Does the BSOD or access violation occur with dgpio3.sys?
Does the BSOD or access violation occur with dgpio3.sys?
Hello,
I’ve faced several issues over the past two months, mainly with access violations and corrupted memory—this problem is new. While BSODs are rare, most times apps or games crash due to an access violation. The troubles began after I reinstalled Windows. Crashes happen in every game I play; some after a few minutes, others after hours of use (Arma always crashes the fastest, even without mods). Chrome also sometimes freezes with an "Aw, Snap!" message, citing access violation as the reason. Temperatures remain normal—GPU stays below 75°C and CPU under 70°C, even during heavy usage.
My system specifications are:
AMD Ryzen 5800X + THERMALRIGHT Peerless Assassin 120
RX 6700 XT Red Devil
MSI B450 TOMAHAWK MAX II
2x 16GB Kingston 3200 MHz (KF432C16BBK2/32)
Seasonic 750W FOCUS-GX-750
Samsung 980 (90% health) as the main drive
Samsung SSD 870 EVO (99% health)
Apacer AS350 (100% health)
Kingston SA400 (69% health)
I’ve tried everything available online: reinstalling BIOS/GPU drivers, updating chipset drivers, running diagnostic commands, using Windows Memory Diagnostic and MemTest86, disabling XMP and enabling Eco Mode (65W), testing with a single RAM stick, updating firmware, and consulting Microsoft support. They suggested using Driver Verifier and provided a memory dump pointing to "amdgpio3.sys".
Could there be another approach before replacing parts? Also, here are some older dumps/minidumps.
Welcome to the forums, newcomer! For relevance, what BIOS version are you using for your motherboard? When you mentioned amdgpio.sys as the problem, did you install the newest chipset driver via an elevated command? I observed WinDBG and Ntfs.sys as the component causing the issue. Are there any pending updates for your operating system?
This link is for downloading your chipset from AMD. Before installing the new version, use Revo Uninstaller to remove any previous installation, ensuring no leftover files or registry entries remain. It's wise to do this with all drivers to prevent conflicts between old and new versions, which can be difficult to detect and resolve.
Hello, so I've tried uninstalling chipset drivers with Revo, but that didn't help. Then I tried the following:
"In the BIOS, locate the Power Supply Idle Control option and set it to Typical (this option should be available in the Advanced section of the BIOS)."
That didn't have any effect either, and then my PC crashed again. I then tried changing the GPU to a GTX 1050 (uninstalling drivers with DDU, then installing NVIDIA drivers), but that didn't help. I also tried changing the OS drive and having only it (Samsung SSD 870 EVO), but the issue persisted.
Is there any way to stress test the CPU or motherboard? (I don't see any pattern that could cause the crashes, and I don't believe a simple stress test would have any effect.)
My current stress test is running ArmA with mods
, where it crashes on an access violation, and without mods, it crashes to a BSOD.
Here is a collection of BSODs with their causes.
Do you happen to have any other RAM that you can try? Memory corruption is a possibility for this issue. Also try updating the bios to the latest version, you never know.
So even after writing this down, I experimented with various RAM options (4GB), and the same issue persisted.
However, when attempting to power on my PC using that RAM, the LED signaling a faulty CPU was illuminated. After resetting CMOS, the system started up, suggesting the CPU might be defective. If it isn't, then I'm seeking help from an exorcist.
When attempting to power on the computer, it immediately crashes and displays a "turning off" message. Accessing the BIOS also results in crashes. After powering off, the system attempts to restart but fails, booting directly into a loop. I had to disable the PSU, wait briefly, then restart again, but it keeps shutting down at login. Clearing the CMOS didn't resolve the issue. I believe the CPU might be faulty, but I'm hesitant to replace it without confirming the problem's root cause.
The only alternative is to bring it to a repair shop for diagnosis. If it's the CPU, you can use the cheapest AM4 available to test it. A first-generation 1300 will also work as a trial unit.
Recent update: After powering off my PSU for a while and using another one, despite the same issues, the system stopped looping. I even tested a different outlet without success. Given this situation, I suspect the problem lies with the CPU or motherboard since only these components were changed. Once I can afford it, I plan to replace the CPU. At least my PC will still function with occasional BSODs and crashes.