Monday, April 29th 2024

Intel Statement on Stability Issues: "Motherboard Makers to Blame"

A couple of weeks ago, we reported on NVIDIA directing users of Intel's 13th Generation Raptor Lake and 14th Generation Raptor Lake Refresh CPUs to consult Intel for any issues with system stability. Motherboard makers, by default, often run the CPU outside of Intel's recommended specifications, overvolting the CPU through modifying voltage curves, automatic overclocks, and removing power limits.

Today, we learned that Igor's Lab has obtained a statement from Intel that the company prepared for motherboard OEMs regarding the issues multiple users report. Intel CPUs come pre-programmed with a stock voltage curve. When motherboard makers remove power limits and automatically adjust voltage curves and frequency targets, the CPU can be pushed outside its safe operating range, possibly causing system instability. Intel has set up a dedicated website for users to report their issues and offer support. Manufacturers like GIGABYTE have already issued new BIOS updates for users to achieve maximum stability, which incidentally has recent user reports of still being outside Intel spec, setting PL2 to 188 W, loadlines to 1.7/1.7 and current limit to 249 A. While MSI provided a blog post tutorial for stability. ASUS has published updated BIOS for its motherboards to reflect on this Intel baseline spec as well. Surprisingly, not all the revised BIOS values match up with the Intel Baseline Profile spec for these various new BIOS updates from different vendors. You can read the statement from Intel in the quote below.
Intel has observed that this issue may be related to out of specification operating conditions resulting in sustained high voltage and frequency during periods of elevated heat.

Analysis of affected processors shows some parts experience shifts in minimum operating voltages which may be related to operation outside of Intel specified operating conditions.

While the root cause has not yet been identified, Intel has observed the majority of reports of this issue are from users with unlocked/overclock capable motherboards.

Intel has observed 600/700 Series chipset boards often set BIOS defaults to disable thermal and power delivery safeguards designed to limit processor exposure to sustained periods of high voltage and frequency, for example:
  • Disabling Current Excursion Protection (CEP)
  • Enabling the IccMax Unlimited bit
  • Disabling Thermal Velocity Boost (TVB) and/or Enhanced Thermal Velocity Boost (eTVB)
  • Additional settings which may increase the risk of system instability:
  • Disabling C-states
  • Using Windows Ultimate Performance mode
  • Increasing PL1 and PL2 beyond Intel recommended limits
Intel requests system and motherboard manufacturers to provide end users with a default BIOS profile that matches Intel recommended settings.

Intel strongly recommends customer's default BIOS settings should ensure operation within Intel's recommended settings.

In addition, Intel strongly recommends motherboard manufacturers to implement warnings for end users alerting them to any unlocked or overclocking feature usage.

Intel is continuing to actively investigate this issue to determine the root cause and will provide additional updates as relevant information becomes available.

Intel will be publishing a public statement regarding issue status and Intel recommended BIOS setting recommendations targeted for May 2024.
Source: Igor's Lab
Add your own comment

266 Comments on Intel Statement on Stability Issues: "Motherboard Makers to Blame"

#1
Hyderz
from what i know the asus and gigabyte have different power (but stable) limits in place... the gigabyte ones have quite the substantial performance hit..
the asus one also have a performance hit but not as much as gigabyte... now is the chance for amd to gain more sales in the time being
Posted on Reply
#2
BoggledBeagle
125W power limit of Gigabyte is completely unnecessary, 14900K already behaves pretty nicely at 180W. Anything below 160W is a waste of the CPU potential.
Posted on Reply
#3
napata
Hyderzfrom what i know the asus and gigabyte have different power (but stable) limits in place... the gigabyte ones have quite the substantial performance hit..
the asus one also have a performance hit but not as much as gigabyte... now is the chance for amd to gain more sales in the time being
The Gigabyte one is a scam. For the 13900K & 14900K they need to be stable at 253W PL1 and not 125W.
Posted on Reply
#4
Zubasa
napataThe Gigabyte one is a scam. For the 13900K & 14900K they need to be stable at 253W PL1 and not 125W.
It seems that Gigabyte for some reason applied the PL of the non-K 14900 to the 14900K. EDIT: Nope, not even that, it is just some weird value.
Also, speaking of "scam" since Intel decided for the 12 gen+ K series PL1=PL2, they should just have labeled the 14900K as 253W TDP.
Posted on Reply
#5
BoggledBeagle
napata... 13900K & 14900K ... need to be stable at 253W ...
The root problem is at what power draw (and frequencies and voltage) the silicon IS ABLE TO RELIABLY WORK LONG TERM. This is a physical problem.

