13900k/Z790 Hero Issue

Today I tried to update my GPU drivers, I used DDU in safe mode restarted, instant fail when I ran the installer, didn't even bother looking for the card.
This happened once before, didn't matter what GPU I had in the machine it wouldn't install, couldn't extract zip/rar either, I had to tighten my Thermalright contact frame, it fixed it, but it hasn't fixed it this time.
I got to the point I reinstalled the default latch, still the same, I actually can't even extract a zip/rar file without it failing, yet Cinebench passes.
Reseated the CPU multiple times, checked for damage (none),

I've tried 2 set of DDR5 one 32GB 6000 set and a 32GB 6400 set both 2x16, still the same, CPU at defaults.
Also ran single sticks at defaults (no XMP).
I'm running 2002 BIOS on my Z790 Hero with a 13900k, I don't know if it's the CPU (unlikely) or the motherboard, this issue has been popping up a few times over the last 12 months, but this time there's no rectifying it.

It boots fine, I get into Windows, I can manually install the GPU driver (4090) through device manager, but the fact a rar/zip fails the CRC and doesn't extract is a major concern.

I will add my 980 Pro NVME is a 87%, so I don't know if that could be causing a issue...

I was playing Nightingale for 5 hours yesterday without a single problem, just went to update the driver today..

Any ideas?

Thanks

 

I don't have much input, except to ask, why on earth are you using DDU to begin with, if you're merely updating you GPU drivers?

 

Well it failed there first so I ddu, it's worked in the past..
Today even stock ram settings are unstable......

I'm typing this with 1 ram stick in at stock and chrome is still randomly crashing..

 

Probably the SSD.

 

Funny enough I can use 7zip to extract fine, I'd been looking to replacing my SSD with a 2 TB.
Samsung Magician don't pick up any errors, neither does Crystaldisk info.

I removed the GPU and am running off the CPU's integrated GPU, drivers went straight on through windows update, and no actual issues (other than Winrar) or Chrome crashes.

I've pounded the CPU over and over again with every stress test I can find, including ram, and it's passing all of it.

 

I tried to type back a really long reply but chrome just closed. so I'm guessing SSD.
Hence the double reply...lol

 

Thinking it's the CPU ShadowBurger does any of this sound familiar?

 

this is similar to my old fauly 13900K but it would fail stress test. I used OCCT

 

What test did you use when it failed, it's the only one I haven't tried yet...

I just ran every 3D Mark benchmark and it all worked, but still can't install the drivers with the actual installer.

NVM "
00:00:00 - Info - Test schedule started at 2024-03-07 12:00:48
00:00:00 - Info - CPU - Started (Duration : 01:00:00)
00:04:21 - Error - CPU - 1 error(s) found on physical core #1 - logical core #3"

This is with default BIOS settings and OCCT set to Medium Data set, mode normal, load type variable and auto instruction set.

Though there is one thing I'm noticing, the 10-15 second pause when the login screen try's to show after a reboot, this is on a 500GB 980 Pro NVME...

 

This happens with enforce all defaults on in bios:


00:00:00 - Info - Test schedule started at 2024-03-07 12:25:53
00:00:00 - Info - CPU - Started (Duration : 01:00:00)
00:11:55 - Error - CPU - 1 error(s) found on physical core #1 - logical core #2
00:12:41 - Info - CPU - Test stopped
00:12:41 - Info - Schedule stopped by user request

 

Have you tried Prime95 and/or LinX? I'm sure you would fail those if you're failing OCCT but worth being certain before going down the RMA route.

Shadow successfully had his 13900K replaced by Intel if I'm not mistaken.

 

Didn't they cancel that recently, I know they were doing cross shipping in the past..

Trying Large FFT on Prime95 now.

There's no way I should be hitting 100c with enforce all limits on, yet I am..

 

No clue, how recently are we talking?

Try small FFT for solely CPU stressing. Which cooler are you using; it's not unusual for a 13900K to hit TjMax with these particular stress tests.

Are you getting any errors or failing?

 

Not yet, am running the small FFT's not the smallest FFT's test..

 

Yep, just small.
Hmmm, curious. If it was faulty I would've expected it to throw an error within minutes.

 

It's lasted longer than OCCT, so I'm kinda back to square one, unless it is the NVME, I bought both the 500GB and 1TB 980 Pro's on launch day, so they'd be a good 4 years old now.
The 500Gb is the boot drive, it only has a 83% rating in crystaldisk info where the 1TB is still 97%...

 

Just got this trying to reinstall the drivers through device manager...

 

I think that issue is too generic. What're Event Viewer and Reliability Monitor saying when you get the driver install errors?

 

Thats the thing there isn't any for the driver fail.

I did a Dism /Online /Cleanup-Image /RestoreHealth then tried installing from the installer, it went through, rebooted.

I tried running the installer again over the ones that just got installed and I get this,


Yet the drivers are installed....

 

I still think it's the CPU. Do you have access to someone else's 12th-14th gen processor to rule this out?