Question BSOD loop caused by CPU

[LegendArtorias]

The past week I have been plagued by constant BSOD's when attempting to boot my computer. I couldn't even make it to my desktop without the thing crashing 90% of the time. This is on a newer computer that I built a month ago, but the thing has been working fine until now. Even when attempting to install windows (fresh, with no OS installed) I was getting BSOD's. I couldn't figure it out for the life of me and I had no idea what the cause suddenly was.

Well, just last night I attempted to disable 3/8 of my P-Cores (Intel i9-13900K) as a last-ditch attempt to find the culprit and what do you know, the thing can boot up and run reliably with no BSOD so far. So does this mean that I have a faulty CPU? Or is something suddenly causing this that I may be able to fix without RMA? I have already tried to re-seat the CPU and it appears to be fine with no bent pins on my motherboard.

The error codes I am getting are:
kmode exception not handled
system thread exception not handled
page fault in nonpaged area
unexpected kernel mode trap
irql_not_less_or_equal
system service exception
registry driver filter exception
critical process died

Along with these error codes are things listed that have failed such as:
ntfs.sys failed
wdf01000.sys failed
fltmgr.sys failed
win32kfull.sys failed

Here is a link to my pc parts: https://pcpartpicker.com/list/qmZsqm

 

[Misgar]

Asus released a BIOS update yesterday for some motherboards (August 4) which improved memory stability. If you are overclocking your RAM with XMP/EXPO/DOCP, disable this setting in the BIOS for the time being and revert to the default (much slower) JEDEC setting. 6400MT/s may be too fast for your current BIOS. Try 4,000 to 4,800MT/s and see if the system stabilises.

Although the NH-U12A is classified by Noctua as being adequate for the 13900K, I would have picked the more powerful NH-D15 (my favourite), or if you want to extract the last ounce of performance, a 360 or 420mm AIO. The 13900K will throttle earlier on the U12A which has an NSPR rating of only 169. The D15 is rated at 183, but even that isn't enough to extract the most out of a 13900K running flat out.
https://ncc.noctua.at/cpus/model/Core-i9-13900K-1642

 

[LegendArtorias]

Asus released a BIOS update yesterday for some motherboards (August 4) which improved memory stability. If you are overclocking your RAM with XMP/EXPO/DOCP, disable this setting in the BIOS for the time being and revert to the default (much slower) JEDEC setting. 6400MT/s may be too fast for your current BIOS. Try 4,000 to 4,800MT/s and see if the system stabilises.

Although the NH-U12A is classified by Noctua as being adequate for the 13900K, I would have picked the more powerful NH-D15 (my favourite), or if you want to extract the last ounce of performance, a 360 or 420mm AIO. The 13900K will throttle earlier on the U12A which has an NSPR rating of only 169. The D15 is rated at 183, but even that isn't enough to extract the most out of a 13900K running flat out.
https://ncc.noctua.at/cpus/model/Core-i9-13900K-1642

Sadly that BIOS does not pertain to my motherboard, I am up to date with the latest BIOS version. I am not using any RAM overclocking either, I have not touched any BIOS settings prior to this happening. Temps do not seem to be causing this either as the temps look quite stable in BIOS and it is not even able to boot long enough to raise those temps.

 

[Misgar]

Have you tried running MemTest86+ and booted up the PC from a USB memory stick? It might be worth a try to see if the problem is memory-related.

Is your RAM included in the motherboard QVL?
Intel processors are usually pretty tolerant of unlisted memory, but you may just be unlucky with your selection, or the RAM might actually be faulty.

When you disabled three P-cores, did you pick them at random? You could try picking a different set of three P-cores and see if anything changes. It might help if you need to RMA the CPU.

 

[LegendArtorias]

Have you tried running MemTest86+ and booted up the PC from a USB memory stick? It might be worth a try to see if the problem is memory-related.

Is your RAM included in the motherboard QVL?
Intel processors are usually pretty tolerant of unlisted memory, but you may just be unlucky with your selection, or the RAM might actually be faulty.

When you disabled three P-cores, did you pick them at random? You could try picking a different set of three P-cores and see if anything changes. It might help if you need to RMA the CPU.

I have run MemTest86 and everything checks out. I even ran the longer, more extensive test. It is unable to boot from a USB with all P-Cores active. It is also unable to boot even with a single stick of RAM being used, cycling between the 2. There are even BSOD issues when installing Windows fresh without an OS installed.

I did not pick them at random as it seems I cannot just select and isolate certain cores. I am only able to choose to disable 1-8 cores and I cannot see which ones are disabled.

 

[smallhan]

The past week I have been plagued by constant BSOD's when attempting to boot my computer. I couldn't even make it to my desktop without the thing crashing 90% of the time. This is on a newer computer that I built a month ago, but the thing has been working fine until now. Even when attempting to install windows (fresh, with no OS installed) I was getting BSOD's. I couldn't figure it out for the life of me and I had no idea what the cause suddenly was.

Well, just last night I attempted to disable 3/8 of my P-Cores (Intel i9-13900K) as a last-ditch attempt to find the culprit and what do you know, the thing can boot up and run reliably with no BSOD so far. So does this mean that I have a faulty CPU? Or is something suddenly causing this that I may be able to fix without RMA? I have already tried to re-seat the CPU and it appears to be fine with no bent pins on my motherboard.

The error codes I am getting are:
kmode exception not handled
system thread exception not handled
page fault in nonpaged area
unexpected kernel mode trap
irql_not_less_or_equal
system service exception
registry driver filter exception
critical process died

Along with these error codes are things listed that have failed such as:
ntfs.sys failed
wdf01000.sys failed
fltmgr.sys failed
win32kfull.sys failed

Here is a link to my pc parts: https://pcpartpicker.com/list/qmZsqm

What a deja vu! Just been through the same issues with you with nearly the same parts. Wonder if you'd like to share your latest updates?

I have I9-13900 (Not K), 280mm AIO, Asus Strix Z790H, Zskill 16GBX2, 850W PSU, 4070 and some NVMEs. We hooked up everything and installed Windows10 and it went quite smoothly, it was around 10PM. Then I installed drivers, updates for Windows 10 and some programs and transferred files from my old computer with ethernet, then around 11PM, I saw my first BSOD. Then it just got more and more frequent, I remember the first three BSOD was 11, 11:30, 11:35, then not soon after that, I could not even enter OS and went into a perfect loop. And moreover, it could not even boot from the USB installation drive. At around 1AM, I decided to make it run MEM test in Bios and it ran flawlessly through the whole night, then I pretty much narrowed down the issues to either CPU or MB.
Today, I got a spare i3 CPU and it worked perfectly, and concluded either the CPU itself is defective, or the CPU+MB combo is bad, perhaps that MB just does not like i9.
Just I could not understand is how the CPU core could degrade that rapidly just within hours?? I have no idea.

 

[Misgar]

Just to clear up a few things, are you running your memory at stock JEDEC speeds around 4200MT/s when the BSOD occurs or are you running an XMP overclock? Also, what power supply are you running (exact make and model number)? Is it a Tier-A unit?

It's still not clear to me if this is a BIOS, CPU or PSU problem.

 

[opa9696]

To anyone that has this problem: Disable Intel Turbo Boost in BIOS and you should be able to use your PC.