Mystery PC crashing

[ToxicChild]

This post is just for theorizing and speculating, any thoughts or advice is welcome.

 

In fall of 2023 I made this build:
CPU: Intel Core i9 13900k

RAM: 2x16 GBs T-FORCE Vulcan DDR5 @ 5200MHz
GPU: ASUS TUF GAMING 4090 OC
MoBo: ASUS TUF GAMING Z790 PLUS WIFI
Storage: 1TB SAMSUNG SSD 980 PRO, 2x 2TB SAMSUNG SATA SSDs 870 EVOs
PSU: CORSAIR RM1000x (2021)
Case: HYTE Y60

Note: the PC uses an aftermarket 12VHPWR to 4x8 ATX cable from CableMod to avoid clutter and stay away from the included controversial NVIDIA adapter in the case

About 12 months after building and daily usage, whenever I run a demanding game at high settings. The system will sometimes crash in a strange way after 5-10 minutes under load;
-The screen will go black, and then display no signal 10 seconds later,
-The PC will crank all fans at max speed
-Audio and sound will continue to play, but no updating in surrounding sound (ex: if playing an FPS game, the ambient/background music will still play but all the shooting and other short sounds will stop or loop)

-Keyboard and mouse become unresponsive; pressing caps lock will still toggle the caps, but other key combos such as ALT+F4 will not stop the music in the game.
-The system will not respond until it has been force shutdown and rebooted with the power button.

Since then, I've chalked it up as being a Thermal Shutdown, assuming the graphics card got too hot and tripped from being stuck inside a cramp case. I went ahead and did some precautions to prevent it from reoccurring.

-I utilized MSI afterburner to adjust the GPUs thermal limit from 86C down to 75C

-Installed an additional NOCTUA Chromax case fan below the GPU hot spot

-cleaned all case filters

After a 3DMARK stress test, the GPU didn't meet its baseline frame stability test (96% to 92%), but didn't shutdown either, so i usually played games on low/medium settings or games that aren't demanding in power.

This week In preparing to play DOOM: The Dark Ages, I've done all the prerequires to playing the game, including updating the BIOs, the CPU Microcode, and NVIDIAs GPU Game Ready drivers (576.22), as well as insuring the thermals didn't skyrocket with afterburner.

 

Here's the strange part. the PC crashed AFTER the 10 hour game session. the GPU was under constant load the majority of the time (~350w) and the game played smoothly at 120fps on the out of the box ultra settings. it gave out 15 minutes later when i was checking my discord messages and reading email.

 

My new theory is the Power Supply (or CableMod cable) is going bad and is not adjusting well to either load changes or temperature changes. as of writing i already ordered a new PSU (CORSAIR 1500x) which would solve both possible failure points.

 

Any ideas on what might be wrong? or things to try? if i left any details out you think might have a factor feel free to ask.
Thank you for reading  

 

[Sawa Takahashi]

I don't want to be the bringer of bad news but JayZtwocents made a video about a similar issue recently. He said that mystery crashes like yours are often a symptom of CPU degradation.  Ironically, it happened to Jay's own computer a few days later. Same issue, same cause. I posted the two links just below.

Hope that helps

 

 

 

[ToxicChild]

On 5/27/2025 at 12:14 PM, Sawa Takahashi said:

I don't want to be the bringer of bad news but JayZtwocents made a video about a similar issue recently. He said that mystery crashes like yours are often a symptom of CPU degradation.  Ironically, it happened to Jay's own computer a few days later. Same issue, same cause. I posted the two links just below.

Hope that helps

 

 

I was afraid that might've been a contributing factor, after all the 13900k is part of early batches in 2023 with the whole microcode scare, i knew to not push the cpu too hard then.

The main culprit as i theorized was the power supply, i swapped the cabling and PSU with a corsair 1000x with a corsair 1500x for a fair price. the random crashes on DOOM TDA disappeared, and i was able to push the GPUs settings without fear with its powerdraw up to 300W+.
 

I see where you got at with CPU degradation, for a while i believed it was that. but then it didn't make sense why any CPU benchmarks went fine, but gaming did not. Pretty sure It was power delivery to the GPU being inconsistent either from the CableMod cable or the Power Supply itself. and the whole reason the cable existed was because the in-box Nvidia adapter would've put unnecessary stress on the connector, which might've caused it to melt.
(it also could've been Power delivery to the CPU too in the end)

The PC continues to function and haven't had an incident since the PSU swap. if it ever occurs again i'll come back to this thread, but I'll still be cautious and be nice to the processor

[_hitman_]

Top reasons for these kid of problems are:

1. Bad GPU power cable or not seated properly

2. Chipset drivers

3. BIOS corruption

4. Bad PSU