These limits are tied to the silicon chips, not Intel paperwork. Intel will probably need to rewrite the paperwork and enforce compliance with it.
Posted on Reply
#6
napata
BoggledBeagleThe root problem is at what power draw (and frequencies and voltage) the silicon IS ABLE TO RELIABLY WORK LONG TERM. This is a physical problem.

These limits are tied to the silicon chips, not Intel paperwork. Intel will probably need to rewrite the paperwork and enforce compliance with it.
This has nothing to do with the topic at hand. How long chips last at what voltage and/or power draw is a very different subject than this issue. These limits get tested by Intel/TSMC/foundries and the "paperwork" from Intel/AMD/Nvidia/Apple/ETC is based on that data.
Posted on Reply
#7
Guwapo77
Gigabyte is making a statement to Intel, that's the only way I an see this at this time. Intel knew about the motherboard makers having their own values for years and never said a word. Now that complaints are coming in, Intel points the finger. Gigabyte in return says they will no longer be liable for any more of Intel chips that are pushed to the limit as is, drop them to Ultra Mega Safe levels. My only question is how long will they remain at this ultra low levels?
Posted on Reply
#8
zmeul
The Intel baseline should've been the factory defaults, not the optional

Most users at home don't update the BIOS and will also won't know where these options are if they do update it
Posted on Reply
#9
stimpy88
This is what AMD has reduced Intel to... Setting motherboard defaults as a last gasp measure to OC just enough extra to try to stay up with the competition, then blame the OEM's when they get caught frying CPU's? So will Intel do the honourable thing and offer replacement CPU's for the ones that have been degraded and no longer run stable at stock?

I find it, and all the fanboi cope... delicious!

Stay classy Intel!
Posted on Reply
#10
john_
It seems everyone knew about the problems (Intel, motherboard makers, Nvidia, maybe even the press? ) except the consumers.

So, we have 3-4 years of benchmarks online that do not correspond in reality, but instead mislead consumers. And that's before we add that those results are possible with ultra expensive cooling system and top of the line motherboard.
Posted on Reply
#11
ratirt
No shock here with the news. The more you push the CPU's to the limit the more possibility these will fail at some point and the CPUs fail number will grow same way degradation will and failed CPUs from the factory will not adhere to the requirements.
Not sure how intel would want to milk that cow but I guess they have just acquired a tipping point with it.
Fix what is there to fix and move on with something new and less power hungry.
Posted on Reply
#12
Keullo-e
S.T.A.R.S.
Pretty much coping when you've put out an Emergency Edition with insane factory overclock for being able to compete with the competitor.
Posted on Reply
#13
john_
Hyderznow is the chance for amd to gain more sales in the time being
Intel is changing platform. Tech sites will power limit 14900K in their reviews, because that's the correct thing to do after this fiasco, making the Intel Ultra models look even faster.
Funny isn't it?
Posted on Reply
#14
ratirt
john_It seems everyone knew about the problems (Intel, motherboard makers, Nvidia, maybe even the press? ) except the consumers.

