPU position on the Motherboard it self .. and I think this is not a bad idea at all , they can choose the best CPU position and make it fixed place , just like the slots places for example..

 

[rubix_1011]

This cooler is 2.2kg and it didn't break the motherboard.

 

[USAFRet]

Motherboard specs should change ... the mounting holes for the CPU cooler should be mounted on the case itself and not o the motherboard . If they do this you can add a huge heatsink eveen upto 2KG without the fear of breaking the motherboard.

And how would one transfer the heat load from the CPU face to the cooler?

Long rigid pipes would not work, because no two cases are identical.
Flexible pipes...Hey, we just reinvented the AIO liquid cooler.

 

[wexir]

man i'd love to put it on top of my r5-1600 and see how far can it clock haha

i wonder if it would be possible to have extra brackets that lets's say you could screw to the top of the case so it would lower the stress on the motherboard

 

[BaRoMeTrIc]

It is possible , the motherboard mounting screws are followed standard already ,and it is just 4 mounting holes more on the case it self where you mount the heat sink screws.

However this will fix the CPU position on the Motherboard it self .. and I think this is not a bad idea at all , they can choose the best CPU position and make it fixed place , just like the slots places for example..

Do you mean to mount the cooler on the motherboard standoffs? even though the motherboard is already occupying those standoffs, and they are staggered in 3 rows of 3 on an ATX mobo?

 

[hotaru251]

so.....its a giant rad pump combo that sits on cpu.

 

[rubix_1011]

No, it is much more similar to a heatpipe cooler. There is not an internal pump...the fluid inside is distributed by thermal convection, and the fluid is designed to change phases rapidly to allow for heat to be expelled quickly.

 

[nofanneeded]

Do you mean to mount the cooler on the motherboard standoffs? even though the motherboard is already occupying those standoffs, and they are staggered in 3 rows of 3 on an ATX mobo?

There will be extra four standoffs in my Idea for the cooler just below its place.

 

[nofanneeded]

And how would one transfer the heat load from the CPU face to the cooler?

Long rigid pipes would not work, because no two cases are identical.
Flexible pipes...Hey, we just reinvented the AIO liquid cooler.

It is the same design , Just screw the cooler on the stand offs on the case instead of the bracket that is fixed behind the motherboard.

today design the motherboard carries the cooler . my way the case carries the cooler. there should be four stand offs behind the four holes on the motherboard that we use to mount the heatsink. that is , make the case carry it not the motherboard itself. and you will not need that cutout anymore to replace the heatsink.

 

[jimmysmitty]

It is the same design , Just screw the cooler on the stand offs on the case instead of the bracket that is fixed behind the motherboard.

today design the motherboard carries the cooler . my way the case carries the cooler. there should be four stand offs behind the four holes on the motherboard that we use to mount the heatsink. that is , make the case carry it not the motherboard itself. and you will not need that cutout anymore to replace the heatsink.

While its plausible it not something they would do. Mainly due to the fact that they would have to standardize this while also cutting more holes into the board and thus having to redesign and reroute traces for the board around the holes. Every inch on a PCB is valuable space. Its why as boards get more complex we get more layers.

Then it would also force Intel and AMD into a maximum socket size which would stop them from being able to throw a larger socket into the board design for more memory channels or larger CPUs.

Then we have to account for ITX designs which give even less space for everything.....

 

[nofanneeded]

While its plausible it not something they would do. Mainly due to the fact that they would have to standardize this while also cutting more holes into the board and thus having to redesign and reroute traces for the board around the holes. Every inch on a PCB is valuable space. Its why as boards get more complex we get more layers.

Then it would also force Intel and AMD into a maximum socket size which would stop them from being able to throw a larger socket into the board design for more memory channels or larger CPUs.

Then we have to account for ITX designs which give even less space for everything.....

The motherboard already has four holes cut for the cooler ... as for the bigger sockets they have monting screws and can be made holes as well ..

