Owners of 13th/14th Gen Raptor Lake CPUs - Media Reports of serious stability issues

[FlyingScot]

This information is being offered with the intention of keeping owners of these CPUs informed, i.e. knowledge is power. Perhaps if enough people become aware of this situation (and the information provided is indeed accurate) then maybe Intel will offer affected users alternative remedies other than the usual warranty replacement process.

Owners of 13th/14th gen i9’s are reporting what looks like rapid degradation over a period of months that leads to sudden instability. Users are then forced to increase voltage and/or drop boost frequencies to stabilize their systems. In some cases, this problem resurfaces and requires the same approach to be repeated. Another major cause of concern raised in the video, and in user comments, is the stability of the Integrated Memory Control (IMC) and the relationship to DDR5 frequencies above 4200. Random instability due to IMC issues, and perhaps degradation of the IMC, cannot be ruled out. This is clearly a story that is still developing. Owners of i9 CPUs, and indeed all Raptor Lake CPUs, should stay tuned. Follow up videos on this story are promised by Steve at Gamers Nexus.

What’s most alarming about this situation is the fact that even users who took every preventative safety measure, i.e. not overclocking, low PL1/PL2, good cooling, even undervolting, etc., are still reporting instability after just a few months of usage, presumably from what could be degradation. And, judging from user comments, these issues may not just be limited to the i9's.

I do not own one of these chips, but I have been following this story with great interest.

UPDATE 1: For folks who want a quick way to see where things stand at the moment, I suggest you start reading this thread from here. In the first post by forum member, CiTay, you will find several relevant links of importance. I will endeavour to keep adding to this shortcut list as events develop.

UPDATE 2: By now, many of you will be aware that Intel and the motherboard manufacturers have collaborated to release new BIOS code to address the Raptor Lake instability and degradation issues. Should you choose to update to the latest BIOS (with microcode 0x129) and then experience thermal and/or performance issues, you may find this new guide helpful in addressing those concerns.

UPDATE 3: [Sep 26, 2024] Intel and the motherboard manufacturers have collaborated to release new BIOS code (0x12B) to address the Raptor Lake instability and degradation issues. For more information, jump to the following post in this thread.

ADDITION 1: Don't forget to stop by our very own online Raptor Lake Survey, where you can see how others have configured their Raptor Lake systems to balance performance and stability, and to reduce the risk of degradation. You can check it out here.

 

[FlyingScot]

For those of you who don't want to watch the video (or may have problems with translation), I offer these two follow-up articles that summarize the situation.

Also Stability Issues in 13th and 14th Gen Intel CPUs on W680 based Servers

Intel's 13th and 14th generation processors, particularly the Core i9-13900K and Core i9-14900K models based on the Raptor Lake architecture, are experiencing significant stability problems not only in gaming scenarios but also in server environments.

www.guru3d.com

Game developer says Intel should recall its defective, crash-prone CPUs

The number of people and organizations forced to experience crashes and general instability on Intel's latest CPU models keeps growing. Now, a game developer is blatantly pointing...

www.techspot.com

 

[edola151e02d2]

Intel’s woes with Core i9 CPUs crashing look worse than we thought – Team Blue really needs to act now to fix this mess

YouTuber highlights further concerns uncovered via game crash logs and more besides

www.techradar.com

https://wccftech.com/warframe-intel-14th-13th-gen-cpus-responsible-for-instability-issues/

 

[Raiken]

Are 13th and 14th generation i7s completely safe? This and the problems with motherboard manufacturer's default profile have me questioning? One of my friends wants to build his new PC with a 14th generation i7.

 

[FlyingScot]

Are 13th and 14th generation i7s completely safe? This and the problems with motherboard manufacturer's default profile have me questioning? One of my friends wants to build his new PC with a 14th generation i7.

The i7 question is a difficult one to answer because we have to consider the fact that some/many/all users who are struggling with i7 instability could be suffering from a whole host of reasons, such as motherboard defaults, XMP challenges, thermals, other hardware issues, or even Windows/Application bugs, instead of degradation.

Maybe this new video by BuildZoid will help you reach a decision. EDIT: The first 19mins are hard to listen to....suggest skipping.

This forum has several threads with i7 owners trying to find the sweet spot between performance and thermals. Those might be worth a read, too.

