i9-13900K : very frequent crashes (Windows 11) with apps, games and benches

[LoloWiwi]

Hi,

I built a setup in April 2023 with :

- Intel Core i9-13900K

- Asus ROG Maximus Z790 Hero

- 2x32 GB Corsair Vengeance 6600 MHz

- Asus ROG 4090

- Asus ROG Thor 850 Watts

From the very beginning, I had sometimes BSODs, and several apps/games crashing very "reliably".

Even though the PC: 

- Not overclocked (no XMP, so RAM is running at 4800 MHz)

- No Tweaks of any sort in the Bios / default values from Asus Bios.

- Windows 11 Pro 10.0.22621

- Windows / Drivers / Bios are up to date with latest versions as of today.

The tests :

- Prime95 : with smallest and small FFT (to only test CPU and CPU cache) -> gives FATAL ERROR (prime numbers errors) on some CPU cores after a few minutes. 

- Cinebench R23 in single core : no problem, no crash during the 10min run

- Cinebench R23 in multi-core : crashes after 2 to 30 seconds systematically.

- GPU tests are fine, they complete with no crash (Furmark)

- Memtest86 : did several runs on the mem at 4800 and 6600 -> no errors, all tests PASS.

- a few games such as Cyberpunk 2077, Horizon Zero Dawn : almost systematically crash when launched.

A couple of days ago, I realized that it's probably the CPU: 

- I use an app for 3D Printing called a "slicer" who prepares the file for 3D printing that would consistently crash on my Windows 11 setup during slicing (after 5 to 10 seconds max), but not on a virtual machine installed with VirtualBox (Windows 10)

- Somebody advised me to try to set the Affinity for the CPU Cores/Threads in Windows 11.

- Also, I found a lot of reports in forums/reddit about problems with i9-13900K...

Since then, when I set the Affinity of the apps for only a few cores:

- Bambu Studio slicing fine, no crash whatsoever if affinity set to 8 first (CPU0 to CPU7)

- Cyberpunk 2077, Horizon Zero Dawn : they both run fine when affinity set to 8 first (CPU0 to CPU7), but crash as soon as I change affinity back to all cores/threads.

- Cinebench multi-core : systematically crashes no matter what subset of cores/threads I set.

Weirdly enough, I tried the Intel Processor Diagnostic Tool: it always passes, but I don't trust its results, since I've so many other apps/games crashing, and reports by people on the web...

So, I need help please with that, I can't trust my CPU even though I need my PC for work every day...

Intel SSU report attached.

[Alberto_R_Intel]

Hi LoloWiwi, You are very welcome, thank you very much for sharing those updates.

Perfect, excellent, it is great to hear that you were able to get in contact directly with your local Intel® Support department in order to claim the warranty of your processor and replace the unit.

We are glad to know that the information provided about the SVID Behavior = Intel Fail Safe was useful and helpful for you as well.

We apologize for any inconvenience.

Any other inquiries, do not hesitate to contact us again.

Regards,
Albert R.

Intel Customer Support Technician

[sibidharan]

Intel agreed RMA for my case.. Individual cores gone faulty in i9-13900k, throwing general protection fault - Intel Community 

I hope you get it too  

[LoloWiwi]

Hi @sibidharan,

Great to hear you finally got the approval for RMA

Same for me, my local Intel Support finally offered to RMA my CPU!

What a relief, after so many hours lost in the process of finding out the problem was my CPU.

[sibidharan]

Congratulations, happy to hear bro

[Geek_sam]

I got a 14900K for 2 weeks now, touchwood not having the as severe problem as you guys, but as I first get it with my old Gigabyte Z690 UD AX board it went unstable in flight sim and occasionally crash to desktop during starfield also, the ram was Gskill 32GB x2 DDR5 6000 which was rock stable with the same board and with the old 12700KF, after some trial it keeps failing one hour OCCT ram test I was wondering if it was the IMC being bad or the Gskill is dying (unlikely), after a few days of trouble shooting it ended up being one of the ram stick might be slightly loose, as I pulled out both sticks, swarp them and press them in hard, bam, all problems gone, occasional <1s studder still happens in MSFS though, but I wonder if that's the core scheduller or something was kicking in, no more BSOD/CTD, what I concern is at "Normal" core voltage, intel LLC and intel POR limit during heavy gaming, all P cores go to 5.7Ghz at 1.40-1.45v, I wonders if that will burn out the cpu quick

[LoloWiwi]

Hi everybody,

My i9-13900K is in the RMA process (I had to provide pictures of all CPU angles, ATPO and FPO numbers). I'm waiting for Intel confirmation to send it back for replacement.

