New CPU, RAM and Motherboard and constant crashes and windows installation error when I try to reinstall any programs.

I kill Bunnies · 11 May 2025 at 14:01

tamzzy said:
Run memtest and see if there's any ram errors

I have ran Windows Diagnostics and there were no issues found with the RAM.

Mcnumpty2323 · 11 May 2025 at 14:03

Long post so kind of lost track
So may ask something already asked before
At least can probably rule out windows now
Unless its a 24h2 issue

Is it only games causing a problem?
All games or specific games?
Does looping a gpu benchmark for an equivalent length
of time cause same problem?
Monitoring temperature when it crashes?
Cpu and gpu temperature

Motherboard on latest bios?
And yes run memtest

Got any friends or family with compatible hardware
Where you could double check stuff?

New cpu have integrated graphics?
If so does it happen on igpu instead of gpu?

Windows ram diagnostic may not be
As thorough a test as memtest
Still better than not testing the ram though

tamzzy · 11 May 2025 at 14:40

I kill Bunnies said:
I have ran Windows Diagnostics and there were no issues found with the RAM.

Windows diagnostics does not show up intermittently faulty ram. You'd need a full memtest to do that

MemTest86 - Official Site of the x86 Memory Testing Tool

MemTest86 is the original self booting memory testing software for x86 and ARM computers. Supporting both BIOS and UEFI, with options to boot from USB.

www.memtest86.com

AndyM001 · 11 May 2025 at 15:27

I kill Bunnies said:
The GPU has 2 x 8 pin connectors. I have slotted in a 6+2 connector into each.

I have PBO enabled now (I removed the 65 wattage) but what is CO?

The cooler is a Noctua NH-U12A and this has been refitted with new new thermal paste, problems persist though.

Just to check, you are using 2 separate cables from the PSU to the GPU, and not one cable that splits in to 2 at the GPU?

Mcnumpty2323 · 11 May 2025 at 15:53

Ideally yes you want 2 separate 8 pin cables
Not a pigtail cable
Though my 3080ti has no issues using pigtail
With an 850w psu
Even in unleashed mode at 400w

And think its original psu and gpu they're using
So while possible it's a less likely issue
Unless new cpu is drawing a load more power

Think they ordered a new psu anyway
So once that's installed
Can see if there's any change to the situation

Edit
3 x 8 pin on my 3080ti not 2
1 separate cable and 1 into 2 x 8 pin pigtail
Wouldn't try 400w through just 2 x 8 pin

I kill Bunnies · 11 May 2025 at 16:41

Mcnumpty2323 said:
Long post so kind of lost track
So may ask something already asked before
At least can probably rule out windows now
Unless its a 24h2 issue

Is it only games causing a problem?
All games or specific games?
Does looping a gpu benchmark for an equivalent length
of time cause same problem?
Monitoring temperature when it crashes?
Cpu and gpu temperature

Motherboard on latest bios?
And yes run memtest

Got any friends or family with compatible hardware
Where you could double check stuff?

New cpu have integrated graphics?
If so does it happen on igpu instead of gpu?

Windows ram diagnostic may not be
As thorough a test as memtest
Still better than not testing the ram though

All games, either crash when playing (Minecraft every time) or crash as I exit the game (PUBG and Cyberpunk).

I have the updated BIOS, updated CPU, motherboard and GPU drivers.

I have watched Adrenaline whilst playing and temps look fine, CPU and GPU look fine. I will run Memetest85 later.

I will see if the integrated GPU has the same problem later.

I kill Bunnies · 11 May 2025 at 16:43

AndyM001 said:
Just to check, you are using 2 separate cables from the PSU to the GPU, and not one cable that splits in to 2 at the GPU?

Two cables, each one is 6 pins with 2 extra pins attached. I put both 6 plus 2 into the 8 pin slots on the GPU.

Mcnumpty2323 said:
Ideally yes you want 2 separate 8 pin cables
Not a pigtail cable
Though my 3080ti has no issues using pigtail
With an 850w psu
Even in unleashed mode at 400w

And think its original psu and gpu they're using
So while possible it's a less likely issue
Unless new cpu is drawing a load more power

Think they ordered a new psu anyway
So once that's installed
Can see if there's any change to the situation

Edit
3 x 8 pin on my 3080ti not 2
1 separate cable and 1 into 2 x 8 pin pigtail
Wouldn't try 400w through just 2 x 8 pin

Yeah new PSU should arrive next week, I'm leaning towards this being the culprit.

tamzzy · 11 May 2025 at 16:51

I kill Bunnies said:
or crash as I exit the game (PUBG and Cyberpunk).

That sounds like memory issues.
Please run memtest as suggested earlier.
Also, what is your vddp voltage running at?

I kill Bunnies · 12 May 2025 at 12:48

I have been running Memtest86 on a single stick of RAM for 5 hours now...

tamzzy · 12 May 2025 at 12:49

ummm...why only a single stick? :confused:

I kill Bunnies · 12 May 2025 at 13:03

The tutorial I watched said to do it that way, but I regret doing so now!

The tests are in for the first stick and everything is fine, doing the second now.

tamzzy · 12 May 2025 at 13:09

I kill Bunnies said:
The tutorial I watched said to do it that way, but I regret doing so now!

that tutorial is bullocks :cry:

you need to test the ram together so it'll run in dual channel mode

Mcnumpty2323 · 12 May 2025 at 13:10

Yeah i have always ran it with all
The modules in situ
Otherwise you'll be at it a long time
If something shows errors with them all in
That would be the time i would only put
Whichever looked like an issue through a test on its own

I kill Bunnies · 12 May 2025 at 14:18

OK restarted with both sticks in.

Tetras · 12 May 2025 at 14:49

Tetras said:
SSD
Do you have a 970 Evo Plus?

If you do, download CrystalDiskInfo and post the SMART results here.

Check your firmware is safe (not one of the affected ones).

I kill Bunnies said:
All games, either crash when playing (Minecraft every time) or crash as I exit the game (PUBG and Cyberpunk).

^^ Done this yet?

Probably not the problem if the drive is old, but worth checking because there's known issue with these drives.

I kill Bunnies · 12 May 2025 at 15:24

Tetras said:
^^ Done this yet?

Probably not the problem if the drive is old, but worth checking because there's known issue with these drives.

It is 970 Evo plus, but that was in the previous unit and worked fine, so is it still worth running these tests?

Mcnumpty2323 · 12 May 2025 at 15:30

Nothing to lose by testing it anyway
While waiting for the psu

Tetras · 12 May 2025 at 15:31

I kill Bunnies said:
It is 970 Evo plus, but that was in the previous unit and worked fine, so is it still worth running these tests?

There are no tests, downloading CrystalDiskInfo or checking on Samsung's Magician takes just a few minutes. Would take longer to post the screenshot for us to look at.

I kill Bunnies · 12 May 2025 at 19:14

New PSU has arrived. Unfortunately it will be another week before Memtest86 has completed its checks.

Tetras · 12 May 2025 at 20:03

I kill Bunnies said:
New PSU has arrived. Unfortunately it will be another week before Memtest86 has completed its checks.

Usually if you have faulty memory it will show up pretty quickly (like within hours, minutes, or even seconds). If it takes a very long time to produce an error, that's more likely to be a configuration issue than a faulty stick.