 

[Raiken]

The i7 question is a difficult one to answer because we have to consider the fact that some/many/all users who are struggling with i7 instability could be suffering from a whole host of reasons, such as motherboard defaults, XMP challenges, thermals, other hardware issues, or even Windows/Application bugs, instead of degradation.

Maybe this new video by BuildZoid will help you reach a decision.

This forum has several threads with i7 owners trying to find the sweet spot between performance snd thermals. Those might be worth a read, too.

Thank you very much. I'll check it out right away.

 

[citay]

We should link to the two other major threads about this issue as well here:

https://forum-en.msi.com/index.php?threads/msi-bios-update-to-address-intel-raptor-lake-instability-enforce-intel-standard-power-limits.396634/

https://forum-en.msi.com/index.php?threads/msi-optimized-defaults-fried-my-cpu-and-it-is-likely-frying-yours.396665/

This whole issue is still in flux, so the later replies will be more up-to-date with what has transpired since the threads were started. Still, there is perhaps some valuable information in some of those replies.

I would say, it might be a good idea to set power and current limits according to the Intel "Performance Profile" recommendations for 13th/14th gen i7/i9, provided that the cooling can deal with it (if not, set lower power limits). The 12th gen CPUs don't seem to be affected because they had lower frequencies (and thus lower voltages to reach them), same goes for the i3/i5 models which tend to run more in the comfort zone of the silicon and are not pushed to the limit anywhere near as for example the 14th gen i7 and i9 models (the worst offenders).

The latest Intel recommendations to date are the following:

So they're saying:
13600K/14600K should have 143W Short / 125W Long / 175A Current limit for the Baseline profile and 181W/181W/200A for the Performance profile.
13700K/14700K should have 188W Short / 125W Long / 249A Current limit for the Baseline profile and 253W/253W/307A for the Performance profile.
13900K/14900K should have 188W Short / 125W Long / 249A Current limit for the Baseline profile and 253W/253W/307A for the Performance profile.
13900KS/14900KS should have no Baseline profile, 253W Short / 253W Long / 307A Current limit for the Performance profile and 320W/320W/400A for the Extreme profile.

The big question mark here is about the Extreme profile for the 13900KS and 14900KS. Despite them being - most likely - sold in much lower numbers than the regular i7/79 models, they show up in the instability statistics with a reasonable percentage, meaning a bunch of them are unstable from stock and/or degrading. Which makes sense, because on average, they use the highest voltages out of all the CPU models. So then why is the 14700K not allowed an Extreme profile at all, then the regular i9 models are allowed an Extreme profile with 253W/400A, but then the i9 KS models are allowed an Extreme profile with 320W/400A? If the degradation happens because of high voltage/currents, then the harder the CPU models were pushed from factory, and the more of that is allowed to be implemented by the BIOS settings, the worse the problem will become in the long run.

For now i would use the cautious approach and only use the performance profile. Which we mustn't forget, a few short years ago, was considered very extreme already, these are figures only overclockers would ever reach. Now these are reached with ease by a mainstream CPU like a 14700K, if let off the leash. And first, people need to have CPU cooling good enough to deal with 253W worth of heat. So this time they were serious when naming the "Extreme profile" that way.

Ideally, one should set power limits according to the individual cooling capabilities though. I explain that in my new thread:
How to set good power limits in the BIOS and reduce the CPU power draw.

I am quite happy that i have a somewhat frugal i5-13500, which is still based on Alder Lake architecture (so, it's really a modified 12th gen, not a true 13th gen), and which runs at sub-5 GHz turbo frequencies that are much more in the silicon's original comfort zone, at reasonable voltages. I was saying time and time again, like here, how this "arms race" has gotten out of hand on the Intel side, even AMD were pushing things a bit too hard sometimes, but not to this extent. This whole approach they did with the recent i9 models (and now even creeping down onto i7) has to end, and hopefully these problems will mark a watershed for Intel in their quest for higher performance. It is no longer viable to squeeze yet another 200 MHz out of the silicon for a new CPU model.

 

[FlyingScot]

Hey, the more info. the better!

Incidentally, we really don't know if we're dealing with different issues for the different SKUs, do we? Also, just how widespread is this potential degradation issue? Judging from the lack of responses to this thread after 500+ views, one might assume not that widespread. Hey, that would be good news for Intel.

