News Intel issues statement about CPU crashes, blames motherboard makers — BIOSes disable thermal and power protection, causing issues

[thestryker]

But there is no default power profile. What does default even mean when you are enabling XMP? In order to enable XMP you HAVE to set your power limits first, you can't proceed unless you choose one of the 3 available options (150 // 280 // unlimited).

XMP and CPU power have nothing to do with one another, so I'm not sure what you're talking about here. What board did you use which forced you to select a power profile in the BIOS before you could access XMP? I've never seen a board force manually selecting a power profile period, but that certainly doesn't mean it isn't a thing.

 

[craigss]

I think some Asus boards on Intel platforms defaulted to either an over volt, I rhink Jayz2cents recently did a segment on it as did GN.
To answer Terrylaze or whatever your called did AMD reduce some setting on mobos yes they did as soon as the problem was apparent, they did not try to shift the blame or cover anything up they just released updates and promised to honour any failures that had resulted, these are the actions of a reputable supplier, I have had personal bad experience from an Intel failure back in the core 2 duo days and as a result have never and will never buy Intel again period.
Intel should take a leaf out of that book as frankly should Nvidia for forcing the stupid connection on folks that is doomed to fail.

 

[_Shatta_AD_]

So this is Intel’s way of tacitly saying, “Not our problem, blame everyone else but us, OR just pay a premium for our K / higher level SKUs if you want more performance AND even if those cause instability, STILL not our fault cause you had a choice to buy a lower SKU that conform to our flexible baseline at our discretion.” This is what lack of competition brings. Full deniability and consumer must pay for company’s mistake and you can’t sue them cause you risk losing even more.

 

[TheHerald]

XMP and CPU power have nothing to do with one another, so I'm not sure what you're talking about here. What board did you use which forced you to select a power profile in the BIOS before you could access XMP? I've never seen a board force manually selecting a power profile period, but that certainly doesn't mean it isn't a thing.

I think all of them do, I've tried 3, an MSI unify X , Asus Apex and an Asrock something something.

 

[TerryLaze]

I think some Asus boards on Intel platforms defaulted to either an over volt, I rhink Jayz2cents recently did a segment on it as did GN.
To answer Terrylaze or whatever your called did AMD reduce some setting on mobos yes they did as soon as the problem was apparent, they did not try to shift the blame or cover anything up they just released updates and promised to honour any failures that had resulted, these are the actions of a reputable supplier, I have had personal bad experience from an Intel failure back in the core 2 duo days and as a result have never and will never buy Intel again period.
Intel should take a leaf out of that book as frankly should Nvidia for forcing the stupid connection on folks that is doomed to fail.

The reputable thing for AMD to do would have been to do a full recall, their cpus were blowing up, it was a faulty product, instead they locked down your unlocked cpu, they took away part of what you paid good money for and you are happy about them stealing from you.

Also what AMD did was to blame all of the OEMs for not using in spec vsoc settings, even though those where the settings they used for years before that so AMD knew that any new CPU would have to be able to use those settings.
AMD cpus also didn't have any internal safeguards, no, actually that was what made it a faulty product, the safety measure they had was burning out because it had no safety measure to protect it from too high voltages...
so how are they better than intel?!
https://www.tomshardware.com/news/a...use-identified-expo-and-soc-voltages-to-blame

Our sources also added further details about the nature of the chip failures — in some cases, excessive SoC voltages destroy the chips' thermal sensors and thermal protection mechanisms, completely disabling its only means of detecting and protecting itself from overheating. As a result, the chip continues to operate without knowing its temperature or tripping the thermal protections.

 

[TerryLaze]

So this is Intel’s way of tacitly saying, “Not our problem, blame everyone else but us, OR just pay a premium for our K / higher level SKUs if you want more performance AND even if those cause instability, STILL not our fault cause you had a choice to buy a lower SKU that conform to our flexible baseline at our discretion.” This is what lack of competition brings. Full deniability and consumer must pay for company’s mistake and you can’t sue them cause you risk losing even more.

Intel' official statement will be issued at the end of next month, this thing was a message to mobo makers and isn't targeted at end users at all.

