Intel processors crashing Unreal engine games (and others)

[IEC]

MLID (yeah, I know) presses X to doubt Intel:

My take:
After months of "nothing to see here, move along" Intel finally makes two big admissions:
1) The problem is real (I still think they are downplaying the scope)
2) Damage is permanent

To their credit, they are saying users can reach out to Intel about affected chips. How the end user diagnoses and proves this for RMA approval is up in the air. Hopefully they don't deny RMAs like they have been, either.

Intel also conveniently punts a fix that supposedly has no performance impact to after the competitor launches their next gen. While quietly launching a a i9-14901KE with only P-cores that peaks with TVB at 5.8GHz.

Wait. 5.8GHz for max boost and max TVB? Almost like they're hitting a hard wall at that frequency... hmm, what could possibly be the reason for that?

 

[GTracing]

I don't know why he wouldn't just tell people to download the intel extreme tuning utility.

Tinfoil hat time, maybe Intel doesn't want to point to a good testing tool because it would cause more degradation and more RMAs. This GPU driver test method might be designed to dissuade people from doing anything to cause the CPU to degrade more or crash.

 

[Rigg]

What they did not factor in, is GN, HUB, Kit Guru, and maybe even LTT, are going to make certain their audience knows there is a HUGE asterisk attached to Raptor Lake.

As far as I'm concerned the upcoming Zen 5 reviews will be a litmus test for review outlets. If the Raptor Lake concerns are not clearly explained, and do not weigh heavily on any purchase recommendations contained within the review, then that reviewer should lose all credibility. You simply can not include any comparisons to Raptor Lake in a review right now without addressing the elephant in the room. I actually hope that the more prominent You Tubers are talking to each other behind the scenes about how to approach review material including Raptor Lake results moving forward.

 

[Sgraffite]

Ya, it feels like they are slow playing this as much as possibly they can for multiple reasons.

 

[jdubs03]

As far as I'm concerned the upcoming Zen 5 reviews will be a litmus test for review outlets. If the Raptor Lake concerns are not clearly explained, and do not weigh heavily on any purchase recommendations contained within the review, then that reviewer should lose all credibility. You simply can not include any comparisons to Raptor Lake in a review right now without addressing the elephant in the room. I actually hope that the more prominent You Tubers are talking to each other behind the scenes about how to approach review material including Raptor Lake results moving forward.

Seems like it’s pretty much unavoidable. Particularly since the main audience of the review sites and YouTubers already know about this situation.

 

[Rigg]

Seems like it’s pretty much unavoidable. Particularly since the main audience of the review sites and YouTubers already know about this situation.

Outlets like GN, HUB, Kit Guru, and L1 techs are pretty much guaranteed to make the Raptor issue a focus in Zen 5 reviews. Hopefully LTT will show some sense of responsibility to their massive viewership and properly address it. I have less faith in the review sites and the less prominent YouTubers to do the right thing.

 

[H433x0n]

MLID (yeah, I know) presses X to doubt Intel:

My take:
After months of "nothing to see here, move along" Intel finally makes two big admissions:
1) The problem is real (I still think they are downplaying the scope)
2) Damage is permanent

To their credit, they are saying users can reach out to Intel about affected chips. How the end user diagnoses and proves this for RMA approval is up in the air. Hopefully they don't deny RMAs like they have been, either.

Intel also conveniently punts a fix that supposedly has no performance impact to after the competitor launches their next gen. While quietly launching a a i9-14901KE with only P-cores that peaks with TVB at 5.8GHz.

Wait. 5.8GHz for max boost and max TVB? Almost like they're hitting a hard wall at that frequency... hmm, what could possibly be the reason for that?

I think people are reading too much into 14901KE. It’s an oddball SKU meant for the embedded market - nothing more, nothing less. It’s not being marketed or pushed by Intel for regular consumers.

 

[Joe NYC]

As far as I'm concerned the upcoming Zen 5 reviews will be a litmus test for review outlets. If the Raptor Lake concerns are not clearly explained, and do not weigh heavily on any purchase recommendations contained within the review, then that reviewer should lose all credibility. You simply can not include any comparisons to Raptor Lake in a review right now without addressing the elephant in the room. I actually hope that the more prominent You Tubers are talking to each other behind the scenes about how to approach review material including Raptor Lake results moving forward.