 

[citay]

The thread is quite new, the issue is rather complex and so far really unsolved, not everyone has an i9, and the crashes don't immediately seem attributable to a degraded CPU (because those things just didn't really happen before this). So you see people posting in game forums for example, saying that the game might be buggy, things like that. But the bigger this issue gets and the more it gets reported about, the more users will connect the dots that it may have to do with their i9 (or even i7). The chances for Intel to come out of this unscathed don't look too good, unless they can really pull a rabbit out of the hat for this issue.

 

[FlyingScot]

The thread is quite new, the issue is rather complex and so far really unsolved, not everyone has an i9, and the crashes don't immediately seem attributable to a degraded CPU (because those things just didn't really happen before this). So you see people posting in game forums for example, saying that the game might be buggy, things like that. But the bigger this issue gets and the more it gets reported about, the more users will connect the dots that it may have to do with their i9 (or even i7). The chances for Intel to come out of this unscathed don't look too good, unless they can really pull a rabbit out of the hat for this issue.

Good point!

 

[FlyingScot]

I’ve seen this rumor before about a possible 15th Gen for LGA1700 (e.g. Z690/Z790) platform; and here it is again. I have an Alder Lake 12700KF, so the thought of extending the Z690/Z790 platform is an enticing one. Is it possible that Intel will do what they did with the 11th gen and backport a newer Core architecture - perhaps one that is more efficient, leading to an increase in P-Cores? I think a lot of current i9 users might welcome an upgrade path that doesn’t mean ditching their motherboard and potentially their memory, too. Of course, this is all just speculation at this point. But I will keep watching for any future developments on a possible 15th gen. Hey, I wonder if that‘s why Intel changed the naming convention for Arrow Lake, i.e. to allow squeezing in a 15th gen on the current platform? Just a thought…

Intel Planning P-core Only "Bartlett" LGA1700 Processor for 2025

In a surprising development, Intel plans to extend the longevity of its Socket LGA1700 platform even as the newer LGA1851 platform led by the Core Ultra 200 "Arrow Lake" remains on track for a late-Q3/early-Q4 2024 debut. This, according to a sensational leak by Jaykihn. It plans to do this with...

www.techpowerup.com

 

[citay]

I've heard this rumour too, but should it come, it likely won't be called 15th gen, it will be 1st gen "Core" / "Core Ultra" like Arrow Lake-S (first CPU models coming in October this year).

In news relating to the current instability, ASUS has released new BIOS updates. After updating, the new versions default to "Intel Default Settings" with the "Extreme" Profile, and the "Asus MultiCore Enhancement" now defaults to Disabled.

With a 14900K, it looks like this now on ASUS:

Alternatively, with manually selected "Performance" Profile (which would be my recommendation):

ASUS kept PL1 = PL2 = 253 W, ICCMax drops to 307A with the 14900K.

Loading the "Asus Advanced OC Profile", which removes most of the safeguards, now comes with a clear warning:

The changes with the "OC Profile" are as follows:

So, ASUS is completely playing by Intel's rules now, albeit opting for the presumably more risky "Extreme profile" instead of the "Performance profile" by default. Once their "OC Profile" is loaded, it reverts back to settings of old, maxing out most limits and disabling most protections. Weirdly enough, the 253W power limit seems to remain active (it's unclear wether this is an oversight or a deliberate choice).

 

[FlyingScot]

That's interesting that ASUS are taking the more conservative approach now. Maybe the days of wide open throttle settings by motherboard vendors are truly over. That would be a good thing for PC builders who are new to the hobby.

 

[citay]

So, by now we have official word from Intel, saying that "elevated operating voltage" is behind all this, which is "stemming from a microcode algorithm resulting in incorrect voltage requests to the processor".

https://www.reddit.com/r/intel/comments/1e9mf04

Another quote in regards to the oxidation rumours: "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."

And: "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."

There are a few questions left, such as:
What has taken Intel so long to find this out?
"Exposure to elevated voltages" is called a "key element" of the instability, meaning, not the only one, just one of the main ones?
If elevated voltages lead to CPU degradation, then bringing the voltage back down will surely just prevent further degradation, not fix existing one?

About the last question, they more or less confirmed that, if you still have instabilities with all the fixes applied (the crucial microcode update is supposed to come in August), then you should contact Intel for a potential RMA of the CPU.

