Question Strange items and I'm not sure if it's a GPU or CPU issue?
Question Strange items and I'm not sure if it's a GPU or CPU issue?
CPU throttling doesn't lead to data corruption in RAM (whether it's the RAM on the motherboard or the VRAM on the GPU). It simply reduces the CPU's clock speed, which lowers its computational power.
If your RAM seems problematic—especially if it was an older model—consider downloading and running memtest86:
https://www.memtest86.com/
A full pass (15 tests) is the minimum. Two passes are preferable, and four are acceptable. If no errors appear during all four passes, your RAM appears intact. However, if errors occur, it may indicate GPU VRAM issues (particularly if the GPU overheats).
Replacing the RAM is often the best solution—it's more cost-effective and simpler than upgrading the GPU.
Keep in mind that memtest86 can take a while to finish: about an hour for 8GB or around two and a half hours for two 8GB DIMMs. With your setup (two 16GB modules), expect roughly five hours per pass, or about twenty hours for four passes. Running it overnight is recommended.
Regarding AIOs versus air coolers: neither provides significant cooling improvement over the other when relying on ambient air. For comparable performance, a radiator size of 240mm or 280mm is ideal. Smaller radiators (120mm and 140mm) are usually outperformed by mid-sized air coolers. Single-slot radiators work well in mini-ITX builds where CPU cooler space is limited.
Here are the advantages of both cooling methods:
Hy again and thank you for your assistance.
On the old motherboard I performed memtest using my 4 stick of HyperX RAM. I ran them one by one. On two sticks I completed four passes, while on the other two only two passes due to lack of patience. The results were clear: if I ran the tests individually without XMP, everything worked fine. However, when all four were tested together without XMP, errors appeared in the first pass. With XMP again, I managed to fix it with a single stick.
pic1
I became frustrated and eventually replaced the old RAM with a new one and upgraded the memory. Even during memtest, I encountered some issues.
Regarding the cables, I connected my TV LG 55NANO913NA via HDMI 2.0 at 120Hz and the LG 27GP850P-B at DisplayPort 1.4. I attempted to use HDMI together, but the problems persisted. So I purchased new cables—specifically HDMI 2.1 and DisplayPort 2.1—to rule out that as the cause.
Other connections include a PowerPlay setup with G502 X Plus mouse, two Logi dongles (one with Prox 2), LightSpeed headphones, and a second mouse using G309. I tried pairing the two mice with PowerPlay but it didn’t work.
On my PC, I encountered errors such as Logi lamp driver issues, Armoury Crate service stopped, and Armory Crate errors, which I couldn’t resolve. I also experienced a blue screen after Blue Screen of Death on my old motherboard (B450e gaming), where the white light appeared on the motherboard until I performed a BIOS reset or removed the graphics card.
Faulting application name: ArmouryCrate.Service.exe, version: 6.1.11.0, timestamp: 0x6789b1c1
Faulting module name: Windows.ApplicationModel.dll_unloaded, version: 10.0.26100.3323, timestamp: 0xfbccc310
Exception code: 0xc0000005
Fault offset: 0x0000000000017a92
Faulting process id: 0x540C
Faulting application start time: 0x1DB9D5376B18953
Faulting application path: C:\Program Files\ASUS\Armoury Crate Service\ArmouryCrate.Service.exe
Faulting module path: Windows.ApplicationModel.dll
Report Id: 976ce3d7-c204-4282-a84f-697591ecf56b
Faulting package full name:
The Armoury Crate Service service ended unexpectedly. This has occurred once. A restart of the service will be performed within 5000 milliseconds.
Today I set a limit in BIOS and after 10–15 minutes of gaming, the crash occurred. I observed the crash dump file at C:\Windows\Minidump\032625-11171-01.dmp (Minidump). Bugcheck code: 0x50(0xFFFFE20AB691D38C, 0x0, 0xFFFFF801832CE186, 0x0)
Bugcheck name: PAGE_FAULT_IN_NONPAGED_AREA
Driver or module involved: amdkmdag.sys (amdkmdag+15de186)
File path: amdkmdag.sys
Description: ATI Radeon Kernel Mode Driver
Product: ATI Radeon Family
Company: Advanced Micro Devices, Inc.
Bug check description: Indicates invalid system memory usage, possibly due to a faulty driver. Antivirus or a corrupted NTFS volume could also trigger this error. Hardware issues like overheating RAM or video memory may be responsible.
Analysis: Likely a software issue, probably caused by a driver bug. Memory corruption is a common cause—see articles on memory issues and thermal problems.
A third-party driver was identified as the likely source of this error. It’s recommended to check for updates for amdkmdag.sys (ATI Radeon Kernel Mode Driver, Advanced Micro Devices, Inc.).
Search query: amdkmdag Advanced Micro Devices, Inc. PAGE_FAULT_IN_NONPAGED_AREA
These items are signs of data corruption, possibly in the RAM or VRAM area.
VRAM testing isn’t possible at the moment; no reliable software exists for validation.
RAM can be checked using memtest86.
If you run memtest86 and detect issues, it suggests your GPU’s VRAM may be failing—consider upgrading your GPU.
Should memtest86 reveal RAM errors, replace the RAM.
This covers the hardware aspect (including potential GPU overheating that can lead to VRAM damage).
Your crash reports might stem from RAM corruption. First, address hardware problems before focusing on software fixes.
For 80C and newer models, CPU thermal throttling is a concern. Set all fans to full speed and observe if CPU temperatures drop under load. If not, upgrade cooling options—such as a better CPU cooler or larger case fans (e.g., from 120mm to 140mm) and additional fans if your case supports it.