Intel Will Not Recall Failing 13th and 14th Gen CPUs

[InVasMani]

Intel will not recall failings...

Silicon Dreams and CPU Screams

"Don’t tell me 'cause it hertz,
My CPU’s got quirks,
Running hot, it’s a fight,
In the middle of the night."

 

[thesmokingman]

This is the price people pay when they want latest and greatest,technically you pay for the privilege to be a test pilot...my moto always buy a generation behind the latest one no matter the product.

What you need to do is stop justifying this shit.

 

[oxrufiioxo]

What you need to do is stop justifying this shit.

I'm kinda with him on that at this point at least wait 6-12 months, Both AMD and Nvidia had gpu issues this past generation regardless of how big or small, and both intel and amd had cpu issues.....

This is the worst of the 4 by far but that doesn't change the fact that stuff seems like it is being rushed to market these days.

I really just want to see how intel handles RMA's at this point before I ever buy a CPU from them again.

 

[phanbuey]

I don't care much, my own 12th gen CPU runs fine, no issues for me.
I'm still waiting for Arrow Lake, I'm sure they don't want to happen this again with their next gen CPU's.

I mean i've had no stability issues with a day 1 13700KF and MSI mobo -- easily one of the most stable platforms / nicest platforms i've had so far.

Really curious to see what the actual numbers of the issue are after the OMFG ZOMG 100% FAIL RATE techtuber clickbaits die down. I doubt these issues are as widespread as they are made out to be.

 

[kondamin]

The 13 line is a generation behind.

in name, 14gen is 13gen

 

[Chaitanya]

Don't worry about that, they just changed the CEO in charge of their foundry manufacturing from a big nosed person to an indian person. I'm expecting them to get stuck on a node again pretty soon, like how they couldn't jump away from 14nm for like six generations.

That person with big nose has leeched millions from tax payers and now is taking his business across the ocean as Made in 'murica chips are destroying themselves.

Intel did recall both cpus and chipsets in past so this is puzzling move.

On a serious note Gelsinger has been a disaster for Intel and this new saga of failing chips just after getting subsidies should call for an investigation by US govt.

 

[Caring1]

Arc is unaffected!

As far as we know so far

 

[InVasMani]

Arc could be dead on arrival with the hill Intel is choosing to die on.

 

[remixedcat]

Hey I'm close to the Columbus fab... anyone wanna crowdfund me a visit, rent a car and a hotel stay there n I can see about talking to some intel ppl there n see if they'll give me a QA job lol...

I'm 3 hours away lol

 

[InVasMani]

A lot is riding on how well the microcode addresses the issues and/or how much damage is already done to hardware and how Intel handles it all.

 

[R0H1T]

Don't worry about that, they just changed the CEO in charge of their foundry manufacturing from a big nosed person to an indian person. I'm expecting them to get stuck on a node again pretty soon, like how they couldn't jump away from 14nm for like six generations.

So then all the OEM BS during the 2004-06 era or contra revenues bribes a decade back & this sh!t is on who exactly? The American in charge right, not the Indian (American) leading the foundry

Really subtle that, good job

 

[JustBenching]

Disastrous handling of the situation... Intel can and should do better. They don't have an universal, or hell, a lead at all over the competition. Best not to get complacent.

It would be impossible to do a recall 2 years after release. How many units are we actually talking about? Hundreds of thousands? Millions maybe?

But seems like the issue is the single rail that feeds pcores / ecores / cache. Remember a couple of weeks ago I said that my 12900k gets immediate degradation on the cache at or above 1.4v? Well, since over 1.4v is kinda common for 13 and 14th gen, it looks like it's the cache that goes kaput.

Probably the microcode update will cap the cache at 4ghz instead of the stock 4.5 and that's it

Nevermind, they said their microcode won't fix the issue. So they will just make sure new chips don't degrade and RMA the problematic ones.

