[Intel] Sept 2024 (Update on Instability Reports on Intel Core 13th and 14th Gen Desktop Processors

[Silent_Scone]

September 25th Update:

Beta BIOS Links (microcode 0x12B)

Note: Some pages may not have propagated, please check back later

• ROG MAXIMUS Z790 HERO

• ROG MAXIMUS Z790 HERO EVA-02 EDITION

• ROG MAXIMUS Z790 DARK HERO

• ROG MAXIMUS Z790 APEX

• ROG MAXIMUS Z790 APEX ENCORE

• ROG MAXIMUS Z790 FORMULA

• ROG MAXIMUS Z790 EXTREME

• ROG MAXIMUS Z790 HERO BTF

• ROG STRIX Z790-E GAMING WIFI

• ROG STRIX Z790-E GAMING WIFI II

• ROG STRIX Z790-F GAMING WIFI

• ROG STRIX Z790-F GAMING WIFI II

• ROG STRIX Z790-A GAMING WIFI

• ROG STRIX Z790-A GAMING WIFI II

• ROG STRIX Z790-A GAMING WIFI D4

• ROG STRIX Z790-H GAMING WIFI

• ROG STRIX Z790-I GAMING WIFI

• PROART Z790-CREATOR
  
  Z690
  

• ROG MAXIMUS Z690 HERO
• ROG MAXIMUS Z690 HERO EVA EDITION
• ROG MAXIMUS Z690 APEX
• ROG MAXIMUS Z690 FORMULAROG MAXIMUS Z690 EXTREME
• ROG MAXIMUS Z690 EXTREME GLACIAL
• ROG STRIX Z690-I GAMING WIFI
• ROG STRIX Z690-A GAMING WIFI
• ROG STRIX Z690-A GAMING WIFI D4
• ROG STRIX Z690-G GAMING WIFI
• ROG STRIX Z690-F GAMING WIFI
• ROG STRIX Z690-E GAMING WIFI

Z760

• PROART B760-CREATOR
• PROART B760-CREATOR D4
• PROART B760-CREATOR WIFI
  
  Intel Community Forum
  Intel has localized the Vmin Shift Instability issue to a clock tree circuit within the IA core which is particularly vulnerable to reliability aging under elevated voltage and temperature. Intel has observed these conditions can lead to a duty cycle shift of the clocks and observed system instability.  

Intel has identified four (4) operating scenarios that can lead to Vmin shift in affected processors:

1. Motherboard power delivery settings exceeding Intel power guidance. 
   a.  Mitigation: Intel Default Settings recommendations for Intel Core 13th and 14th Gen desktop processors.  
2. eTVB Microcode algorithm which was allowing Intel Core 13th and 14th Gen i9 desktop processors to operate at higher performance states even at high temperatures. 
   a.  Mitigation: microcode 0x125 (June 2024) addresses eTVB algorithm issue.  
3. Microcode SVID algorithm requesting high voltages at a frequency and duration which can cause Vmin shift. 
   a.  Mitigation: microcode 0x129 (August 2024) addresses high voltages requested by the processor.  
4. Microcode and BIOS code requesting elevated core voltages which can cause Vmin shift especially during periods of idle and/or light activity. 
   a.  Mitigation: Intel is releasing microcode 0x12B, which encompasses 0x125 and 0x129 microcode updates, and addresses elevated voltage requests by the processor during idle and/or light activity periods.  

Regarding the 0x12B update, Intel is working with its partners to roll out the relevant BIOS update to the public.

Intel’s internal testing comparing 0x12B microcode to 0x125 microcode – on Intel® Core™ i9-14900K with DDR5 5200MT/s memory1  - indicates performance impact is within run-to-run variation (ie. Cinebench* R23, Speedometer*, WebXPRT4*, Crossmark*). For gaming workloads on Intel® Core™ i9-14900K with DDR5 5600MT/s memory2, performance is also within run-to-run variation (ie. Shadow of the Tomb Raider*, Cyberpunk* 2077, Hitman 3: Dartmoor*, Total War: Warhammer III – Mirrors of Madness*). However, system performance is dependent on configuration and several other factors.

