RTX 5090 FE Liquid Metal Swap: Thermal Paste Performs Just Fine

[usiname]

Are there really differences between different liquid metals?

There are many brands with different ratio of the used metals and materials, just like the thermal pastes

 

[Outback Bronze]

From what I have been reading about LM, it tends to dry up quicker than a good paste. Sure, temps will be fine for a while but over time as the LM gets absorbs by the HSF temps will start to degrade.

I have seen this statement many times from CPU delidders over at OCN that use LM and state it needs re-applying after several months.

I have also seen it happen on a 1080Ti FE that I have personally done. I opened it up and had to change it to a decent paste as the LM seemed to have dried up.

 

[Mysteoa]

And what about VRAM? I saw temps reaching around 90+ C degrees on some tests.

LM is just on the core. It is still using Paste on the VRAM, so no change other than difference from disassembly.

 

[Lycanwolfen]

What thermal paste was used I wonder. I am using thermal grizzly extreme on few of my card it lowered the temps lower than liquid metal.

 

[Knight47]

LM is just on the core. It is still using Paste on the VRAM, so no change other than difference from disassembly.

That looks like thermal putty to me.

 

[blufeb95]

From what I have been reading about LM, it tends to dry up quicker than a good paste. Sure, temps will be fine for a while but over time as the LM gets absorbs by the HSF temps will start to degrade.

I have seen this statement many times from CPU delidders over at OCN that use LM and state it needs re-applying after several months.

I have also seen it happen on a 1080Ti FE that I have personally done. I opened it up and had to change it to a decent paste as the LM seemed to have dried up.

Apparently the nickle plated base of the Nvidia cooler should prevent diffusion into the copper and the hermetic sealing dam Nvidia is using should prevent oxidation to the point that the application should last the reasonable service life of the card.

 

[Lycanwolfen]

Also what gets me is for a 2000.00 card why on earth are they using a aluminum heat sink. I mean why copper would have been much better.

 

[Hecate91]

PTM is still paste, and will eventually degrade.

PTM would take years to degrade, it doesn't separate out like paste does. Most thermal pastes degrade on high end GPU's after a few months.

 

[blufeb95]

Also what gets me is for a 2000.00 card why on earth are they using a aluminum heat sink. I mean why copper would have been much better.

Copper radiators are unecessarily heavy and make the card sag more and put unecessary stress on the PCIe slot and the motherboard for minimal gains. a good example of this is the old Thermalright Ultra 120 copper version really didn't perform much better than the aluminum fin version but it was much heavier and there were major concerns whether it could warp the motherboard.

 

[Outback Bronze]

the nickle plated base

A lot of delidders reuse the original IHS which is nickel plated copper and the LM will still need replacing.

Me personally from what I have seen and tested will probably never use LM again. Paste is also way more practical to apply and clean up.

 

[blufeb95]

A lot of delidders reuse the original IHS which is nickel plated copper and the LM will still need replacing.

Me personally from what I have seen and tested will probably never use LM again. Paste is also way more practical to apply and clean up.

The point where the die is soldered to the IHS is gold plated, I don't know how gold plated copper behaves vs nickle plated in terms of liquid metal diffusion. Also de-lidded CPUs aren't really hermetically sealed, and we don't know exactly what alloy Nvidia is using vs what's available aftermarket.

 

[Outback Bronze]

Also de-lidded CPUs aren't really hermetically sealed

This is true and also direct die would fall into that compartment so I'll give Nvidia engineers the benefit of the doubt atm but time will tell

 

[freeagent]

PTM would take years to degrade

Chatter says 6 months or so.

 

[Gasaraki]

Also what gets me is for a 2000.00 card why on earth are they using a aluminum heat sink. I mean why copper would have been much better.

Sure copper can extract and hold more heat but the main issue is copper loves holding on to that heat so fans have a harder time extracting that heat from the copper.

 

[csendesmark]

LOL
That risk just to have .5% better temperature

 

[blufeb95]

Sure copper can extract and hold more heat but the main issue is copper loves holding on to that heat so fans have a harder time extracting that heat from the copper.

Copper is better than aluminum at dissipating heat, the biggest problem is copper is 3 times as dense as aluminum and is impractically heavy on a big heatpipe radiator and while it cools better it's typically within a few C and not worth making the card weight twice to three times as much.

 

[RH92]

Maybe the huge area also plays a role - there is relatively small advantage in using LM compared to good paste between CPU heat spreader and cooler, but a large difference in using LM to interface small delidded CPU die to heat spreader.

The (NVIDIA GeForce RTX 5090) GB202 graphics processor is a large chip with a die area of 750 mm²

AMD 9800X3D die size: 70.6 mm²

Intel Core Ultra 9 285K die size: 243 mm²

Could explain some of it but i doubt it's the main factor . For instance even back in the 1080Ti ( 471 mm² ) days when people started experimenting with LM on GPUs you would see 15-20C diff between LM and paste. Basically the more power hungry the chip the more LM should scale ( provided the heatsink has the capacity to take the load ) but here we see almost no difference .