In the interim, I got a 14th gen Intel Core i7-14900K, and I triple confirm the problem was my i9-13900K, because with the i7-14700K :

- Cinebench R23 runs its multi-core stress test like a charm, no crashes

- My slicing softwares work perfectly now, no more need to set affinity to UC0 to UC7 for them not to crash during slicing

- Prime95 2 hours stress test for Small FTT (to test CPU and Cache) PASSES, as opposed to my i9-13900K that used to crash systematically on some threads

Funny enough, Cinebench R23 score is better with the i7-14700K than with my i9-13900K :

- i7-14700K score : 33891 (all params set to Asus default in Bios)

- i9-13900K score : 32597 (BIOS SVID Behavior=Intel Fail Safe and MultiCore Enhancement = disabled)

 

I'm soooo disappointed with my i9-13900K purchase...

[KCLam]

Hi, 

May I know about what are your RMA procedures? Like if they have guaranteed refund/exchange or they just said you have to wait for the testing results.

I have contacted with my local distributor for RMA, and they asked me to send back the CPU to them and it would take 6-8 weeks for troubleshooting. I would totally be fine with buying a interim CPU, but it would be annoying if they still could not troubleshoot after that period, I am quite confused what to do next.

[n_scott_pearson]

Here's my standard response to this type of query:

After verifying that your warranty is still active (here: Determine if Your Product Is Still Under Warranty), you need to directly contact Intel Customer Support to initiate the RMA process (you cannot do this here). You can start here: Contact Intel Support or use this page: Intel Support Phone Directory to look up local/country phone numbers.

You can use the online forms service or email to initiate this process, but I recommend calling Intel directly. The above links provide access to local or toll-free numbers for most countries and these numbers provide support for most local languages. You could also use the chat service (start here: Intel Customer Support Chat), but understand that this service is offered only in English and is subject to the normal (9:00-5:00 M-F) office hours in the Pacific (UTC-8) time zone.

Hope this helps,

...S

[KCLam]

Hi Scott,

Thanks for the useful reply. I initiated the RMA via phone call yesterday, it has been accepted today and my 13900k has been picked up by Fedex, my part is done.

I got a laptop to use in the mean time, so I decided to wait for some days before they informed me about whether I could get a replacement or I could opt to be reimbushed, it would be annoying if I buy a new CPU to use now and get a replacement 13900k later, I will have no clues what to do with two CPUs then.

Anyway, I am happy with the efficiency of the customer service. Hopefully I don't get another faulty 13900k and go through all this stuff again after two weeks.

KC

[LoloWiwi]

Hi,

Just a quick update : DHL picked up my package today for return to the Intel facility in the Netherlands.

As the i9-13900K is out of stock at the moment and I'd have to wait 4 weeks for them to restock, they offered a reimbursement.

I'm fine with this, I was not happy with the i9-13900K, and would have been afraid to have to do another round of investigation if the new i9-13900K had similar problems 

(I saw some posts on Reddit where people had the same problems with 2 different i9-13900K...).

Bye-bye, so long i9-13900K

I won't get my hours/days/weeks/months back of pulling my hair, but at least a solution has been found.

Now, I don't know if I should go for the i9-14900K or if I should keep the temporary i7-14700K I got in the interim.
I'm afraid that the Intel's "formula 1" CPUs are not that reliable...

[LoloWiwi]

Just a quick update : I got fully reimbursed by Intel. It was quick: about a week from sending the faulty i9-13900K to getting the reimbursement.

I was afraid that it would be a long process with a lot of hoops, but no.
Even though I was very disappointed with my CPU, at least the RMA process was quick and flawless!

[Banksy]

I think I might be in the same boat as you with this. I have an Asus Rog Maximus z790 Apex and have had so much trouble with this Intel i9-13900kf. Can't get it to boot even on optimised defaults of various bios revisions. I have had it running okish for a couple months with Intel Fail-safe SVID behaviour which seemed to allow it to run though at higher voltages and temps. I've been playing Star Citizen but with a lot of instability. I'm beginning to believe it's the CPU after reading through this thread.

[MrZajic]

Wow! I was stunned if I found it should do i9 13900K. Yesterday, we started to figure out why some games are nonsensically crashing with a friend. And all clues are ending at the i9-13900K as a source of problem.

Reducing the processor frequency immediately solved the problem and I am little shocked with this in 2023 year. Is the situation going to get worse? RMA? Will a complaint and a new piece solve it? There are a lot of reports that people's problem reappeared after 1-2 months.

Reducing the processor frequency cannot be a permanent solution! After all, it's a high end processor!

Are only certain versions affected? Will Intel publish a list of defective serial numbers of i9-13900? Please Intel support, do you have more information about this? Any statement?

[LoloWiwi]

Hi MrZajic,

You need to contact your Intel local support. They will ask you to go through a list of tests to confirm your CPU is faulty (I guess it's going to be Intel Extreme Tuning Utility and stress testing with AVX2).

Important : at first, they ask to test with IPDT (Intel Processor Diagnostic Tool) which is total crap and cannot reliably test CPU (the full test runs in less than 5mn...) -> this test would always PASS even though my CPU was faulty.

As you said, reducing clock speed is not a good solution -> why buying an i9, then !!??