There is already GN coming out strongly against buying any Intel CPUs while this issue is unresolved.

Anandtech, in their July CPU recommendation article did not recommend any Intel CPUs. IIRC, this article on Anandtech came out even before the Gamers Nexus video.

So it seems to be spreading. We will see how other review sites conduct their Zen 5 reviews, if they consider Raptor Lake to be a valid competitor, even though competitive results of Raptor Lake may induce instability or whether they recommend staying away from Raptor Lake "until we know what's going on"

 

[gorobei]

Outlets like GN, HUB, Kit Guru, and L1 techs are pretty much guaranteed to make the Raptor issue a focus in Zen 5 reviews. Hopefully LTT will show some responsibility to their massive viewership and properly address it. I have less faith in the review sites and the less prominent YouTubers to do the right thing.

LTT have talked about it a few times on the wan show podcast. they havent talked about how they will handle the benchmarks for zen5 comparison, but Linus has been as harsh about the lack of communication and response from intel as the rest of the techtubers like L1T GN PCworld HWUB. his main concern is how they make 13/14 customers whole, given there could a shortage of replacement chips if the problem is truly widespread. the mobile situation also worries him.

 

[DAPUNISHER]

Outlets like GN, HUB, Kit Guru, and L1 techs are pretty much guaranteed to make the Raptor issue a focus in Zen 5 reviews. Hopefully LTT will show some sense of responsibility to their massive viewership and properly address it. I have less faith in the review sites and the less prominent YouTubers to do the right thing.

Our sister site Tom's has been getting away with shilling for decades, so I have little faith in traditional reviews. However, any techtuber that shills for dollars is going to get ratioed hard, is my best guess. That robeytech douche that just did the fluff job is approaching a 50% downvote.

 

[jdubs03]

There is already GN coming out strongly against buying any Intel CPUs while this issue is unresolved.

Anandtech, in their July CPU recommendation article did not recommend any Intel CPUs. IIRC, this article on Anandtech came out even before the Gamers Nexus video.

So it seems to be spreading. We will see how other review sites conduct their Zen 5 reviews, if they consider Raptor Lake to be a valid competitor, even though competitive results of Raptor Lake may induce instability or whether they recommend staying away from Raptor Lake "until we know what's going on"

Yeah GN was very strong on that. They pretty much gave the roadmap on how to do it.
My assumption is that they will test 13th/14th gen as-is against Zen 5, but then disclaim that many of those Intel chips are unstable and are awaiting a microcode update. I feel like doing anything else just introduces too many variables, i.e. any performance reduction from microcode, lower voltage, etc.

 

[DAPUNISHER]

LTT have talked about it a few times on the wan show podcast. they havent talked about how they will handle the benchmarks for zen5 comparison, but Linus has been as harsh about the lack of communication and response from intel as the rest of the techtubers like L1T GN PCworld HWUB. his main concern is how they make 13/14 customers whole, given there could a shortage of replacement chips if the problem is truly widespread. the mobile situation also worries him.

Even if Intel took all this time to stockpile CPUs, are they solid? Or will it be what Wendell reported, with tray after tray having a bunch go bad too? Presuming the MC patch really does retain performance that is. Otherwise who cares? A stable but noticeably slower CPU is a hard fail in its own right.

 

[Rigg]

Our sister site Tom's has been getting away with shilling for decades, so I have little faith in traditional reviews. However, any techtuber that shills for dollars is going to get ratioed hard, is my best guess. That robeytech douche that just did the fluff job is approaching a 50% downvote.

Oh god don't get me started on Tom's. Paul Alcorn is a hack and a shill. I'm not one to throw that accusation around willy nilly either. I got a vacation from the forums for getting heated in a review comment thread over Alcorn giving a POS MSI X570 board an editors choice award. Decided to sign up here shortly after.

 

[Saylick]

Oh god don't get me started on Tom's. Paul Alcorn is a hack and a shill. I'm not one to throw that accusation around willy nilly either. I got a vacation from the forums for getting heated in a review comment thread over Alcorn giving a POS MSI X570 board an editors choice award. Decided to sign up here shortly after.

Lmao, anyone can buy a participation trophy from a “tech” website these days. All you gotta do is give them a review sample and voila! It gets an Editor’s Choice Award, I guess because it’s easy for an editor to choose/recommend when they don’t have to pony up the money themselves.

 

[DAPUNISHER]

Lmao, anyone can buy a participation trophy from a “tech” website these days. All you gotta do is give them a review sample and voila! It gets an Editor’s Choice Award, I guess because it’s easy for an editor to choose/recommend when they don’t have to pony up the money themselves.

Not having any skin in the game is part of it. But you have to give Intel props. For decades their sales force has cultivated the buddy buddy dynamic with everyone they can. So that saying negative things about them makes you feel like you are betraying a friend. I am certain they act all wounded afterwards too. They should have travel agent added to their biz cards, because they will send you on a guilt trip.

Anyways I am ready for the show

 

[coercitiv]

At this moment I'm in full Nvidia bumpgate mode:

• partial dissemination of information using different sources, including confirmation of the oxidation issue while hand waving to make it seem not-news
• offering some random youtuber with 30k subscribers advanced access to information in order to prepare a video that should have been prepared by Intel themselves
• all kinds of steps to "help prevent damage, but no guarantees" presented as mandatory if you want to protect your CPU while "final" microcode update comes in Aug, but not assumed by Intel, voiced through a 3rd party instead.
• recommending users to install Nvidia drivers 10 times ?!?! after months of delays, this is the method they come up with

I don't think I've ever proposed or endorsed conspiracy theories on the forum, but all the behavioral signs from Intel are telling me they're in cover up mode, more exactly in delay and damage control mode. Minimize RMAs. The official press release was meh and I was ready to let it go, but the reddit post and the 3rd party video endorsed by Intel stink from a mile away.

 

[Saylick]

At this moment I'm in full Nvidia bumpgate mode:

• partial dissemination of information using different sources, including confirmation of the oxidation issue while hand waving to make it seem not-news
• offering some random youtuber with 30k subscribers advanced access to information in order to prepare a video that should have been prepared by Intel themselves
• all kinds of steps to "help prevent damage, but no guarantees" presented as mandatory if you want to protect your CPU, but not assumed by Intel, voiced through a 3rd party instead.
• recommending users to install Nvidia drivers 10 times ?!?! after months of delays, this is the method they come up with

I don't think I've ever proposed or endorsed conspiracy theories on the forum, but all the behavioral signs from Intel are telling me they're in cover up mode, more exactly in delay and damage control mode. Minimize RMAs. The official press release was meh and I was ready to let it go, but the reddit post and the 3rd party video endorsed by Intel stink from a mile away.

The way I see it, it reminds me of the Narcissist’s Prayer:
- Intel CPUs don’t have a problem.
- And if it did, it didn’t affect a lot of people.
- Even if it affected a lot of people, the solution doesn’t reduce performance.
- Even if it does reduce performance, it’s not my fault. It was the mobo maker’s fault.
- And if it was my fault, we didn’t mean it.
- And if we meant it, you the consumer deserved it because all you care about is who has the longest bar or biggest number in the benchmark.

I’m on Xitter and even I start to see the usual Intel defenders go down this Prayer, one step at a time. First it was denial, then it was minimization, etc etc. At some point, I wouldn’t be surprised to see them say RPL is in the past and we shouldn’t care anymore, so let’s just focus on ARL, ya? Please? Pretty please??

 

[coercitiv]

Since it's "not-news", I'll quote the Intel confirmation of oxidation issue one more time:

We can confirm that the via Oxidation manufacturing issue affected some early Intel Core 13th Gen desktop processors. However, the issue was root caused and addressed with manufacturing improvements and screens in 2023. We have also looked at it from the instability reports on Intel Core 13th Gen desktop processors and the analysis to-date has determined that only a small number of instability reports can be connected to the manufacturing issue.