Igor's Lab seems to have obtained a message originally destined for motherboard manufacturers

Intel will be publishing a public statement regarding issue status and Intel recommended BIOS setting recommendations targeted for May 2024.

 

[thestryker]

I think all of them do, I've tried 3, an MSI unify X , Asus Apex and an Asrock something something.

Very interesting as it's something I've never heard of before, and no review I've seen mentions it. I've only used a couple of lower end Asus Z series boards and a Gigabyte board none of which forced any power choices. Curious if anyone else has had the same experience.

 

[35below0]

So this is Intel’s way of tacitly saying, “Not our problem, blame everyone else but us, OR just pay a premium for our K / higher level SKUs if you want more performance AND even if those cause instability, STILL not our fault cause you had a choice to buy a lower SKU that conform to our flexible baseline at our discretion.” This is what lack of competition brings. Full deniability and consumer must pay for company’s mistake and you can’t sue them cause you risk losing even more.

They are to blame, but it was the motherboard manufacturers that introduced "intelligent" overclocking and sold it as a feature. None of us consumers really knew what it did but we assumed it was sensible and under control and generally a good thing.
And it wasn't.

That is the part where intel is blameless. What they should have done is cracked the whip at Asus, Gigabyte, MSi and the rest and warned them not to take things so far as to cause crashes, instability, or do things that mess up flagship intel CPUs.

 

[TheHerald]

Very interesting as it's something I've never heard of before, and no review I've seen mentions it. I've only used a couple of lower end Asus Z series boards and a Gigabyte board none of which forced any power choices. Curious if anyone else has had the same experience.

I'd take a picture for you but I really don't want to default everything. But this has been the case on Intel motherboards as far back as z490 if I remember correctly.

HWunboxed has kinda mentioned it as well, if you go his twitter account, back on the 13900k review, he said he specifically selected the unlimited option to run his review. Also igor's lab, ill find a link from his review and show you.

TLDR, reviewers (most of them) choose the unlimited option and then they run efficiency comparisons, lol



EG1. Ok this is a quote from igorslab

"I don’t stick rigidly to PL1 and PL2 with Intel and Alder Lake, but also run the whole thing with PL1 with the motherboard’s respective PL2 corner data. I also didn’t punish Raptor Lake with the motherboard’s all-in setting for water cooling and 4096 watt limit (i.e. unlimited) because of this; the tower cooler’s 288 watts will have to suffice here as usual."

Those are exactly the options im talking about, they are called "air cooler (125w pl) // tower cooler (250 / 280w depending on mobo) // AIO (which is the unlimited option)

Intel Core i9-13900K and Core i5-13600K Review - Showdown of the 13th Generation and a 3/4 Crown for the last big Monolith | Page 3 | igor´sLAB

With Raptor Lake, Intel now also ends the era of monolithic processors. This of course comes with another decent performance boost and it will be the purpose of this review to compare the new CPUs…

www.igorslab.de

 

[TheHerald]

They are to blame, but it was the motherboard manufacturers that introduced "intelligent" overclocking and sold it as a feature. None of us consumers really knew what it did but we assumed it was sensible and under control and generally a good thing.
And it wasn't.

That is the part where intel is blameless. What they should have done is cracked the whip at Asus, Gigabyte, MSi and the rest and warned them not to take things so far as to cause crashes, instability, or do things that mess up flagship intel CPUs.

The problem isn't with the unlimited power, CPUs are crashing because mobo manafacturers are using custom SVID settings (basically, LLC). What that does is - it lowers power draw , and therefore thermal throttling -which makes the CPUs score higher in cinebench numbers (since it thermal throttles at higher clockspeeds). And the CPU most likely IS stable with those best case SVID settings at full tilt (like cinebench), but it's probably highly unstable at mixed workloads with crazy overshoots up and down.

 

[thestryker]

HWunboxed has kinda mentioned it as well, if you go his twitter account, back on the 13900k review, he said he specifically selected the unlimited option to run his review.