Intel® reaffirms that both Intel® Core™ 13th and 14th Gen mobile processors and future client product families – including the codename Lunar Lake and Arrow Lake families - are unaffected by the Vmin Shift Instability issue. We appreciate our customers’ patience throughout the investigation, as well as our partners’ support in the analysis and relevant mitigations. 

______________________________

August 2024

Hi all,

This update will be pinned here for clarity on the ongoing situation with some 13th and 14th-gen CPUs experiencing instability exhibited at stock.

Intel has issued the following statement (07/22/2024)

Based on extensive analysis of Intel Core 13th/14th Gen desktop processors returned to us due to instability issues, we have determined that elevated operating voltage is causing instability issues in some 13th/14th Gen desktop processors. Our analysis of returned processors confirms that the elevated operating voltage is stemming from a microcode algorithm resulting in incorrect voltage requests to the processor.

Intel is delivering a microcode patch which addresses the root cause of exposure to elevated voltages. We are continuing validation to ensure that scenarios of instability reported to Intel regarding its Core 13th/14th Gen desktop processors are addressed. Intel is currently targeting mid-August for patch release to partners following full validation.

Intel is committed to making this right with our customers, and we continue asking any customers currently experiencing instability issues on their Intel Core 13th/14th Gen desktop processors reach out to Intel Customer Support for further assistance.

https://community.intel.com/t5/Processors/July-2024-Update-on-Instability-Reports-on-Intel-Core-13th...

For CPUs that exhibit unstable behaviour

Ensure you're on the latest BIOS for your motherboard

Update your motherboard BIOS to the latest revision containing microcode 0x125. Whilst this ucode revision does not resolve the issue, Intel has clarified that it contains fixes within microcode pertinent to Enhanced Thermal Velocity Boost (ETVB) behaviour that results in unexpected boost behaviour under certain thermal conditions.

Use Intel Base Profiles
The latest BIOS for your motherboard should contain Performance Preference options for Intel Base profiles. Ensure Intel Default Settings is selected. If using a 13th-Gen or 14th Gen i9 processor, choose Profile Performance.
If not selected already, ensure that SVID Behaviour is set to Intel Fail Safe

My CPU is running hotter using Intel Base Profiles than Asus Optimised, is this normal?
Yes, it can be normal for your CPU to run hotter than usual when using Intel base profiles. This occurs because following Intel's guidelines more closely often results in higher core voltages under certain workloads. Previously, motherboard vendors set the Load-Line Calibration (LLC) lower than Intel's recommendations to achieve lower temperatures while still delivering high performance. However, depending on your motherboard adhering to Intel's guidelines can lead to increased voltages whilst operating within Intel's current limits. This can result in higher operating temperatures depending on the workload.

Disable XMP

XMP is considered overclocking. To eliminate overclocking instability XMP should be disabled if experiencing crashing or unstable behaviour. Be sure to retest the behaviour at DRAM defaults.

Evaluate Cooling

Ensure your CPU cooling is working within the expected realms. If still experiencing instability under the conditions described above, consider reapplying the thermal application and removing any third-party CPU contact frames which may impact signal integrity.

Is my CPU damaged?

If still experiencing instability at Optimised Defaults [Intel - Performance / Extreme Profile] with XMP overclocking disabled, you will need to contact Intel.

I've read there was a fabrication issue with oxidisation, Intel has confirmed it.
This issue is unrelated to the ETVB behaviour and firmware and is a physical defect. Currently,  Intel has gone on record to state that a number of 13th-generation CPUs from 2023 are potentially impacted by this issue

ASUS FAQ

[Vatergascoigne]

First of all, thank you very much for testing!
After a restart and a few test runs, I no longer get any errors.
So it seems to have been a false alarm.
🙂

[dudifergus456]

I am posting a 7-day review of Bios 1801. After a week of gaming and browsing I feel that the temperature has gone a bit down during game play and the fans go noticeably slower after exiting CS2. The overall experience is a good one temps are down and fans are spinning slower. I have not experienced any instability at all, and the pc is running cooler than before   

[Strife21]

Well I just built a Asus Z790-H Strix Gaming WIfi and I5-13400f computer for my best friends kid.   The board and processor were purchased a while back as he got the different components over the last year or so for his birthday and christmas.