Let's translate this using the sarcasm filter to upscale important details:
We had a production issue that led to accelerated CPU degradation. We did not disclose this to customers, we juts let them figure out by themselves, eventually. To this day, this degradation issue still produces errors in some consumer systems, but it's a small number so you shouldn't care.

 

[poke01]

it’s times like this I wish x86 was open to more companies.

Seriously what is up with the up down trend of these two companies. When Intel is good, AMD produces bad products and vice versa.

 

[gorobei]

Even if Intel took all this time to stockpile CPUs, are they solid? Or will it be what Wendell reported, with tray after tray having a bunch go bad too? Presuming the MC patch really does retain performance that is. Otherwise who cares? A stable but noticeably slower CPU is a hard fail in its own right.

part of his thinking is that there may never be a way to fix this in a way that doesnt screw up all of intel's future schedule.

if it isnt something a microcode can fix, they would have to

• assign engineers that originally worked on that gen who are likely already transferred to working on an upcoming gen. that would delay future chips.
• if the numbers are small they can try to offer replacements and eat the money loss, but if it is large scale then they would have to do a whole wafer run to make enough 'fixed' chips. unless intel runs their older fabs well past the next gen release to have spares for their corporate clients, this would seriously affect the profit margins.
• they cant even offer to upgrade them to arrowlake since they are on lga1700.

all of this makes it look like intel cant afford to do the right thing.

 

[jdubs03]

The way I see it, it reminds me of the Narcissist’s Prayer:
- Intel CPUs don’t have a problem.
- And if it did, it didn’t affect a lot of people.
- Even if it affected a lot of people, the solution doesn’t reduce performance.
- Even if it does reduce performance, it’s not my fault. It was the mobo maker’s fault.
- And if it was my fault, we didn’t mean it.
- And if we meant it, you the consumer deserved it because all you care about is who has the longest bar or biggest number in the benchmark.

I’m on Xitter and even I start to see the usual Intel defenders go down this Prayer, one step at a time. First it was denial, then it was minimization, etc etc. At some point, I wouldn’t be surprised to see them say RPL is in the past and we shouldn’t care anymore, so let’s just focus on ARL, ya? Please? Pretty please??

Yeah, they need Arrow Lake out there as soon as possible just to try and put this debacle behind them as fast as they can. But I don’t think they’re going to be able to get away with that. They need to do the right thing and return each and every one of those defective units, and not play games. They have already lost a lot of goodwill from customers. And if they keep messing around it will just continue to get worse. Can’t afford to throw their brand completely down the toilet.

 

[coercitiv]

The way I see it, it reminds me of the Narcissist’s Prayer:

To me it's prime material for The Four Stage Strategy:

 

[jdubs03]

part of his thinking is that there may never be a way to fix this in a way that doesnt screw up all of intel's future schedule.

if it isnt something a microcode can fix, they would have to

• assign engineers that originally worked on that gen who are likely already transferred to working on an upcoming gen. that would delay future chips.
• if the numbers are small they can try to offer replacements and eat the money loss, but if it is large scale then they would have to do a whole wafer run to make enough 'fixed' chips. unless intel runs their older fabs well past the next gen release to have spares for their corporate clients, this would seriously affect the profit margins.
• they cant even offer to upgrade them to arrowlake since they are on lga1700.

all of this makes it look like intel cant afford to do the right thing.

Quite the dilemma they’re in. I was thinking that last point, maybe discount an upgrade to ARL, but then customers would have to eat the cost of a new motherboard, unless they were to do some like half-off an upgrade for both CPU and mobo.

I bet they’re holding out on hope that the microcode update will be the fix they need so they can avoid these thorny issues.

 

[poke01]

To me it's prime material for The Four Stage Strategy:

Best show everrrr

 

[coercitiv]

Since it's "not-news", I'll quote the Intel confirmation of oxidation issue one more time

Also, thanks to Andreas Schilling pointing it out, the reddit post from Intel contains another important element:

For the Instability issue, we are delivering a microcode patch which addresses exposure to elevated voltages which is a key element of the Instability issue. We are currently validating the microcode patch to ensure the instability issues for 13th/14th Gen are addressed.

Elevated voltages are a "key element" of the instability issue, they're not described as the root/only cause. Looking forward to learn the other reason(s).