Hi,
I recently bought a 5900x to upgrade my 2600x. I bought a new motherboard too (MSI MPG X570S EDGE MAX WIFI), a Samsung 990 PRO, a Samsung 980 PRO and reused my 850 PRO. Initially I used my Corsair RM650X with a molex to 8 pin CPU adapter (RM650x only has one CPU 8 pin connector) and my Radeon RX580. Windows was clean installed. After a couple of days I had my first reboot with WHEA LOGGER 18 error (Cache Hierarchy error CPI ID 10). After that I had other random reboots, so I decided to change my RAM (I had 4 8GB@3000 LPX modules from Corsair, now 4 8GB@3200 Kingston Fury Beast, both with XMP enabled), but again another WHEA LOGGER 18 error. So I changed my PSU with a Corsair RM1000x, but still another WHEA LOGGER error. I upgraded the BIOS of the mobo to the last available, but than again another WHEA LOGGER crash. I searched the Internet and found different threads with a lot of settings to set in BIOS "to solve" this problem, so I modified them:
- PSS Support disabled
- Global C-State Control disabled
- Power Supply Idle control typical
- Power Down Enabled for RAM disabled
- Gear down mode for RAM disabled
- XMP enabled
- Spread Spectrum disabled
- ERP Ready disabled
- CPU Switching to 500 KHz
- SOC Switching to 500 KHz
In Windows I set minimum processor state to 15% (as I had these crashes when my computer is on idle, for example when I'm watching a video. Under load, for example during Cinebench R23, other test like Prime95, or a video rendering nothing happened).
It worked for 3 weeks or so, but today it crashed again with a WHEA LOGGER error.
Every error has 10 or 11 as CPU ID, so the problem should be on core 5.
Do you think the CPU is defective?
What are you using for cooling? How's the airflow thru your case? (you have to keep your VRM's and RAM cool too)
Have you tried simply bumping up the voltage on your RAM fron 1.35v to 1.36v, 1.37v ..etc (I wouldn't go above 1.40v personally )
Have you tried just running 2 sticks of RAM in slots A2/B2?
The CPU cooler is a Noctua NH-D15 SE-AM4, the case is an Antec P101 Silent with 3 Noctua fans on the front e 1 on the rear. During Cinebench CPU temp is always under 75°C. But, as I said, the reboot happens when the system is not under load, never crashed under load.
I haven't still tried to bump the RAM voltage... I'll try this too.
Unfortunately, after 45 days it rebooted again with a WHEA-LOGGER 18 error and CPU ID 11.
I bought another 5900x, hoping to solve this situation
End of February 2023, I have bought exactly the same combination of the AMD Ryzen 9 5900X CPU with the MSI MB MPG X570S Edge MAX Wifi motherboard and the Nocuta NH-D15 SE-AM4 CPU cooler. And I also got crashes with random reboots with WHEA-LOGGER 18 in the Event logger and more recently BSODs with WHEA_UNCORRECTABLE_ERROR. I wonder whether the motherboard might have gotten damaged during the fan installation? In any case this is very frustrating. Please let me know whether replacing the CPU helped for you.
The new CPU arrived today. I tried with the only way I knew to trigger WHEA-LOGGER 18 (i.e. using a blur background with a videocall application like Zoom or the background removal OBS plugin) and it didn't happened.
As I said, I didn't have that error under load, but I tried a couple of run of Cinebench R23 and nothing happened. I'll see if it is solved in the next days. :crossfingers:
I don't know if it's similar. But I have an "MSI MPG x570 Gaming Plus" motherboard and my Ryzen 5950x has started freezing at random times. I can be in Safe mode, Windows pre-boot environments and have it still freeze. It will also freeze and reboot sometimes. Sometimes I get bluescreens for ntkrnl and it being a watchdog event.
What is odd is this system has been running smoothly for 2 years. I ended up doing bios updates, swapping memory, messing with XMP. But my board has a "gaming boost" mode that is set in the bios. I decided to enable that mode since it will then do its auto overclocking on certain things. System is now running stable and no longer freezing, rebooting randomly. I'm not sure why the stock settings quit working right... Was it a recent windows scheduling change in an update with Windows 11? I did try a Windows 10 WinPE boot environment on the computer and it was 100% stable as well. I think it has something to do with Windows 11 and core tweaks...
Replaced 3950X with 5950X = WHEA and reboots | Overclock.net However it seems like maybe it's much bigger than that? I just don't know why it has taken so long to manifest for me and why only Windows 11 in my case and why the Gaming Boost mode in my boards motherboard fixes it...
Was there a Microsoft Edge update before the crash today? Usually after such updates, failures occur. Try disabling hardware acceleration in Internet Explorer, it has not been completely removed yet in Windows 10 and 11. WHEA LOGGER 18 error (Cache Hierarchy error) is not the root cause of this failure, but is only a consequence of an unplanned power outage. Try turning off the surge protector or isolating the power wire and you will get exactly the same entry in the Windows event log. And I'm sure it's not the marriage of the processor.
I have a ryzen 9 5900x and I have been getting the following error in my event viewer:
"The WHEA LOGGER processor has reported a non-fatal hardware error."
I have tried to update my BIOS and reinstall my OS, but the error persists. Has anyone else experienced this issue?
I'm waiting for a new 5900X: I tried a lot of solutions, but noone solved for me.
I found that my system cis particulary unstable when I use blur background on my webcam (for example with Zoom or using the backgroud removal plugin for OBS).