There are videos like this, linked by the Intel guy in the reddit thread, which cover this latest development a bit:

Everyone (including the Intel guy from the thread) basically advises to update the BIOS to the latest (beta) version to have a minimum amount of protection, as it will apply Intel's recommendations from the blue table posted further up, which is already much more safe than the maxed out limits that the board vendors used to allow in a lot of their older BIOS versions.

As for checking wether an individual 13th/14th gen CPU is affected, this particular video linked above recommends to install the NVIDIA drivers 5-10 times or so, as it contains a bunch of decompression workload, which is supposedly good to reveal any instability in this case. I think this is not the last word on a good method. For example WinRAR has an integrated benchmark (ALT-B) that also checks for errors, presumably doing compression/decompression, maybe that is something that could also work to detect problems, i don't know.

12th gen is completely unaffected and doesn't require a microcode update, as there was apparently never a problem with its microcode, nor do those CPUs need as high of a voltage to begin with, as, say, a 14900K needs for its high clock speeds.

 

[FlyingScot]

CiTay, I haven't read your post yet but will be interested in your thoughts.

I posted my response to Intel’s ”announcement“ in this thread https://forum-en.msi.com/index.php?...wifi-undervolt-experience.399899/post-2271948 if anyone is interested. I think there are actionable items in that thread that relate to this whole instability issue. I doubt this new microcode update will fix things without some kind of action by the user, like undoing settings that had previously worked.

 

[citay]

I have a problem with the above announcement that I cannot resolve. If high voltages were an issue, why would the recent BIOS changes in connection with the Intel Baseline/Performance/Extreme advisory (whatever they called it) introduce a mechanism (i.e. the “Intel Defaults” CPU LL Mode) that would set AC_LL at the max setting (i.e. 1.1 mOhm)? Would this not lead to a massive increase in voltage versus the previous defaults of CPU LL = 9, 12, etc.? …especially under light workloads… I mean, had you not undervolted like crazy, where would your voltages be? And how many users would know they needed to manual undervolt or be proficient at it? Plus, with manual undervolting, the responsibility for a stable system falls entirely to the user.

Some users seem to experience this very thing, see this thread and especially the one i linked there (even though in the first thread we also have a different issue about the cooling performance). I can't be sure this is the new BIOS version's fault, but it's not a good look. It certainly seems like this issue is gonna keep everyone busy for quite a while.

 

[FlyingScot]

Some users seem to experience this very thing, see this thread and especially the one i linked there (even though in the first thread we also have a different issue about the cooling performance). I can't be sure this is the new BIOS version's fault, but it's not a good look. It certainly seems like this issue is gonna keep everyone busy for quite a while.

Yep. We're all seeing the same fallout, albeit from different starting points. In the UK, we would give this whole saga a very technical term: "A dog's breakfast!" - i.e. very messy... At this point, even if Intel does finally come up with a permanent fix for this situation, the damage to customer satisfaction (and loyalty) might also be permanent.

And to think that they are cheeky enough to suggest that Arrow Lake will be perfectly fine humming along at 105C. Madness, I tell you...madness!!!

 

[FlyingScot]

Someone else made the same observation that I did. Glad I’m not going mad…yet.
This user comment was posted against the video in CiTay's above post:

 

[RemusM]

At this point, even if Intel does finally come up with a permanent fix for this situation, the damage to customer satisfaction (and loyalty) might also be permanent.

A software/firmware change cannot be a "permanent fix" for a hardware issue (bad designed feature?).
It can be a mitigation in the best case scenario.
As I said before, these days there are not more overclockers because now Intel and AMD are the overclockers.
And there is always a price to pay when you push something to the limits.
The main problem here is that the customers are paying the price (and it's not their fault).

 

[FlyingScot]

A software/firmware change cannot be a "permanent fix" for a hardware issue (bad designed feature?).
It can be a mitigation in the best case scenario.
As I said before, these days there are not more overclockers because now Intel and AMD are the overclockers.
And there is always a price to pay when you push something to the limits.
The main problem here is that the customers are paying the price (and it's not their fault).

Hey, don't shoot the messenger! It was mainly just a play on words. I think we're all in agreement with your sentiments. Intel has been playing fast and loose and maybe greedy, too. And this whole situation was very predictable. Many of us saw it coming when the max temps (advised rather than specified) and voltages in Intel's specs suddenly jumped. Will they make the same mistakes with Arrow Lake? Time will tell. But you can count me out if tech companies continue to expect us to be their BETA testers.