 

[R0H1T]

And how are measuring it's just degradation of the cache? Is that even possible without "degrading" other parts of the core? You're just speculating much like the rest of us

 

[JustBenching]

And how are measuring it's just degradation of the cache? Is that even possible without "degrading" other parts of the core? You're just speculating much like the rest of us

Testing on my alderlake it's exactly what happened, P and ecores are still fine, but I have to downclock the cache to work at my previous stable settings. This seems to align with the rumors / leaks MLID (im not posting links cause i don't want to give clicks) is posting on his latest video.

EG1. Also crashes are usually connected to the memory / cache subsystem. A degraded core will usually result in clock_watchdog BSOD.

 

[Dr. Dro]

It would be impossible to do a recall 2 years after release. How many units are we actually talking about? Hundreds of thousands? Millions maybe?

But seems like the issue is the single rail that feeds pcores / ecores / cache. Remember a couple of weeks ago I said that my 12900k gets immediate degradation on the cache at or above 1.4v? Well, since over 1.4v is kinda common for 13 and 14th gen, it looks like it's the cache that goes kaput.

Probably the microcode update will cap the cache at 4ghz instead of the stock 4.5 and that's it

Reducing spec after the product launched is not an acceptable solution, IMO. Especially for SKUs targeted at the enthusiast segment. But I don't think it'll come down to that. And well, may be millions - but Intel is still responsible for replacing every damaged chip.

 

[john_]

That would be true in the quad core era, but there are excellent competitor chips now.

It doesn't really matter. People have preferences. People who are used to Intel platforms will remain to Intel platforms. The same for people who are used to AMD. Logic suggests that Intel will not mess for a second time in a row so people will rush to get the next Intel model.

I don't expect Intel to suffer much from this, even when seemingly mishandling this case so much by not accepting RMAs. And they know it, that's why they aren't going to do RMAs or recalling the whole bunch of CPUs sold. A recall will remain in history and get much more press coverage, RMAs will cost them greatly and probably drag the problem for years because CPUs will probably keep failing. Saying that "we throw out a microcode and we are done", makes it look like a typical case and nothing more. They react as politicians where being honest and admitting something is sometimes much worst than pretending publicly that it doesn't exist.

 

[JustBenching]

Reducing spec after the product launched is not an acceptable solution, IMO. Especially for SKUs targeted at the enthusiast segment. But I don't think it'll come down to that. And well, may be millions - but Intel is still responsible for replacing every damaged chip.

Yeah I was wrong, just read their statement more closely, microcode won't change clocks since it won't fix the problematic cpus. They will just RMA those on demand.

I don't expect Intel to suffer much from this, even when seemingly mishandling this case so much by not accepting RMAs. And they know it, that's why they aren't going to do RMAs

???

Who said they are not doing RMAs? The article literally says that they are in fact replacing the problematic products.

 

[R0H1T]

Si level degradation will rarely be limited to just that part of the chip. Pretty sure it will spread, now whether you can measure that accurately or not is a different topic.

Reducing spec after the product launched is not an acceptable solution, IMO.

Huh? Pretty sure that's a legal case in 99% of the countries out there!

 

[usiname]

Beside the already damaged CPUs, the micro code update most likely will reduce the performance of every existing Raptor Lake well bellow what was advertised in the reviews

 

[remixedcat]

Beside the already damaged CPUs, the micro code update most likely will reduce the performance of every existing Raptor Lake well bellow what was advertised in the reviews

Imagine paying for an i7 or i9 and getting an i5...

 

[stimpy88]

Wow Intel, the balls on you!!! With a near 50% failure rate, you dig in... Such inspire, much BS.

 

[Gatt]

Somewhere I've read that even some non-K CPU are affected. Is that true?

 

[R0H1T]

Possibly & quite likely, K chips are the same just OCed & OVed siblings.

 

[john_]

Who said they are not doing RMAs? The article literally says that they are in fact replacing the problematic products.

Not for everyone and I am expecting their policy after the new firmware is out to change to
"User Error, because the user didn't applied the new firmware on time to prevent degradation".

 

[Gatt]

Thank you. I stress tested my i9-14900 with and without any limit. So far so good. But, who knows ...