In my case, when I RMAed my i9-13900K I could use an i7-14700K that was having higher scores on Cinebench R23 (I had to disabled some params in the BIOS with my i9, resulting in subpar performances).

The RMA process was very quick and I got reimbursed within a week of sending the faulty CPU (I was lucky because they didn't have i9-13900K in stock so they offered to reimburse), and I got an i9-14900K.

Scores in Cinebench R23 (everything identical between both, apart from CPU):

- i9-13900K with limitations in BIOS (MultiCore Enhancement disabled, SVID Behavior = Intel's fail safe) -> score ~ 32500

- i7-14700K with all BIOS settings to ASUS default -> score ~33000

- i9-14900K with all BIOS settings to ASUS default -> score ~39500

Performances +22% with the I9-14900K Vs the "gimped" i9-13900K, and even the i7-14700K was having better perfs...

[MacDaddy]

Today I contacted Intel about this same exact issue! 

You are correct, they run the basic tools and say we'll get back to you...BUT... I showed him this post and he wanted to see Cyberpunk crashing on boot. 

SO...I did just that. He was remoted into my PC through 123Rescue. 

He seen what I meant, and he attached this post to my ticket. 

I hope I'll receive a reimbursement or remedy of some kind. I can just go for the RMA but I don't have the time to wait around for over a replacement. 

I am not sure if it's a good idea to share my ticket number, but if anyone from Intel can contact me directly I'll be more than happy to share it. 
I had trouble even typing this reply up because my browser keeps closing on me

[Powerboy]

i have same issue with 14900k, its really annoying, did you had luck this time not get faulty cpu? its totally solved the issue? you had "out of memory error on shader compiling right? or watchdog bluescreeen?

[MrZajic]

Next tip for workaround I getted about set Long Duration Power Limit and Short Duration Power limit.

BIOS setting:

We setted PCore back to auto.

Long Duration Power Limit: 150 (from 4000 )

Short Duration Power limit: 260  (from 4000 )

(by default motherboard have something unlimited value like 4000 etc)

I'm still trying to understand how such a workaround could have an impact (+- values correspond to the TDP values). But really after this change crashes dissapeared again. But we tested it in the short term. We'll see how stable it is in the longer term. OCCT also passed without crashing.

And what is strange is that mostly DX12 applications are affected. DX11 looks okay. It is still the same problem as at:  https://www.reddit.com/r/intel/comments/13o29w5/13900k_will_no_longer_run_dx12_games_crashingctds/

Any official Intel statement? Thank you.

[KCLam]

Hi all,

Seeing that this topic has drawn some attention again, I would like to share my RMA experience for your reference.

I contacted my local intel customer support (not the distributor) via the procedures mentioned by n_scott_pearson (page 5), the timeline is like this:

Nov 6 - Phone called Intel Customer Support, ticket created

Nov 7 - Recieved email getting further information from me, I replied, and they offered RMA for a new 13900k

Nov 8 - I prepared the package, package picked up on the same day

Nov 11 - They recieved and shipped the new 13900k to me (they have stock of 13900k in the Malaysia spot, unfortunately, no refund option was provided as LoloWiwi had)

Nov 14 afternoon - Recieved the package, the cpu works fine afterward.

---

My old and new 13900k comparison:

Before (in event viewer):

589 0xc0000005 application error in 3 months (Sep to Nov, limited by the storage limit of events), not counting countless times of 0xc0000005 error in chrome internally.

After:

7 0xc0000005 application error in 1 month (14 Nov to 17 Dec), without banning any of the cores. Not once 0xc0000005 in chrome.

So yeah, I think I got a not faulty CPU now. 

Although I truely truely truely believe that I deserve compensation for all the suffer, confusion, and time lost on figuring out and resolving this issue starting from March, I have no better option but to RMA with the only option intel provided. Before more time is wasted on something you cannot fix, just RMA it. By compensation I mean at least a choice for a full refund

I am actually quite happy with the RMA process, the customer support guy is very polite and friendly, and the whole process took way shorter time than I expected. They said it may took 4 weeks for the RMA, but I recieved the new cpu in less than 1 week after I shipped the old one, so this part of the whole thing is quite good.

[Powerboy]

whats strange to me thats its same exactly issues with my 14900k it pass avx2 test without xmp profilebut games out of memory the same, so it not depend from avx2 test, what you advice to test for check faulty cores? curious thing, all work good when i set intel fail safe, and if i drop 1 point core ratio from 57 to 56x, but really, i just bought that cpu and dont want it give me problems when in 3 4 years going to oc it.

[LoloWiwi]

You should look other messages in this thread, myself and somebody else provide a way to select cores when running an app.

It allows sometimes to narrow down to which core(s) is/are faulty, if any.

[Powerboy]

you know my issue to launch ue5 games was related by cooler? seems its quite impossible cool a 14900k with 240 mm, so on shader compliling went in trottling, tomarrow i will speak with intel assistance