cancel
Showing results for 
Search instead for 
Did you mean: 

Z390-F CPU errors at default RAM speed (XMP disabled)

test_lab
Level 7
Hi,

I started getting random BSOD's and freezing on my Z390-F board. The mini dumps weren't telling a useful story because each of them was triggered by something else. I then did all kind of tests and it seems to be a problem with running the RAM at default (2133) speed.

I ran the following tests:

- chkdsk
- sfc
- Windows Memory Diagnostic
- Intel Processor Diagnostic Tool
- memTest86
- OCCT (here i started to get errors)

When i run OCCT cpu test i immediately get physical core errors thrown when XMP is off, but not when XMP is on.

My specs:
9900K
Z390-F
2x32GB Corsair CMK64GX4M2E3200C16
or
2x8GB Corsair CMK16GX4M2B3000C15

Both RAM kits seem to be fine. I didn't mix the RAM kits, just installed either 2x32GB or 2x8GB.
Voltages are fine in idle and under load. No cooling/temperature problems.
BIOS settings are optimized defaults, no overclocking.

Any ideas what's going on here?

Thanks in advance
1,198 Views
5 REPLIES 5

test_lab
Level 7
Can anyone help or has an explanation for my issue?

May i ask if someone would run OCCT with XMP disabled and see if errors occur?

I'd like to understand why i have to enable XMP.

xeromist
Moderator
XMP is just a set of parameters provided by the memory module. It's possible that the default parameters the board is trying to apply are unstable for that memory (or perhaps the CPU). Even if you did not change or flash the BIOS there are still a few reasons why something might start spontaneously. It's rare but a BIOS can become corrupted just sitting there. So it might be worth a fresh flash of the latest firmware from the support page.

It's also possible that the silicon is aging and things that used to work no longer do. This is most noticeable when overclocking and running higher voltages but it's not impossible for aging to impact a system running at stock. The only way to really isolate this type of issue is to swap out components. You've already tried a different memory kit so it would come down to either the CPU or MB.
A bus station is where a bus stops. A train station is where a train stops. On my desk, I have a work station…

xeromist wrote:
XMP is just a set of parameters provided by the memory module. It's possible that the default parameters the board is trying to apply are unstable for that memory (or perhaps the CPU). Even if you did not change or flash the BIOS there are still a few reasons why something might start spontaneously. It's rare but a BIOS can become corrupted just sitting there. So it might be worth a fresh flash of the latest firmware from the support page.


Thanks for chiming in and offering your help.

Do you mean the BIOS by saying 'firmware'? Should i try to overwrite the currently installed BIOS (2004) by flashing a fresh BIOS (2004)?

xeromist wrote:
It's also possible that the silicon is aging and things that used to work no longer do. This is most noticeable when overclocking and running higher voltages but it's not impossible for aging to impact a system running at stock. The only way to really isolate this type of issue is to swap out components. You've already tried a different memory kit so it would come down to either the CPU or MB.


I sadly don't have spare components to try swapping things out. Is there no other method to get to the core of my problem? What i wonder why Intel Processor Diagnostic Tool doesn't throw any errors. Their tools says the CPU is running fine at stock 2133 MHz.

test_lab wrote:
Thanks for chiming in and offering your help.

Do you mean the BIOS by saying 'firmware'? Should i try to overwrite the currently installed BIOS (2004) by flashing a fresh BIOS (2004)?



I sadly don't have spare components to try swapping things out. Is there no other method to get to the core of my problem? What i wonder why Intel Processor Diagnostic Tool doesn't throw any errors. Their tools says the CPU is running fine at stock 2133 MHz.


Yes, a fresh flash of the same BIOS would correct any corruption. But again, don't get your hopes up as this is rare.

I'm not familiar with the internals of that tool and what it's actually checking. It may be that your CPU is perfectly fine and something else is going on.
A bus station is where a bus stops. A train station is where a train stops. On my desk, I have a work station…

xeromist wrote:
Yes, a fresh flash of the same BIOS would correct any corruption. But again, don't get your hopes up as this is rare.


Thanks for elaborating. I just flashed the latest BIOS again and it looks like it's working. I'm not sure yet because i just did a short OCCT run which would always throws CPU errors after runnning for a few seconds. It now runs 2 minutes with XMP disabled. I'll keep XMP off for now and see if bluescreens return.

I wonder whether there was a problem because the computer was not powered on for 4 months at all or it is unrelated.

My BIOS before flashing again were :

TPM Device Selection: Firmware TPM
HD Audio: disabled
Fast Boot: disabled
Chassis Fan 1 Q-Fan: PWM

I now have
TPM Device Selection: Discrete TPM
HD Audio: disabled
Fast Boot: enabled
Chassis Fan 1 Q-Fan: PWM