News Intel Arrow Lake and Lunar Lake CPUs are unaffected by crashing issues — Vmin Shift Instability issue only impacts 13th and 14th Gen CPUs

[Admin]

Intel releases an official update on its 13th and 14th Generation Core CPU instability issues and claims that future Arrow Lake and Lunar Lake CPUs are unaffected.

Intel Arrow Lake and Lunar Lake CPUs are unaffected by crashing issues — Vmin Shift Instability issue only impacts 13th and 14th Gen CPUs : Read more

 

[Mattzun]

The real question is why should we trust that there won't be other issues should have been caught and fixed before release?

If Intel had just been better at accepting RMAs and had extended its warranty earlier, I could be confident that I'd be covered if there was an issue. Intel's behavior and financial issues makes me way less likely to consider Intel for my upgrades next year.

 

[MacZ24]

The semiconductor designs are on a complexifying treadmill, and also, perhaps, the old guard has retired/left and some expertise was lost ?

That such a problem could happen at Intel is not reassuring at all, but complexity has drawbacks. And for me noone in the semiconductor industry is immune to this.

 

[Gururu]

Intel certainly lost a few fans with this one. I had to RMA some Corsair memory yesterday, never buying them again.

 

[ThisIsMe]

The real question is why should we trust that there won't be other issues should have been caught and fixed before release?

If Intel had just been better at accepting RMAs and had extended its warranty earlier, I could be confident that I'd be covered if there was an issue. Intel's behavior and financial issues makes me way less likely to consider Intel for my upgrades next year.

It’s always the most fickle of people that like to boast of their choices. Sadly they’re merely trying to justify their decisions to themselves by convincing others of why they did what they did. All your reasoning is based on rumors and hearsay beginning with a handful of stories of RMA issues perpetuated by trolls and modern tech news outlets with juicy leaks FOMO.

 

[watzupken]

No company will deliver products or services that’s flawless consistently. The problem is how they choose to handle the situation at that point. The fact that Intel knew of the issue, sat and deny it for more than 1.5 years, and finally admitting because the noise is growing out of proportion. This is bad handling of a severe issue. For them to now call out that the crashing issue is only limited to 13 and 14 gen Intel processor is a clear attempt to isolate the problem and minimise impact on their processor sale. But this is an insensitive statement because it’s clear a lot of their customers are still experiencing system crashes using the affected CPUs and Intel is not recalling but relying on BIOS update or user initiated RMA process that somehow may get rejected.

 

[watzupken]

It’s always the most fickle of people that like to boast of their choices. Sadly they’re merely trying to justify their decisions to themselves by convincing others of why they did what they did. All your reasoning is based on rumors and hearsay beginning with a handful of stories of RMA issues perpetuated by trolls and modern tech news outlets with juicy leaks FOMO.

Were those accusations against Intel rumors and hearsay? If so, why is Intel acknowledging that it’s an issue now? Like in their statement, the spokesperson clearly acknowledge there are issues with 13th and 14th gen Intel CPUs, that started with the rumors and hearsay. Unhappy customer tends to make the most noise. If for example you paid 500 bucks for a high end CPU that’s great at benchmarks, but crashes when running games, I don’t think you will be a happy customer to be honest. Fortunately for me, I stuck on to my my Alder Lake CPU as I was not convinced spamming E-cores is meaningful to my workflow.

 

[baboma]

>Were those accusations against Intel rumors and hearsay?

It's not a binary "good/bad" CPU issue. Not all RPL CPUs are bad, but some. The rumormongering is about how much is "some." One bloke gets on reddit and says MY CPU BLOWS UP! and it gets magnified manifold.

13th-gen CPUs have been around for two years. 14th-gen is almost a year. Both have a track record. You can peruse on public forums like reddit to see how many "bad CPU" are reported.

https://google.com/search?&q=raptor+lake+instability+report+reddit

From cursory manual scan, there's a lot of teeth-gnashing and anxiety, and rightfully so, but precious few actual reports of "bad CPUs." The rational conclusion to draw is that "some" equates to a small percentage.

My gauge is that 13900K/14900K are most at risk, within that small percentage. 13700K/14700K are substantially less so, and 13600K/14600K's risk is even smaller still.

Reddit is a valid metric, because its population are by and large regular consumers, which mirrors people here. If you were a server shop with intensive 24/7 use, then you'd use a more rigorous test method.

Second, there's a distinction between people with existing RPLs, and people thinking of buying RPLs. Assuming that the fix is conclusive--and there is no reason to think that Intel is lying--then people buying RPLs now, and immediately applying the microcode fix, would have zero risk.


>The real question is why should we trust that there won't be other issues should have been caught and fixed before release?

Both AMD and Intel have a long track record of making CPUs. Their track records show that all CPUs can have issue post-release--to wit, AMD's Sinkclose vulnerability--and the vast majority of those issues can be fixed or mitigated with a microcode patch. The Intel RPL degradation--more correctly, accelerated degradation, as all CPUs degrade over time--is an outlier. The possibility of something like it to happen again is infinitesimally small.