That's just about review settings the same as Tom's which specifically cites:

We followed our standard policy of allowing the motherboard to exceed Intel's recommended power limits, provided the chip remains within warrantied operating conditions. Almost all enthusiast-class motherboards come with similar default settings, reflecting the out-of-box experience. These lifted power limits equate to more power consumption and more heat, but you get faster performance in exchange.

Steve (HUB) has spoken about varied default settings in his Z790 roundup and there's no reason he'd lie about the default settings especially since they vary by manufacturer.

EG1. Ok this is a quote from igorslab

"I don’t stick rigidly to PL1 and PL2 with Intel and Alder Lake, but also run the whole thing with PL1 with the motherboard’s respective PL2 corner data. I also didn’t punish Raptor Lake with the motherboard’s all-in setting for water cooling and 4096 watt limit (i.e. unlimited) because of this; the tower cooler’s 288 watts will have to suffice here as usual."

He's just talking about how he won't test them with unlimited power.

I'd take a picture for you but I really don't want to default everything.

I wouldn't expect you to.

Those are exactly the options im talking about, they are called "air cooler (125w pl) // tower cooler (250 / 280w depending on mobo) // AIO (which is the unlimited option)

MSI has that newbie trap bs naming in their BIOS, but until 12th gen there was no unlimited. I very much hate that MSI decided to use cooling for power profiles as that's a ridiculous approach they should just be transparent about it so people with less experience can't get confused.

 

[TheHerald]

That's just about review settings the same as Tom's which specifically cites:

Steve (HUB) has spoken about varied default settings in his Z790 roundup and there's no reason he'd lie about the default settings especially since they vary by manufacturer.

He's just talking about how he won't test them with unlimited power.

I wouldn't expect you to.

MSI has that newbie trap bs naming in their BIOS, but until 12th gen there was no unlimited. I very much hate that MSI decided to use cooling for power profiles as that's a ridiculous approach they should just be transparent about it so people with less experience can't get confused.

No, igors lab is specifically talking about the tower cooler option, that's what the 288w power limit option is called on his msi board. But he is using an auto translator from German to English since the original review is in German.

I know motherboards have varied default settings but since we are talking about XMP enabled systems, there is no default anymore, the user has to specifically choose one of the 3 options. Default means just pressing the power button and booting straight into windows, and in this case yes, most high end mobos use unlimited power.

MSI also shows you the power limits for those options but in small letters 😁

Eg1. Hwunboxed just replied to my comment and he verified that indeed those options exist. In his mobo the air cooler is called intel stock cooler but still, it's there. He is claiming that those options only exist on lower end mobos but that is NOT the case unless asus apex and msi unify x are low end

 

[thestryker]

No, igors lab is specifically talking about the tower cooler option, that's what the 288w power limit option is called on his msi board. But he is using an auto translator from German to English since the original review is in German.

I know motherboards have varied default settings but since we are talking about XMP enabled systems, there is no default anymore, the user has to specifically choose one of the 3 options. Default means just pressing the power button and booting straight into windows, and in this case yes, most high end mobos use unlimited power.

MSI also shows you the power limits for those options but in small letters 😁

Eg1. Hwunboxed just replied to my comment and he verified that indeed those options exist. In his mobo the air cooler is called intel stock cooler but still, it's there. He is claiming that those options only exist on lower end mobos but that is NOT the case unless asus apex and msi unify x are low end

The options exist, but that doesn't mean you're forced to pick one before you can do anything else. Enabling XMP is not some sort of excuse for other settings and I don't know why you keep getting hung up on it. This isn't manually tweaking memory settings to get the right combination of clocks and latency.

 

[35below0]

Eg1. Hwunboxed just replied to my comment and he verified that indeed those options exist. In his mobo the air cooler is called intel stock cooler but still, it's there. He is claiming that those options only exist on lower end mobos but that is NOT the case unless asus apex and msi unify x are low end

Speaking of of lower end motherboards, the GB z790 UD r1.0 has never been accused of being a high end motherboard, and yet Gigabyte stuck it's PerfDrive onto it.

https://imgur.com/iNHJyIC

View: https://i.imgur.com/iNHJyIC.png

Instant 6Ghz?
E-Core disable? mmk...