The i5-13400f used has the B0 stepping so its actually a raptor lake variety and not alder lake thus could be susceptible to this stability issue.   The board had an older bios on it, one that defaulted the Asus Multi-Core Enhancement to "Auto- Lets Bios Optimize" instead of "disabled - enforce all limits".    

That being said I never left the bios because this computer was brand new so no operating system was installed or any functions that would stress the cpu.  Before doing anything else I updated the BIOS to the latest version from Asus's site and restored the factory defaults using "F5".  This set the intel profile and then defaulted the MCE to "disabled - enforce all limits".

My question is was any damage done to the CPU potentially from the Intel stability issue by booting to the bios with the older bios version that came on the board, since it didn't have latest processor microcode update.  I would hope that simply going into the bios, wouldn't push the processor too hard.  Is there any chance of degradation occurring in this circumstance?

[Daytrader]

I am 99.9% sure zero damage was done, just for reference, i have the Asus Z790-H Strix Gaming WIfi also, but with the dreaded 13900K, i was using the board and cpu with the very old bios for months and months before updating, and that was gaming and stress testing alot, and i still have no problems.

[johnhellstar]

Hello guys! 1801 bios for Apex Encore has some very strange behavior when i lock the cores to let's say 56p-45e the system is extremely unstable,cinebeches crashes like hell..(this is not happening with 1102 bios i used for months)...I spent many hours tinkering the bios and i find that to make it stable i have to enable sync acdc to vrm loadline,ac and dc to 0.01ohms and load line to level to 5....After that i have to set manual cpu voltages in my case 1.26v cpu is ok.Otherwise on auto all hell breaks loose 😅

Despite that i found another problem...What ever i do to disable cpu downclocking to 3.2ghz when idle doesn't work... especially the ecores...disabling all states,shifts and whatnots won't help 🤔 i found another thread at OCN with the same downclocking problem on the Apex Encore.... Strange 🤔 i'm digging it more,i will try to reflash the bios again and star over.

[Taint3dBulge]

Been running 1503 on my Apex Encore since it came out. Just noticed today that my Ecores are downclocking from 4400 to 4300 even during gaming with performance power mode on, have C1E turned off and Speedstep turned off. Any ideas why this is happening? I have the cores locked at 57 and 44.   LLC7 with voltage on auto TVB is off and ACLL set to .20 DCLL set to .24

[Taint3dBulge]

Heh, I dont think they will ever be able to fix the instability unless they just have the clocks turned down to 5.5ghz max boost.

[inge70]

If the instabilities and degradations were and are already present, NOTHING can be fixed.

The only solution is an RMA and CPU replacement.
19 processors of the 13th and 14th generations were most affected.
The clock speed limitation that Intel had partially recommended only affected already degraded/damaged CPUs so that they could continue to be used for the time being.

However, this notification only came in mid-2024, when the actual cause was still largely unknown.

However, due to public pressure, Intel had already more or less confirmed everything in 2024 and subsequently extended the warranty periods for boxed CPUs to 5 years.

The microcode updates, which address the partial causes and now likely the actual cause, are intended to prevent unaffected 13th and 14th generation CPUs from being largely spared from this defect as long as they are operated within Intel's specified specifications.
However, anyone who absolutely wants to overclock, i.e., goes beyond the specs, will still risk damaging their CPU in the long run, because no manufacturer guarantees unrestricted overclocking.

Overclocking always carries a certain residual risk for the CPU, no matter how well it is cooled.

But this should be clear to all overclockers.

[Vynra]

The issue now though is that intel (before) had basically said in multiple interviews and videos for 13th gen that you should lift power limits to what your cooling is capable of. They wanted you to get more performance if you had the cooling. Basically if you were under 100c then you were missing out on performance. As long as you didnt alter clockspeeds manually the cpu still controlled clock speeds and had safety mechanisms...which technically was not an overclock. Just unlocked power. Hell they even said their own rules were just "guidelines"

Then they completely backtracked that statement and said that anyone who is not adhering to the power limits is overclocking. 