Now, you can say, well, no matter how small the possibility is, I just don't trust Intel anymore, and I won't buy any of its CPUs from now till forever. This is known as cutting off your nose to spite your case, because your future CPU options are then cut by half. People who think like this are fanboys and idiots.

 

[Mattzun]

Intel has spent over a year denying that there were issues and decided to blame motherboard vendors instead of microcode when they first acknowledged the issue.

They have rejected RMAs that should have been accepted and didn't analyze the failures when they did accept them.

If they had done an analysis of the earliest failures and determined that they had a voltage issue, this wouldn't have been a problem.

Its not that Intel had an unusual failure. Its that Intel intentionally pushed the envelope and treated its customers and partners like shit when that caused problems.

 

[baboma]

>Intel has spent over a year denying that there were issues...

Incompetence doesn't equate to malfeasance. Intel is in turmoil right now. I'm not surprised that it isn't the most responsive to customer concerns, or product issues.

Paying for a CPU is different than paying for a service. A CPU is likely the most reliable component out of all PC components, the present Intel fiasco notwithstanding. You don't need to have a "responsive" company for after-sale support. You don't need after-sale support, period. The only thing you need is warranty service if the part is bad. No matter Intel's dithering, the fact is that you can now RMA the part for 5 years, from 3 years. If anything, you should feel more assured about the buy.

 

[JRStern]

Intel Arrow Lake and Lunar Lake CPUs are unaffected by crashing issues

... so far!

 

[thestryker]

Given the nature of the problem facing Raptor Lake it makes sense that it wouldn't affect future products. I get why they're stating it's not a problem, but people who don't believe them aren't going to believe this and those who do likely already realized it.

Intel has spent over a year denying that there were issues and decided to blame motherboard vendors instead of microcode when they first acknowledged the issue.

This should tell you how low the failure rates were more than anything else. Intel also had to actually step through to find the issues going on. It makes sense that motherboard manufacturers ignoring power settings might have had something to do with it. If this changes default behavior across the industry going forward I'd call this initial focus a net positive for consumers.

They have rejected RMAs that should have been accepted...

This is undoubtedly true, but they also very likely accepted many as well.

... and didn't analyze the failures when they did accept them.

We don't know what their process is, but I'm assuming for most of the time this issue had been going on the failure rate was low enough they used their standard RMA analysis policy.

If they had done an analysis of the earliest failures and determined that they had a voltage issue, this wouldn't have been a problem.

This is unlikely given how long it has taken them to narrow down the first fix. It may have been resolved faster, but it has taken them months to figure out even that much.

 

[rluker5]

I like how Buildzoid has been analyzing the excessive overvoltage leading to premature degradation and related by being motherboard setting undervolting by introducing excessive vdroop instability issue.

He's been measuring using external monitoring equipment and seeing what his motherboards are doing with the different microcodes and settings.

A much more scientific approach than the qualitative speculations and fud spreading so common elsewhere. I wonder what happened to make technical analysis in a seemingly technical community so rare?

High voltage may lead to degradation and low voltage may lead to instability seemed like things that were so well known that one didn't have to state them. Now it seems like many refuse to believe these in hopes of something else.

 

[TheHerald]

It’s always the most fickle of people that like to boast of their choices. Sadly they’re merely trying to justify their decisions to themselves by convincing others of why they did what they did. All your reasoning is based on rumors and hearsay beginning with a handful of stories of RMA issues perpetuated by trolls and modern tech news outlets with juicy leaks FOMO.

People don't understand that there are hundreds or even thousands of RMAs done daily, but the 2 stories on reddit (which btw, nothing really happened anyways and the rmas went through) are all that's needed to call intel bad at RMA.

Either that, or it's the usual amd bias everyone has.

 

[TheHerald]

They have rejected RMAs that should have been accepted and didn't analyze the failures when they did accept them.

What do you mean by that? Can you be more precise? Is there any other company that has rejected RMAs that shouldn't have? AMD for example. Or is it just an intel thing?

Just so you know, intel is known for how fast their RMA is for more than 10 years. There is the likelihood the current situation put more pressure on their support logistics - that's understandable, but saying they flat out deny rmas, come on now. There are hundreds of complaints about RMA process for every single company.

 

[YSCCC]

Intel has confirmed that thanks to their new architecture, the company's next-generation Core Ultra 200 Lunar Lake and Arrow Lake CPUs aren't affected by the Vmin Shift Instability issue.

So, It basically means they admit that it IS architecture issue that causes the degradation

 

[Ogotai]

yea, too bad its not a new architecture

 

[Hotrod2go]

The semiconductor designs are on a complexifying treadmill, and also, perhaps, the old guard has retired/left and some expertise was lost ?

That such a problem could happen at Intel is not reassuring at all, but complexity has drawbacks. And for me noone in the semiconductor industry is immune to this.

Yes, reminds me of an Einstein saying from long ago;
"Everything should be made as simple as possible, but not simpler."
Hence why AMD's non hybrid architecture for AM5 SKUs is a winner today.