I don't know how hard any CPU needs to be pushed in a 60A 8+8 VRM motherboard. It's a pretty silly feature but i suspect Gigabyte calculated that not including it would make the motherboard sell worse.

At least it's user enabled and not default.

The problem isn't with the unlimited power, CPUs are crashing because mobo manafacturers are using custom SVID settings (basically, LLC). What that does is - it lowers power draw , and therefore thermal throttling -which makes the CPUs score higher in cinebench numbers (since it thermal throttles at higher clockspeeds). And the CPU most likely IS stable with those best case SVID settings at full tilt (like cinebench), but it's probably highly unstable at mixed workloads with crazy overshoots up and down.

Gigabyte highlights exactly this. And they call it "optimization" which is misleading. Optimal means stable, not over-the-top.

 

[TheHerald]

The options exist, but that doesn't mean you're forced to pick one before you can do anything else. Enabling XMP is not some sort of excuse for other settings and I don't know why you keep getting hung up on it. This isn't manually tweaking memory settings to get the right combination of clocks and latency.

You can't do anything else on the bios until you either select one of the 3 options or hit the small cancel button. It's an image that basically takes up the whole screen on both msi and asus. I'll send you a picture when I'm back so you understand exactly what we are talking about

 

[TerryLaze]

The options exist, but that doesn't mean you're forced to pick one before you can do anything else. Enabling XMP is not some sort of excuse for other settings and I don't know why you keep getting hung up on it. This isn't manually tweaking memory settings to get the right combination of clocks and latency.

Just an example, but there have been several similar stories over the years.

MSI intros 'CPU Cooler Tuning' BIOS feature on Intel mainboards

Makes it easy for its board users to gain additional performance from improved cooling.

hexus.net

. 'CPU Cooler Turning' is automatically invoked when a system boots for the first time or after the CMOS has been cleared.

 

[TheHerald]

Just an example, but there have been several similar stories over the years.

MSI intros 'CPU Cooler Tuning' BIOS feature on Intel mainboards

Makes it easy for its board users to gain additional performance from improved cooling.

hexus.net

Yeap, that's the one, it's very similar on my mobo

 

[Order 66]

Folks, time to stop all the fanboy/shill nonsense. Don't make me have to pull this car over!

Civility is required by all. It is okay to disagree, but not to be disagreeable.

Thank you.

On the topic, shouldn't these CPUs be "smart" enough to know what will cause them to crash or not? I mean shouldn't CPUs be able to know when something isn't stable and at least try to recover from it, sort of like how GPU drivers can, or is that not possible?

 

[-Fran-]

The reputable thing for AMD to do would have been to do a full recall, their cpus were blowing up, it was a faulty product, instead they locked down your unlocked cpu, they took away part of what you paid good money for and you are happy about them stealing from you.

Also what AMD did was to blame all of the OEMs for not using in spec vsoc settings, even though those where the settings they used for years before that so AMD knew that any new CPU would have to be able to use those settings.
AMD cpus also didn't have any internal safeguards, no, actually that was what made it a faulty product, the safety measure they had was burning out because it had no safety measure to protect it from too high voltages...
so how are they better than intel?!
https://www.tomshardware.com/news/a...use-identified-expo-and-soc-voltages-to-blame

You're wrong on both accounts though.

AMD did acknowledge they didn't communicate with their partners correctly and fixed that, but also stated the values were never "in spec".

Relevant quote:
"The issues are known to occur either when a user adjusts the SoC voltage manually to a value beyond 1.3V or when the motherboard firmware automatically increases the SoC voltage beyond 1.3V when an EXPO memory overclocking profile is engaged (The EXPO profile itself does not increase the SoC voltage, the board vendors assign a pre-determined value of their own to support the increased speed of the EXPO profile)".

https://www.tomshardware.com/news/a...t-on-ryzen-burnout-issues-limits-soc-voltages

Their fix was just making sure motherboard makers would not push the limits beyond what AMD themselves considers "safe" (or so I read in between the lines). As this is my interpretation, I could be wrong.

Also, the CPUs were never faulty and proof of that is everyone that could have been potentially affected is still using their CPUs just fine, no?

Do you think Intel's 13 and 14 series affected by this need to be recalled or something? Because that's what your "logical" reasoning makes me believe?

Regards.

 

[TheHerald]

You're wrong on both accounts though.

AMD did acknowledge they didn't communicate with their partners correctly and fixed that, but also stated the values were never "in spec".

Relevant quote:
"The issues are known to occur either when a user adjusts the SoC voltage manually to a value beyond 1.3V or when the motherboard firmware automatically increases the SoC voltage beyond 1.3V when an EXPO memory overclocking profile is engaged (The EXPO profile itself does not increase the SoC voltage, the board vendors assign a pre-determined value of their own to support the increased speed of the EXPO profile)".

https://www.tomshardware.com/news/a...t-on-ryzen-burnout-issues-limits-soc-voltages

Their fix was just making sure motherboard makers would not push the limits beyond what AMD themselves considers "safe" (or so I read in between the lines). As this is my interpretation, I could be wrong.

Also, the CPUs were never faulty and proof of that is everyone that could have been potentially affected is still using their CPUs just fine, no?

Do you think Intel's 13 and 14 series affected by this need to be recalled or something? Because that's what your "logical" reasoning makes me believe?

Regards.

This whole situation is a big fat nothing burger. That applies equally to amd and intel. You buy an unlocked cpu and an unlocked motherboard, you are on your own. If you don't know what the hell you are doing stick to locked parts with xmp off.

Enabling xmp alone can absolutely kill your cpu. We've seen that happening with the amd parts, but it's also the case with intel parts (they are just much more tolerable to high voltages but still). It's been the case for at least 10 years with the first skylake parts. Try to boot a 4000 MHz kit on any skylake or skylake derivative and the auto applied IO voltage can exceed 1.5 volts (I've seen 1.65 for 4400 MHz kits).

Right now if I try to boot my 12900k with pcore frequencies of 5.6 ghz the auto applied core voltage is 1.63 volts. I'm not blaming Intel, amd or mobo makers. With that said, sure there should be an option somwehwere to set intel defaults, but who would use that anyways? Isn't everyone basically enabling XMP?

Is there a single motherboard that allows you to run jedec? Even without enabling XMP mobos from both amd and intel tighten timings (IO rtls etc) when they should be just running jedec specs.

 

[palladin9479]

Wow so much wrong in here....

Firstly, Intel has zero control over what Motherboard manufacturers do in the BIOS, zero. They ship CPU's with a default set of operating parameters and those parameters are changeable on unlocked SKU's to enable consumers to overclock. Motherboard manufacturers in an attempt to win more market share have built "automatic overclock" capabilities into their boards. They then set that as "default" to win more youtube reviewer benchmarks, and since that can make systems unstable they then did the overclocker favorite trick of cranking up the voltage to compensate. That worked for awhile but recent generation of CPU's are far more sensitive then previous generation with the predictable results of instability or in AMD's case just burning up.

Seriously this happened to me on a recently purchased Asrock 650E board with AMD 8600G and DDR5-6400 memory. Put it all together, and set the memory profile, reboot and it can't boot or anything. Find out that the Asrock bios was setting the memory controller VCore to 1.4v, which also sets the fabric voltage to 1.4v and is outside the 1.3v limit AMD recommends and way beyond the stability point of 1.10~1.15v. Set the memory back to default and the controller voltage goes back to default. It is impossible to set the memory profile without the bios attempting to OC to CPU, I had to manually change all the other voltages back to safe ranges to get the system stable.

This is not a CPU manufacturer problem, it's a MB vender problem with how they set the automatic OC algorithms inside their BIOS's. They do it in an attempt to win "benchmarks" from reviewers like J2C, GN and Toms.

 

[-Fran-]

Wow so much wrong in here....

Firstly, Intel has zero control over what Motherboard manufacturers do in the BIOS, zero. They ship CPU's with a default set of operating parameters and those parameters are changeable on unlocked SKU's to enable consumers to overclock. Motherboard manufacturers in an attempt to win more market share have built "automatic overclock" capabilities into their boards. They then set that as "default" to win more youtube reviewer benchmarks, and since that can make systems unstable they then did the overclocker favorite trick of cranking up the voltage to compensate. That worked for awhile but recent generation of CPU's are far more sensitive then previous generation with the predictable results of instability or in AMD's case just burning up.

Seriously this happened to me on a recently purchased Asrock 650E board with AMD 8600G and DDR5-6400 memory. Put it all together, and set the memory profile, reboot and it can't boot or anything. Find out that the Asrock bios was setting the memory controller VCore to 1.4v, which also sets the fabric voltage to 1.4v and is outside the 1.3v limit AMD recommends and way beyond the stability point of 1.10~1.15v. Set the memory back to default and the controller voltage goes back to default. It is impossible to set the memory profile without the bios attempting to OC to CPU, I had to manually change all the other voltages back to safe ranges to get the system stable.

This is not a CPU manufacturer problem, it's a MB vender problem with how they set the automatic OC algorithms inside their BIOS's. They do it in an attempt to win "benchmarks" from reviewers like J2C, GN and Toms.

It's about communication and how they tell the AIBs how to do things. On a case-by-case we'll find, as you point out, situations where AMD or Intel do not know the AIB is not following their guidance or specs, and that is fully on the AIB, but in most situations both AMD and Intel control the flow of information into the AIBs. When it is just one that does it wrong, it's on the AIB (think EVGA and the 3090 issue, for instance), but when all of them have this issue, that falls on AMD/Intel and their communication to them.

It's not more complicated than that, I'd say.

Regards.

 

[t3t4]

Allowing any chip to draw unlimited amounts of power seems like a DUH factor to me..... Meaning, it's going to cause a problem, DUH!

When I first built my 13900k system I ran all kinds of bench tests on it. Dumping more power into the thing does NOT make it faster, it only makes it hotter! I limited mine at 225 watts if memory serves with an overclock. But I had it pulling high 300's at one point and it was just stupid hot with no benefit, only loss! I don't care what Intel says in their specs, I blame the mother board manufacturers for doing an obviously stupid thing!

With unlimited power comes unlimited problems..... DUH!

 

[palladin9479]

It's about communication and how they tell the AIBs how to do things. On a case-by-case we'll find, as you point out, situations where AMD or Intel do not know the AIB is not following their guidance or specs, and that is fully on the AIB, but in most situations both AMD and Intel control the flow of information into the AIBs. When it is just one that does it wrong, it's on the AIB (think EVGA and the 3090 issue, for instance), but when all of them have this issue, that falls on AMD/Intel and their communication to them.

It's not more complicated than that, I'd say.

Regards.

They already do that. Intel provides the chip specifications to the motherboard manufacturers and that's where it ends. It's the motherboard manufacturers job to implement those specifications. Motherboard manufacturers doing silly stuff is the fault of the motherboard manufacturer, not the chip maker.

People's brand hate needs to stop. The reason most all the board manufactures do this is to "keep up" with each other on the youtube reviews. Reviewers love benchmarks, "board A is 5% faster then board B" then everyone wants to buy board A. Since almost everything that used to be board specific has moved into IMC's and dedicated vendor chipsets, there isn't much left for motherboard vendors to differentiate themselves with so they have started to do overclocking by default.

 

[palladin9479]

Allowing any chip to draw unlimited amounts of power seems like a DUH factor to me..... Meaning, it's going to cause a problem, DUH!

When I first built my 13900k system I ran all kinds of bench tests on it. Dumping more power into the thing does NOT make it faster, it only makes it hotter! I limited mine at 225 watts if memory serves with an overclock. But I had it pulling high 300's at one point and it was just stupid hot with no benefit, only loss! I don't care what Intel says in their specs, I blame the mother board manufacturers for doing an obviously stupid thing!

With unlimited power comes unlimited problems..... DUH!

The CPU manufacture has zero control on the power to the CPU, they can only write the specification then ask the motherboard makers to follow it. Voltage control and power distribution is something the motherboard does and why they like to use "X/Y/Z phase power distribution etc.." as selling points.