The Case can have multiple holes for standoffs .. four holes for small socket and four holes for large socket , or a grid of holes , and you put the four standoffs where you want , and the motherboard manufacturer follow the grid of wholes when postioning the socket.

Actually you can do it yourself for this motherboard , get a case without socket cutout , and locate these four holes on the case with a pen , get a drill and drill four holes , and put standoffs there , and use them to mount the CPU cooler instead of the supplied bracket that comes with the cooler. you can mod it . be sure that the stand offs are compatible with the cooler mounting screws.

 

[nofanneeded]

This cooler is 2.2kg and it didn't break the motherboard.

Try moving the case around with a 2.2KG cooler and see what will happen. I ve seen 1.5KG break ..

thats why some case makers add a support mechanism for large coolers.

 

[rubix_1011]

I moved the case several times around my office, didn't seem to be an issue with their mounting collar. It is far more substantial than other mounting brackets.

Any PC component can break if moved violently enough. If you are just trying to be right, sure, you can win.

 

[AtotehZ]

It is impossible since all the case are different and doesn't follow standards.

It's not impossible. There's a point of contact between case and motherboard that has been standard for decades. The I/O.

So if you have mounting holes at the top of the case with the I/O as a reference, you'll have a limited number of variants to take into consideration. The same way you do with current CPU sockets.

The rest is just developing the mounting mechanism. My suggestion would be 2-4 rods being mounted at the top of the case, going down around the cooler, giving it stability and taking the load of the motherboard. They would have to have some fine-tuning available of course.

 

[AtotehZ]

And how would one transfer the heat load from the CPU face to the cooler?

Long rigid pipes would not work, because no two cases are identical.
Flexible pipes...Hey, we just reinvented the AIO liquid cooler.

Noone was talking about disconnecting the Cooler from the board. Everything is fine as it is now, except the motherboard shouldn't be carrying the load.

 

[AtotehZ]

After some consideration I think the best system would be 1-2 rails at the top of the case, both going front to back.. Then you insert the mounting mechanisms into the rail and tighten them in the right position. At this point you're dialed in on the X and Y axis. X(side to side) because of the I/O and Y(front to back) because of the rails. Now you just need rods in the correct length and clamps or whatever to hold the cooler at the correct point on the Z axis..

 

[nofanneeded]

I moved the case several times around my office, didn't seem to be an issue with their mounting collar. It is far more substantial than other mounting brackets.

Any PC component can break if moved violently enough. If you are just trying to be right, sure, you can win.

Yea right moving 1/2 kilos will break as fast as moving 2.5 kilos right ? if moved violently? sorry this is bad physics. you will need more power to break the smaller one.

This needs testing not just few moves around , btw the 2.2 KG is way off limit the recommendation standard. I remember reading in specs sheets some ~1KG maximum limit standard for coolers...

No one will sell you built PC with this cooler. it is out of specs, by two folds.

 

[rubix_1011]

Remember, this is a prototype. It is not what the retail version will be, as I mentioned, it will be much smaller in size yet reportedly perform better due to 3 condenser cores vs 2 in the prototype.

Also, recall that the TR4 socket has built-in mounts for coolers, not just holes for a backplate. Dispersion of this weight over a larger socket area allows for better retention and support than having the mounting holes closer on smaller sockets. It's the same reason you can mount a large TV on drywall if you have a larger mounting bracket which covers larger area vs. just having it hang on a single, small foot.

 

[nofanneeded]

Remember, this is a prototype. It is not what the retail version will be, as I mentioned, it will be much smaller in size yet reportedly perform better due to 3 condenser cores vs 2 in the prototype.

Also, recall that the TR4 socket has built-in mounts for coolers, not just holes for a backplate. Dispersion of this weight over a larger socket area allows for better retention and support than having the mounting holes closer on smaller sockets. It's the same reason you can mount a large TV on drywall if you have a larger mounting bracket which covers larger area vs. just having it hang on a single, small foot.

This is guessing ... stop guessing and contact AMD and ask about their socket recommended max cooler weight.

you can contact them easier than me. because you review stuff and they will answer your emails faster.