 

[Steevo]

PTM would take years to degrade, it doesn't separate out like paste does. Most thermal pastes degrade on high end GPU's after a few months.

Really? My 7970 would like a word, its still on its original thermal paste under a waterblock.


I think those who have issues with thermal paste that needs reapplied after a few months are doing something wrong or using the wrong product. I have build hundreds of computers over almost 30 years and never had an issue with machines in service even after 10 years, my own old machine was under the same cooling loop without changing the coolant, thermal paste for almost 8 years.

 

[remekra]

Really? My 7970 would like a word, its still on its original thermal paste under a waterblock.


I think those who have issues with thermal paste that needs reapplied after a few months are doing something wrong or using the wrong product. I have build hundreds of computers over almost 30 years and never had an issue with machines in service even after 10 years, my own old machine was under the same cooling loop without changing the coolant, thermal paste for almost 8 years.

My 7900XTX would like to have a word as well, reaching 110C after 6-7 months. Not applied by me even, from factory from Gigabyte. You can also search for threads here or reddit or elsewhere, many people with the same problem. But it happened couple of months after reviews so not much coverage on the issue.
Red Devils even from Powercolor had the same issue. Paste leaks out and hotspot rises to 110.

 

[Waldorf]

@Outback Bronze
ignoring trouble with application etc, as long as you do it a few times within a year, its fine after 3rd application.
its mainly a problem with new HS that never had it, and i remember a few brands/reviews recommending to apply/remove/reapply it ~3 times so it can saturate.
one the first cpu i used it i had the same problem, on the next one i did as i mentioned and it was still working without trouble almost 2y (sold cpu), and even block removal was easier (vs 1st time).

 

[Steevo]

My 7900XTX would like to have a word as well, reaching 110C after 6-7 months. Not applied by me even, from factory from Gigabyte. You can also search for threads here or reddit or elsewhere, many people with the same problem. But it happened couple of months after reviews so not much coverage on the issue.
Red Devils even from Powercolor had the same issue. Paste leaks out and hotspot rises to 110.

Mines been running F@H for a month now after gaming and doing other things and my hotspot is lower now than before. Maybe some MFGs are using garbage.

I used Céramique on my GPU since it was bare die when I installed my waterblock and the cards before that too, and I tried it on the CPU and MX4 worked better.

If they sell a million 7900XTX cards and 1000 users have an issue with improperly applied or defective thermal paste I wouldn't be surprised.

 

[remekra]

Mines been running F@H for a month now after gaming and doing other things and my hotspot is lower now than before. Maybe some MFGs are using garbage.

I used Céramique on my GPU since it was bare die when I installed my waterblock and the cards before that too, and I tried it on the CPU and MX4 worked better.

If they sell a million 7900XTX cards and 1000 users have an issue with improperly applied or defective thermal paste I wouldn't be surprised.

Even in the RDNA3 thread on this forum there were people that had the hotspot temp rising to 110, then instead of RMAing it doing the repaste themselves with good quality paste only to have it rising again soon.
Don't know if it's because it was designed with MCDs, or because of the fact that you can run it at 460W with some cards.
PTM simply is better than a thermal paste so if I'm spending 1000$ or more on a GPU I would expect the card to use it, instead of thermal paste.
We never had GPUs with such high wattages, dual cards like 690 had 300W and that was spread to dual chips. So for new cards with those kinds of TDPs its better to use PTM.

 

[mkppo]

The issue isn't transferring the heat from 750mm2 of surface area to the cooler, but rather the cooler itself close to being maxed out when trying to dissipate 600W.

LM makes a bigger difference in cases where the former is an issue eg. on Ryzens where the die is too small to transfer heat to the cooler which itself has headroom left for dissipating more watts.

I'm guessing PTM will perform close enough to LM in this case considering how a paste performs.

 

[chrcoluk]

I expect the way it was decided is the new cooler, has a bunch of incremental improvements that might be 2-3C here and there, so alone nothing to shout about, but in tandem, worthy.

 

[R-T-B]

The conclusion:
Nvidia is using the cheapest $hitty available liquid metal

Which still beats pretty much any and all thermal paste. Liquid metals cost as an OEM is more in application at scale costs than material cost. They are all basically Galinstan at heart anyways.

PTM is still paste, and will eventually degrade.

Yeah but it takes like 10 years, really a nonissue.

Chatter says 6 months or so.

Honeywell themselves and my experience contradict whatever "chatter" you have seen.

Also keep in mind they is A LOT of counterfeit PTM. Helios even, well regarded generally, is a clone and not real honeywell PTM. The clones vary from excellent to very bad.