So, we have 3-4 years of benchmarks online that do not correspond in reality, but instead mislead consumers. And that's before we add that those results are possible with ultra expensive cooling system and top of the line motherboard.
I'm not sure about the misleading. From what I understood, about the media presenting Intel CPUs (not all media) they have mentioned these issues just to stay competitive. Intel knows how much they can push the cpus and apparently they have miscalculated. I knew it is not a good thing to get these and push them to the limit due to power they suck to run. Again. not a biggie for me suddenly some CPUs crapped themselves up and stopped working properly.
Posted on Reply
#15
pressing on
BoggledBeagleThe root problem is at what power draw (and frequencies and voltage) the silicon IS ABLE TO RELIABLY WORK LONG TERM. This is a physical problem.

These limits are tied to the silicon chips, not Intel paperwork. Intel will probably need to rewrite the paperwork and enforce compliance with it.
This issue arose mainly from problems that some users were having with games using the Unreal engine. The hardware involved typically was a 13700K/14700K or 13900K/14900K CPU with a high end NVIDIA card such as the 4080 Super and the 4090. The integral ultra high speed decompressor in Unreal games - Oodle - was detecting errors in decompressed data. Epic's policy with Oodle is that decompression errors are fatal errors and users experienced "crashes" where the game terminated with an Oodle error message. Epic say that the problem was caused by overclocked Intel CPUs going out of spec.

To give an example of what was happening, a gamer with an undervolted overclocked 13700K upgraded their graphics card to a 4080 Super. A game using the Unreal engine would not longer run, crashing constantly. They reduced the overclock by 100 MHz - game then runs again without errors. So this seems to be a configuration issue rather than anything else. This user was able to fix the error. If the cause had been motherboard default settings and they were not familiar with BIOS settings it would have been a different matter.
Posted on Reply
#16
john_
stimpy88So will Intel do the honourable thing and offer replacement CPU's for the ones that have been degraded and no longer run stable at stock?
I don't think this is just "honorable". 14th series is not that long in the market and warranties are probably still valid. If Intel plays the game of "overclocking the CPU will invalidate your warranty", what will follow will be pretty bad. They will replace those CPUs and hope this gets forgotten fast and quietly.
Posted on Reply
#17
ARF
Oh, the irony, which "normal" company would sell CPUs with the insane value of 253W PL1 limit?!
Simply too much wrong doing, corruption, greed and hunger, instead of admitting the defeat and saying they are the inferior option... :mad:
Posted on Reply
#18
Ferrum Master
I was actually telling it from the start. OEM parters are the ones at blame for the most of time.
Posted on Reply
#19
64K
Intel had to have known that mobo manufacturers were pushing their CPUs beyond specs while reviewers were posting such great results from the i9s. Intel isn't directly to blame for what has happened but their silence about what was going on for months before the issue hit the news is on them.
Posted on Reply
#20
Crackong
So Intel admitting they don't have a default profile and relies on motherboard manufacturers to make their own 'Default' .

And also it is Intel themselves using PL1 = 253W in their own CPU performance index,
  • Increasing PL1 and PL2 beyond Intel recommended limits
Maybe every review site should honor Intel's decision and re-do 12/13/14 gen benchmark with PL1&PL2 = 125W, I bet the results will be fascinating.

Posted on Reply
#21
dgianstefani
TPU Proofreader
Ferrum MasterI was actually telling it from the start. OEM parters are the ones at blame for the most of time.
dgianstefaniUnnecessary for stability. Just run the actual Intel stock settings. Not the motherboard "stock" settings.

You can lock your clocks/voltage for consistent performance, but I don't see how underclocking will give better performance than actual stock settings. Especially as the CPU will still have variable clocks unless you do a static tune.

It's the motherboards overvolting chips past limits that cause this issue, not the chipmakers. I'd argue the strongest criticism you can legitimately make is that Intel and AMD need to be stricter at enforcing their stock settings with their board partners.

The AMD melting chips/socket issue a while back was something you could actually blame on AMD, because it was their AGESA/EXPO algorithms that were causing the overvolting.
Yup.

Reminder that TPU reviews are all done using Intel Spec limits, not motherboard defaults.

I find it incredible that even after the Intel memo, board makers are still making up values, rather than literally just following the Intel baseline spec.