Intel also has screws on the socket for the HEDT and their max weight as well is ~1KG

 

[rubix_1011]

This is guessing ... stop guessing and contact AMD and ask about their socket recommended max cooler weight.

you can contact them easier than me. because you review stuff and they will answer your emails faster.

Intel also has screws on the socket for the HEDT and their max weight as well is ~1KG

What am I 'guessing' about? Nothing in what I said and what you quoted has anything to do with the spec for AMD Threadripper socket support in weight; this is your argument.

Re-read the content I've posted: it discusses the prototype cooler mass and dimensions as well as decrease in size and weight of the prototype cooler and the retail cooler as well as spreading a larger weight over a larger area. Nothing of what I said has anything to do with the cooler being in-spec or out of spec.

Again, the cooler tested was a prototype and will not be available for retail purchase as tested.

Your argument is whether a prototype cooler falls within specs (which aren't easily found) and I've already provided evidence (in the form of the manufacturer/vendor website + data) stating that the cooler will be much smaller for end-users. You are debating a point which serves little purpose.

If you want to know, I would recommend that you contact AMD.

 

[Ia21J^CfxIMeDu@ZUVGP]

It is possible , the motherboard mounting screws are followed standard already ,and it is just 4 mounting holes more on the case it self where you mount the heat sink screws.

However this will fix the CPU position on the Motherboard it self .. and I think this is not a bad idea at all , they can choose the best CPU position and make it fixed place , just like the slots places for example..

Here's the problem with that: Is the optimal CPU position the same for TR4 as it is for LGA 1151? Remember, one typically has DIMM slots on both sides of the socket, while the other doesn't. Also, would this "optimal" positioning be the same for every single standardized form factor? I'm guessing no.

In addition, how are you addressing mounting hardware? Are you saying every case should be able to mount everything from sTR4 to LGA 1151 to AM4 to LGA 2011v3? Would you just machine in all of the mounting hardware, increasing the cost of every single case, would you include adapters for every single socket, increasing the cost of the case, or would you just break up the product stack where you can only install one socket type into a case? Hint: None of those are good ideas.

This would also make installation a gigantic pain in the bum. Are you suggesting every single case should have a removable motherboard tray (increasing the cost more) to make installation easy? Or do you have to install the motherboard and CPU, then try to finagle a 2+ kg cooler into the case without busting anything? That sounds worse than the solution we already have.

Lastly, how many people are using 2+ kg tower coolers? Scratch that, how many 2+ kg tower coolers even exist? You're talking about upsetting the entire motherboard/case industry for a tiny fraction of edge use cases. In what world would be completely redesigning your entire product stack to accommodate such a small group of people be financially advantageous?

 

[AFDsads89la9Z]

It is impossible since all the case are different and doesn't follow standards.

https://en.wikipedia.org/wiki/ATX

Not sure how people can't know there is a standard.

Seems odd so many people are hung up on the weight of the cooler. Torque is probably the word you should be using here, as the weight and the height of the unit becomes a force that can crack traces on the motherboard. Seeing how there's a solid mounting bracket for this thing then they are obviously looking to compensate for some of the flex with that. I have seen really light but tall heat sinks crack a motherboard before because users dont really grasp how flexing works.

The ATX spec does dictate roughly where placement of components need to go. For the spec to allow for through-hole mounting of heat sinks it would require the CPU placement to be the same on all motherboards. If an ATX spec was that specific then I can't see manufacturers being overly happy with that.

I previously had a Thermoiphon cooler and it was great, allowed passive cooling as the heat moved away quite nicely. I have seen DIY based units that exit the case into a tall finned cooler and they work well. So while I think it's good that a company is now making this product, it seems odd to me and the tech isn't that groundbreaking.

The real "feature" of this unit is that it is able to be mounted horizontally or vertically with no loss (tested that anyone?) between the two modes.

I can't see why it would outperform water cooling at all. If they do make it smaller for production then Water Coolers are going to beat it, tho a simple all in one solution like this without the worry of leaks is a huge plus to some people.