I9-13900K, Cannot get my system stable. CPU most likely the cause.

[K03212]

Please help!

I bought 2 I9-13900K and build the same computer twice. Both can not run stock speeds without crashing. Yes, XMP is running, but no, it will not influence the crashes since the CPU manages to crash completely without any XMP active or overclocking. It seems to crash randomly under loads. Sometimes when gaming, sometimes transferring data to another folder crashes it. Crashes usually get the sound and picture stuck until the PC reboots itself. Usually no Bluescreen is shown. Bluescreen logs indicated an issue with the CPU. I did several Memtests. The memory seems to be in pristine coniditon. No issues found after several tests (OCCT, Memtest86).

I currently run the system half stable (crashes every 15 minutes, sometimes runs up to 3h) while I deactivated all the boost mechanics in my BIOS which I could find.

The event logger logs critical hardware failures (of certain cores) and the only way to avoid them seems to underclock the CPU heavily on less than 4 Ghz per Performance Core.

I am going to be honest; I am not well versed with all these BIOS Settings and I never touched OCing since I did not see the benefit in doing so for myself so I never touched any voltage settings as well. I do not know if these issues resolve around the CPU getting too much or too little voltage and all I really know is that this **bleep** thing will not run the XMP profile on the full advertised potential nor does the CPU can run its own stock clock speeds on the default BIOS settings (with XMP being off!).  

For some strange reason it worked just fine on both PCs for months on an underclocked RAM speed to get XMP stable, then one had issues and weeks after that the other got the same. I first thought this might is some OS issue and properly installed Windows again but the crashes remained. It is truly some sort of hardware related issue and stuff like OCCT reports many CPU related crashes during torture tests on both CPUs when running stock speeds.

Anything thermal related can be excluded, since I monitor the heat live with warning messages while caping the max temp of the CPU on 95°C. But since I got a strong watercooler and proper Case airflow the CPU usually runs under 60°C on very heavy loads (unless it is a long-term stress test of course).

I am grateful for any help or advices I can get. I am not just gaming on my PC, I also work on it which turns this into an absolute nightmare of an issue since the only thing I can think of is to make use of the guarantee since both CPUs seem to have some kind of hardware issue. But how is that possible? They ran cool on optimized default BIOS settings without any OCing?

Specs
Gigabyte Z790 AERO G
RTX 4080
I9-13900K (32 CPUs)
Memory: 32768MB RAM@6000Mhz 
2 TB M.2 Storage
Windows 11

Let me know if you want to know anything else. 

[AlHill]

And, your motherboard model number is?

Doc (not an Intel employee or contractor)
[Maybe Windows 12 will be better]

[K03212]

My motherboard is an Gigabyte Z790 AERO G. 

[IsaacQ_Intel]

Hello @K03212

Thank you for posting on the Intel️® communities. We would be more than glad to help you with your issue.

In order to have a better understanding of your issue, please provide me with the following:

SSU: Please download and scan your system using our tool, Intel® System Support Utility so we can help check your system info first. Once finished downloading, open the tool, click on scan 'Everything', click next, then save the system information to a .txt file. Please provide the .txt file.

This document will be necessary to check your entire system, to find possible errors or incompatibilities.

As part of our initial troubleshooting process, we recommend the following actions:

a. Ensure that your system's BIOS is up to date. It is important to note that issues pertaining to compatibility with your CPU may often be rooted in the motherboard. As such, updating the BIOS to the latest version is essential, as newer releases may resolve compatibility concerns.

b. Reset your BIOS settings to their default values. This step is intended to eliminate any instability resulting from inadvertent configuration changes.

c. Inspect all hardware components, including the motherboard, power supply, and RAM, for potential faults or loose connections. A detailed guide on this matter is available in the article titled Scenarios for Using a Processor Swap Test to Help Isolate an Issue.

d. Ensure that your operating system is both up to date and functioning correctly. It is important to recognize that software-related issues can sometimes manifest as hardware instability.

If, after implementing these recommendations, your issue persists, we advise you to consult your motherboard's original equipment manufacturer (OEM) to validate the proper functioning of your motherboard. Alternatively, if you wish to initiate a return merchandise authorization (RMA) with us, please do not hesitate to inform us of your decision.