Another thing to mention, some of the crashing is due to too low voltage. Motherboard makers change the voltage curve when they fiddle with settings, and idle/low load voltages can drop below the Intel spec, enough to cause a crash. It's not as simple as "CPU uses too much power", although you can also crash from the voltage going too high.
Analysis of affected processors shows some parts experience shifts in minimum operating voltages
Posted on Reply
#22
close
This is the oldest trick in the book. Make your product look good at launch, make it work well later. Intel and OEMs held hands while benchmarks were saying "king of the single threaded performance". Turns out that was a less than sustainable setup.

Let's see how many reviewers update their reviews and benchmarks to account for this. A lot of them get some freebies if they just "forget" to do it, or they're just afraid Intel will blacklist them. I still remember that guy who eventually left AnandTech through the backdoor enthusiastically writing about the incredible 5GHz all core Intel chip that turned out to run under a 1kW chiller. Even for AT, it took him forever to come up with an article which was basically excusing Intel. If it happens at once-relevant media outlets, it can happen to others too. Intel is probably "blackmailing" honest reviewers or rewarding the dishonesty. The lesson is if you want to buy an Intel CPU, stay away from the top end, and buy it 1+ years after launch, you may get a slim chance of seeing the real situation by then. (I'm posting from one of AMD's "fake 8-cores" right now but they seem to have cleaned up their act a bit in the past decade).
Posted on Reply
#23
dgianstefani
TPU Proofreader
closeThis is the oldest trick in the book. Make your product look good at launch, make it work well later. Intel and OEMs held hands while benchmarks were saying "king of the single threaded performance". Turns out that was a less than sustainable setup.

Let's see how many reviewers update their reviews and benchmarks to account for this. A lot of them get some freebies if they just "forget" to do it, or they're just afraid Intel will blacklist them. I still remember that guy who eventually left AnandTech through the backdoor enthusiastically writing about the incredible 5GHz all core Intel chip that turned out to run under a 1kW chiller. Even for AT, it took him forever to come up with an article which was basically excusing Intel. If it happens at once-relevant media outlets, it can happen to others too. Intel is probably "blackmailing" honest reviewers or rewarding the dishonesty. The lesson is if you want to buy an Intel CPU, stay away from the top end, and buy it 1+ years after launch, you may get a slim chance of seeing the real situation by then. (I'm posting from one of AMD's "fake 8-cores" right now but they seem to have cleaned up their act a bit in the past decade).
Anandtech is one of the other review sites that also tests using Intel Spec, not the motherboard defaults.
dgianstefaniYep. Because most Z series motherboards aren't running Intel stock spec. They're doing their own thing. Most users don't know how to tune either, or properly navigate a bios, meaning the problematic "stock" settings as set by the board manufacturers cause issues like this.

Since it's not the board maker who has to deal with returns etc. There's no real consequence if they screw up their "ai boost" or whatever they choose to call their "optimised defaults" etc.
I wonder if board makers are taking this seriously since again, they're not the ones who have to deal with returns, most of the time, unless people realise it's the motherboard.
Posted on Reply
#24
stimpy88
dgianstefaniAnandtech is one of the other review sites that also tests using Intel Spec, not the motherboard defaults.


I wonder if board makers are taking this seriously since again, they're not the ones who have to deal with returns, most of the time, unless people realise it's the motherboard.
The trouble is that nobody knows exactly what other settings have been manipulated. Did you read that list of things Intel posted the other day? There is more than power settings, which seems to be all people are talking about... So, has this site followed these additional recommendations when reviewing these CPU's? Not accusing, just wanting to get a clear view.

Posted on Reply
#25
Crackong
dgianstefani Intel Spec limits, not motherboard defaults.
I want to ask.
By Intel Spec Limit do you mean PL1 = 253W ?

Since it is the performance Index listed on Intel's own website.
Posted on Reply
Add your own comment
May 15th, 2024 10:55 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts