Intel denies RMA due to the use of Liquid Metal as a TIM

[chrcoluk]

Dont see an issue with it being denied, you go down this route, but is consequences to it.

 

[DirtyDingusMcgee]

Any extreme form of modding has always been a warranty killer. I just always accepted that as a risk we take.

 

[de.das.dude]

Its not the fact that the RMA was denied which is surprising, but the more surprising fact is that so many intel CPUs are getting RMA'd in the first place!

 

[dont whant to set it"']

Also, unless the IHS did not have scratches or some deep grooves before LM(to work its way with more surface area)> Intel is now penny pinching on the IHS coating?

 

[claylomax]

So one of the Strix laptops premium features is the fact that they use LM on the CPU and GPU. Does that mean if you have a Strix laptop with an Intel chip that they will not honour the Warranty if the CPU fails?

With or without Liquid metal you don't get to RMA, it's Asus

 

[Lewzke]

but why the metal paste hassle? we have proven and safe industrial alternatives like the PTM7950, there is not much gain going from this to a metal layer, no wonder why Nvidia used this for 4090 or Lenovo for gaming laptops

 

[Bill_Bright]

Liquid metal as a TIM only reaches it's full potential on a bare die, with all of the other options we nowadays why use it on an IHS in the first place?

I don't understand why people go crazy over TIM choice.

Bragging rights. Just as some are totally obsessed with getting 1 more MHz out of their CPU, or 1 more FPS out of their GPU, some people are totally obsessed with achieving 1 more degree of additional cooling. If they could submerge their computers in liquid nitrogen so they could brag to their friends, they would. I am not saying that obsession is unhealthy - but I would not call it healthy!

That said, to Chrispy's "generalized" comment (not specific about LM) about TIM choice, I totally agree. Of MUCH MUCH MORE importance than which TIM formula is being used is just the absolute necessity to (1) use TIM AND (2), apply it properly. By properly, I mean clean surfaces and use only the minimum amount necessary for complete coverage.

The facts are, typically the difference in cooling efficiency between the most efficient and the least efficient TIM (assuming proper application) is just a small handful of degrees. And the reality is, if your device "needs" that small handful of degrees to prevent crossing thermal protection thresholds, then its temperature is already too high in the first place, and you have bigger, more urgent cooling issues to deal with first, like case cooling.

Some might want to jump in to say if you "lap" the mating surfaces, you don't need TIM. I disagree. You likely can get away without TIM "IF" the lapping was done correctly. But there will still be microscopic pits and valleys (just not as many or as deep) that could trap insulating air between the mating surfaces - not good. Plus, lapping definitely will obliterate any markings on the CPU and void any warranty.

 

[eidairaman1]

End user was overclocking it using that crap and it failed resulting in denied warranty, hope the idiot learned.

 

[freeagent]

Erryone wants to be hardcore, till its time to pay the hardcore price..

Surprised we are on page 2..

 

[Bill_Bright]

Erryone wants to be hardcore, till its time to pay the hardcore price..

Right. And then they blame someone else for their mistake.

Surprised we are on page 2..

I am surprised someone hasn't come in to blame Microsoft for this.

 

[Calenhad]

Have you ever tried an Asus warranty? It was one of my first posts (rant) on TPU.

I have. But I am lucky enough to be in a part of the world where it actually works as intended.

And either way it is not possible to yank out the cpu in a modern laptop and send it to Intel for RMA, instead of the manufacturer. Which is basically what the post I replied to in the first place said.

 

[Dr. Dro]

Right. And then they blame someone else for their mistake.

I am surprised someone hasn't come in to blame Microsoft for this.

What do you mean, Bill? Microsoft and Intel, as members of SPECTRE, have to keep the Wintel racket going. Nobody's talking about it because everyone already assumes it's happening!

Betcha Bill Gates and Michael Dell are both personally in to help Pat Gelsinger on this one.

 

[:D:D]

Any extreme form of modding has always been a warranty killer. I just always accepted that as a risk we take.

Me too.

Erryone wants to be hardcore, till its time to pay the hardcore price..

Not everyone.

 

[freeagent]

Not everyone.

Using LM is petty hardcore.

Just like flashing your GPU bios to another vendor for X reason.

Or using LN2 for a few hours of playtime on your CPU.. its all relative.

 

[RootinTootinPootin]

I direct die so I use LM on my 2 chips, if it kicks the can, relid it nice and tidy and like nothing ever happened.

 

[freeagent]

S462 was all about direct die lol..

You used to get a few good crunches off the core before it died.. I doubt you would be so lucky these days

 

[RootinTootinPootin]

S462 was all about direct die lol..

but the current TIMs were non-existent then, I only use arctic silver before, still that's too dangerous too..haha

 

[freeagent]

but the current TIMs were non-existent then, I only use arctic silver before, still that's too dangerous too..haha

Heck no, AS5 is good stuff. Probably not the best anymore, but it can still handle a couple of hundred watts..

Remember.. it isn't conductive

 

[Bill_Bright]

What do you mean, Bill? Microsoft and Intel, as members of SPECTRE

Nonsense. Get real. Why degrade the thread with falsehoods? Everyone knows they were founding members of KAOS.

 

[R0H1T]

I am surprised someone hasn't come in to blame Microsoft for this.

What do you mean? Did you forget the user had to do LM because MS' EULA asked forced him to

Next up EU finds MS liable for Intel's performance "degrading" patches

 

[Count von Schwalbe]

Nonsense. Get real. Why degrade the thread with falsehoods? Everyone knows they were founding members of KAOS.

That's why Intel released a patch to mitigate the impact of SPECTRE.

Now back on topic, I think the issue is not corrosion but amalgamation. I know that is a bit of a nitpick, but there is a difference and it can be an important distinction. Amalgamation will continue to degrade the IHS even after the LM is removed, as the gallium diffuses through the copper. An aluminum IHS might actually be fine with it, if the oxide coating is unbroken.

 

[JustBenching]

Intel wouldn't handle that RMA anyway......?

It doesn't matter if you have an Intel or AMD desktop CPU, if your IHS is severely damaged or defaced to the point of removing all markings on it, both RMAs would be refused. They can't identify any, y'know, identifying markings that they require in the RMA process. LM doesn't do anything of that sort to the bare die, and the bare die doesn't have any identifying info on it.

Isn't the serial number also printed on the actual chip though? I swear I must have it seen somewhere outside the IHS.

I've been saying all along LM is not a TIM you really want to use because of the damaging effects it has on stuff over time.
Already proven as fact this stuff WILL eventually render things like coolers and stuff in a state that makes it not work as well, if at all.
Nickel plating for example, while it does slow down it's destructive effects it still fails to stop it period from doing it.

I can see in a couple of years folks that bought PS5 consoles suddenly running into problems caused by the use of LM in those machines, in fact I woudn't doubt it's already started happening by now.
If there are different formulated kinds of LM, I'm not aware of them but did see this and noted how this is related to "Wetting", or how well it embeds itself into other materials.
Liquid metal - Wikipedia

That's not true though. You can definitely clean the lm, even from ihs. You need to be careful, but you can bring the printing back.

I've been using it for 3 years now on my 12900k + u12a combo. No issues.

 

[Bones]

Nonsense. Get real. Why degrade the thread with falsehoods? Everyone knows they were founding members of KAOS.

This founding member even polishes his suit with LM since it's so "In" style these days.



Now, getting to this:

Using LM is petty hardcore.

Just like flashing your GPU bios to another vendor for X reason.

Or using LN2 for a few hours of playtime on your CPU.. its all relative.

I agree and must also agree to what you said is relative about it.

Funny how some apply it and then think all their cooling problems are solved..... Only to watch things go to hell over time, OR they just clock it up right from the start and wonder why their temps are still "Too damned high" like rent in NYC, leading to the inevitable.

If you're gonna clock things up to the moon, get what does the job in the first place - LM ain't it guys and certainly isn't suitable for sub-zero temps either if going that way about it.

As for the current trend for some companies to use LM in their products, as long as it's cheap(er) than standard TIM and works until the warranty expires they'll do it and keep on doing it too.....
Looking at you and your PS5's over that one Sony.

The destructive effects of LM are just that.
You can argue all you want but it happens, guys like the one that used it and apparently abused it too are one reason why things like being denied, such as a case of the etching on the chip's lid is no longer there to say what the chip "Is".

LM DOES work as a TIM but in the face of what it is, what it does, how well it does the job, the cost of it and all the problems you can name about it to the contrary is why I'll never touch the stuff.
If I ever get something that has it applied, that crap is being removed on the spot - Warranty be damned over it.

 

[eidairaman1]

That's why Intel released a patch to mitigate the impact of SPECTRE.

Now back on topic, I think the issue is not corrosion but amalgamation. I know that is a bit of a nitpick, but there is a difference and it can be an important distinction. Amalgamation will continue to degrade the IHS even after the LM is removed, as the gallium diffuses through the copper. An aluminum IHS might actually be fine with it, if the oxide coating is unbroken.

Dissimilar metals touching then add an electrolyte= galvanic corrosion→ pitting→ intergranular→ exfoliation, other types are fretting and filiform.

Pitting leads to component failure when vibrated (work hardened enough) 1 reason why you won't see sharp corners/edges on aircraft (have radius for a reason)

 

[Count von Schwalbe]

Dissimilar metals touching then add an electrolyte= galvanic corrosion→ pitting→ intergranular→ exfoliation, other types are fretting and filiform.

Pitting leads to component failure when vibrated (work hardened enough) 1 reason why you won't see sharp corners/edges on aircraft (have radius for a reason)

True, but there is no electrolyte at play here. That is a real concern for mixing metals in a water loop.