Windows 10: Windows 11 MSI MEG x570 AMD Freezing with DPC_WATCHDOG_VIOLATION

Windows 11 system has been freezing up several times a day for the last month. Sometimes this creates a BSOD and minidump, other times it's just unresponsive and requires a hard reboot. Problem may reoccur within a few minutes, or a half a day of steady use. It occurs when in constant use; and when idle or overnight. This system worked for over a year without issues. Nothing identifiable changed proceeding this issue.Basic ConfigCPU: AMD Ryzen 9 5900X 12-Core Processor, 3701 Mhz, 12 Cores, 24 Logical Processors\MB: MSI MEG x570 Godlike updated firmwareOS:

:)

 

Cannot install windows 11 22H2 upgrade or a fresh install. BSOD

Hello,

I cannot install a new version or upgrade to windows 11 22H2

A fresh install of windows 10 or windows 11 21H2 works, but not for 11 22H2.

I receive a BSODs like KERNEL_MODE_HEAP_CORRUPTION or IRQL_NOT_LESS_OR_EQUAL during installation.

The latest HW change I did is replacing the graphics card to RTX 4070Ti.

I have upgraded to the latest BIOS on my motherboard.

HW:

AMD Ryzen 9 3950x

MSI Meg x570 Unify

32GB RAM

Nvidia RTX 4070Ti

 

Ryzen 5800x + X570 tomahawk WIFI BSOD

Hello there,

I recently purchased a MSI x570 tomahawk wifi and after assembling i am facing lots of BSOD issue. After days of reading troubleshooting steps on linustechtip, MSI, AMD, reddit forums i am still facing BSOD UNLESS i turn off Core Performance Boost (CPB) in BIOS. Hope i can get some further help here. Please find below a summary of my setup below:



My new setup:



CPU: AMD Ryzen 7 5800X (NEW)

MOBO: MAG X570 TOMAHAWK WIFI (NEW)

DRAM: Corsair Vengeance LPX DDR4 16GB X 2, 3600Mhz (CM4X16GD3600C18K2D-CN ver 3.44, CM64X16GD3600C18K2D-CN ver 5.49) (NEW)

GPU: ZOTAC GEFORCE RTX 2070 Super Twin Fan 8GB (OLD)

PSU: Cooler Master MWE Gold 750 Full Modular 80+ Gold (OLD)

SSD: Samsung EVO 980 PRO M.2 (NEW)

BIOS version: E7C84AMS.150

CPU cooler: AMD Wraith (NEW) *Planning to upgrade to a AIO soon after i get this bsod sorted out.





Issues faced:

1. Constant BSOD (WHEA) before reaching Windows.
   
2. Sometimes able to go into Windows but after 1-2min of idling (low load) will BSOD (WHEA)
   
3. Windows event viewer showing Kernel power (41)
   

Troubleshooting steps taken:

1. Reinstalling windows from flash drive.
   
2. Updating bios, chipset, NVIDIA and windows /drivers update to latest.
   
3. Reseat CPU, PSU, RAM, GPU (basically all connections).
   
4. Following AMD troubleshooting steps.
   
5. i) Clearing CMOS by removing MOBO battery.
   
6. ii) Set power supply idle to Typical in BIOS.
   
7. iii) Select balanced power plan in Control Panel.
   
8. iv) Sliding performance and energy slider to middle.
   

1. Swapping DIMM slots between 2,4 and 1,3 and trying with 1 ram only.
   
2. Turn off C-state BIOS.
   
3. Turning off XMP.
   
4. Turning off PBO. (Precision Boost Overdrive)
   
5. Manually setting Vram to 1.35V.
   
6. Running memtest86 using OLD MOBO and OLD INTEL CPU with NEW RAM. 4passes with 0 errors.
   
7. Chip temperature should not cause BSOD as im constantly checking and only reaching 35-45deg on idling. Setup in aircon room with ambient 23-24deg.
   
8. Running memtest86 on NEW MOBO with NEW RAM.
   
9. Testing with a different RAM, MOBO, and PSU model (borrow from friend)
   

Further troubleshooting using friend's mobo and chip:

1. Able to run under load and idle after installing my 5800x onto my friend's X570 gigabyte gaming x with CPB turned on and XMP.

2. My friend's 3700x is able to run in my x570 tomahawk wifi board with CPB turned on and XMP.

3. I have RMA the X570 tomahawk wifi mobo and still facing the same issues with the new 5800x CPU.



I'm confused as to why the 5800x is able to run stable on the gigabyte board but not on the MSI board, and the 3700x cpu is able to run stable on the both boards. The ONLY way to get the pc (5800x + MSI x570) to run windows without BSOD is to TURN OFF CPB. (Tested for more then 5hours with iding load no BSOD). As soon as i turn on CPB, BSOD (WHEA) appears at the login screen.

Hope to get some advice and thank you in advance.

 

Fix random freeze for AMD CPU, MSI motherboard and nVidia Video Card

Windows 10 Pro 64-bit

GeForce GT 710

AMD FX-8320E CPU

MSI 970A-G46 motherboard

The system freezes randomly without trace in event viewer. Sometimes the freeze will recover. Sometimes only a hard reset will do. No clear usage pattern causing this. It can happen right after reboot or idling. Another issue was there is random buzz
during audio playback

I have done a lot research and experiment. And finally I have a stable system for over a week. Here is a quick summary. Hope this can help others.

• Clean installation of Windows 10 with network unplugged. Initially I did the upgrade from Windows 7 so was not sure if it's because of that. So I did clean installation this time but still got the random freeze.
  
• Install nVidia Driver. I tried several versions and got random freeze with each of them. So I settled with one that I used before 364.72. This is before step 3. After that, I did not go back to upgrade the driver again.
  
• nVidia driver installer will corrupt opencl.dll. I used
  
  SFC and DISM and fix it. I suspect this also causes the random buzz during audio playback but I can't confirm it since I did not play audio at this point
  
• Plug in the network and update. Freeze continues
• Upgrade SSD firmware. Freeze continues
• Upgrade motherboard BIOS. I've checked MSI website many times and it never showed an update. It turns out that you have to install MSI Live Update. It showed available update and I installed it
  
• System fails to start after BIOS update. Had to pin-reset and it rebooted fine. Went into BIOS and configured as I would normally do. Reboot, freeze continues.
• Run memtest86+ overnight. No error. Reboot, freeze continues.
• Went back into BIOS, disabled C6 State in power management (kept C1E enabled). Rebooted, no freeze in a week

I can't say everyone has the same problem. I had to spent several days to troubleshoot and had essentially 3 issues

• nVidia driver corrupts opencl.dll
• BIOS update
• Disable CS State

One thing worth note. I have 2x8 GB Patriot RAM. In BIOS, I've enabled AMP. According the manufacturer spec and CPU-Z, CAS under AMP is 10. However, CPU-Z showed the actual CAS is 9. I did notice this until the system became stable. Since memtest86+
passed overnight, I did not slow the RAM down.

I started with a Radeon video card. It had many crashes. I blamed it on the lack of updated video driver (AMD abandoned the old cards) and switched to nVidia. I did not go back to test this with Radeon card.