Additionally, we encourage you to peruse the following articles, as they may contain valuable insights into resolving your concern:

·      Random System Crashes on an Intel® Boxed Processor.

·      Tips for Random System Freezing Issue.

Thank you for your patience and cooperation in addressing this matter.

Best regards,

Isaac Q. 

Intel Customer Support Technician

[K03212]

Hello Isaac, I am so glad someone responded!

I am desperate by now. Nothing seems to work and the system seems to be very unstable unless I extremely underclock the CPU. By now I am very certain that the CPU is somehow faulty. Crashes happen randomly, but much quicker under load. Even with XMP off nothing is stable and when the CPU is on default settings it crashes very fast under load. Only downclocking the speed helps.  TXT file is attached. Spent days by now, trying to find out whats wrong. 

But please keep in mind most crashes are not visible anymore since I reinstalled windows to exlcude potential software errors. I had plenty of crashes before reinstalled windows and my event logger claimed that the source been several of my CPU cores which had critical errors. It first began with hardware errors that the computer was able to compensate (it did not crash the PC back then, just single applications), then it turned into crashing the whole computer at some point with bluescreens. By now the whole computer just freezes without bluescreens and after 1-2 minutes it usually boots back up. So, there are no detailed crash logs anymore. The crash logs usually reported this for several APIC IDs when the computer was still able to log things:

Event ID 19

Fixed hardware error.

Reported by component: processor core

Source of error: Machine check corrected

Error Type: Internal parity error

Processor APIC ID: 32

The detailed view of this entry contains further information.

Regarding the suggestions:
a. Already using newest BIOS.
b. Done so several times. In fact, this will increase the frequency of the crashes and crashes the computer under heavy load. Downclocking helps.
c. I build two identical computers which began to have the same issues over time. Both been running mostly well for months and there was no physical tampering done during that time. The one that got used more started to have these issues earlier but both CPUs did stop to run stable on the default BIOS settings at some point. They are in closed proper cases and I knew what I did when I build them, nothing is lose, everything is well cooled. 
d. It is up to date and was before. I reinstalled windows to make sure of that. Tried everything with only the most basic software on it. Nothing seems to fix it.

Since I have these issues on both identical motherboards it is most likely safe to assume that it is not the motherboard. And it would be unlikely that both mainboards became more faulty over time.

An RMA for a refund would be great and helpful! The I9-13900K was a disappointment for me to be honest. But I am an Intel costumer for over two decades now. One flawed release will not make me turn to the competition since I bought plenty of Intel CPUs and never had these issues before. The I9-14900K is using the same socket isn´t it? And people have not reported the same issues with it compared to the I9-13900K I could just buy the 14900K after the refund. 

Since I work with one of these computers this would be a better solution than having to face this mess again after a few months due to it not working for long with default BIOS settings. 

But I never had an RMA case before. How would that work?

[fredde23]

My cpu crashed also completely random underload and i found a fix by lowering down the active core tuning by 1 unit

[K03212]

Sadly, that will not fix my issue. I already tried to run every performance core on 5 Ghz already. Still crashed on both systems...

[IsaacQ_Intel]

Hello @K03212,

Thank you for your reply. We regret to learn that the problem persists despite having followed the recommendations and troubleshooting steps outlined in the previous message.

This time we would like to recommend you process the RMA with us, after verifying that your warranty is still active here, directly contact Intel Customer Support to initiate the RMA process (you cannot do this through the forums). Here are pages where you can look up contact information, including local/country phone numbers, by geography:

U.S. and Canada: Intel Customer Support

Europe, Middle East, and Africa: Intel Customer Support EMEA 

Asia-Pacific: Intel Customer Support APAC 

Latin America: Intel Customer Support LAR  

 As the discussion is currently within the purview of our live support services, we will proceed to formally conclude this conversation. Kindly note that this thread will no longer be subject to ongoing monitoring by Intel. We appreciate your comprehension in this matter.

Best regards, 

Isaac Q. 

Intel Customer Support Technician