The thing is, if i am on the asus oc profile, but adhere to intels power guidelines guidelines, am i overclocking? 

There is absolutely no clarity, as intel has also said that the Intel baseline profiles are for CPUs that showed instability.

[inge70]

You're right that Intel initially publicly stated that the 13th and 14th generation parts in "K", "KF", "KS", etc., had full overclocking potential.
Ultimately, ALL manufacturers always talk a lot to keep or maintain their performance crown, even though their hardware/CPUs are already running at their limits out of the box, and then not much happens with overclocking without causing damage because the things were hastily put together.
We saw that in the Intel chaos.

For me, however, it's now also a fact that the Asus OC profile, when activated, also means proper overclocking and no longer adheres to Intel specifications.
However, the Asus OC profile still handles overclocking cautiously and sets all important parameters right away, thus allowing fewer errors to occur due to manual user input.
But even this profile ultimately represented overclocking and was no longer in Intel's best interest, even though Intel did force manufacturers to open the UEFIs.
Until Intel had to backtrack.
Because only the later introduced Intel Performance Preferences profiles:
Extreme (only applies to I9 "K", "KF", and "KS" CPUs)
Performance (for the remaining "K" CPUs)
were apparently in accordance with Intel's SD specifications.
At least Intel released the relevant data, etc.

The Asus OC profile, however, was not applicable, as it already set all PL1 and PL2 to unlimited.
If you wanted to use the Asus OC profile after the Intel profiles were released (I think it was mid-2024), there was a warning message in the UEFI that pointed out precisely the circumstance of overclocking and the potential damage to the CPU that could result.
That wasn't the case at first.
I "only" use a 13700K and had the "Asus OC Profile" active for three months shortly after purchasing my hardware (October 2023).
It was quite nice, but I didn't see any added value, except for benchmarks. So I went back to the defaults, and it was great.
Fortunately, there were no visible side effects, which I noticed with Unreal Engine 5 games at the latest when the CPU was freaking out, as the shader compilation crashed.
When the Intel profiles with the "Performance Preferences" switch came along, there were only "Performance" and "Asus OC Profile" for my CPU, although I've had and still have everything set to Intel's "Performance" profile and "Intel's Standard" since then (e.g., this sets PL1 and PL2 to 253W, etc.).
I never had the baseline profile active when it was still in the UEFI, as there were conflicting reports and it probably wasn't the best solution.
I ran various tests several times last year that might indicate a damaged CPU. Most notably, a game with Unreal Engine 5, because it was with this engine that the whole mess first became truly apparent and publicized at the beginning of last year.
Unfortunately, Intel never developed and released a test tool to clearly diagnose damage. If it had even been possible, given all the partial problems with these two Intel series, then who knows what else might have come to light.

Ultimately, Intel didn't do itself any favors with all this and deserved the rewards accordingly.

In my opinion, however, the Asus OC profile is an overclock outside of the specifications, as it deactivates various Intel protection mechanisms (I believe this affected TVB, eTVB, SA CEP, etc.) and sets the power limit to unlimited.
If you then change something here back to Intel standards, you will no longer be fully using the Asus OC profile as intended, as you will then be overriding it again.
Ultimately, however, you are still overclocking with the active "Asus OC Profile" and should receive a corresponding warning message when activated in the UEFI, which has been appearing since at least June/July 2024.

I believe the baseline profiles were briefly introduced in May/June 2024 to allow the continued use of already damaged CPUs. However, this also increased the power supply, which in turn was not good for the CPUs.
The i9 processors in particular suffered from this.
This topic could be discussed indefinitely, but unfortunately, it doesn't provide any new insights. If you have a defective CPU, you can replace it via RMA, and that's it.
Due to the recent chaos and the fact that there won't be a new, bug-fixed revision, the 13th and 14th generations can't be recommended without a doubt.