F5F Stay Refreshed Hardware Desktop This issue could indicate a GPU problem.

This issue could indicate a GPU problem.

This issue could indicate a GPU problem.

Pages (2): 1 2 Next
C
ColinPlays19
Member
57
12-09-2025, 11:33 PM
#1
CPU: Ryzen 9 5900x
GPU: Powercolor Radeon RX 6700XT
Motherboard: Asus ROG Crosshair VIII Dark Hero Wifi
RAM: G. Skill Trident Z Royal 32gb
SSD: Corsair MP600 PRO LPX 1TB
PSU: Thermaltake Toughpower GF A3 (1200W)
Cooler: Thermalright Peerless Assassin 120 SE
Case: Be Quiet Pure Base 500DX
BIOS and GPU drivers are all up to date.
PC experiences crashes from WHEA 18 error.
Error details: Machine Check Exception, Cache Hierarchy Error, APIC ID changes each crash, automatic restart after failure.
Issue began after playing Ghost of Tsushima a few weeks prior; frequent crashes occurred. Even with other games like My Time at Sandrock, Monster Hunter Rise, and Monster Hunter Wilds, crashes happened before the intro screen loaded.
Stardew Valley runs without problems despite over 100 mods installed (per SMAPI). This suggests a GPU rather than CPU issue.
I purchased a Ryzen 9 5900x from Amazon to replace the 5900x. Problems remained identical, though WHEA 18 errors were replaced by Kernel 41 errors temporarily. Still, crashes occurred when attempting to launch demanding games.
BIOS voltage was adjusted +0.1, possibly causing a transition to Kernel 41. I also modified settings such as enabling optimized mode, turning on DOCP, undervolting the GPU, disabling Global C states, using or removing AMD Adrenalin, reseating components, and securing cables.
Despite these efforts, the same WHEA 18 error persisted with changing APIC IDs.
I ran CPU stress tests via OCCT and RAM diagnostics with Windows Memory Diagnostic, both passed without issues.
However, GPU testing with OCCT caused an immediate crash upon starting.
This leads me to suspect a GPU problem, though I lack deep technical knowledge. Should I proceed with replacing the GPU and returning my 5950X, or consider upgrading to something like the 9060XT?
C
ColinPlays19
12-09-2025, 11:33 PM #1

CPU: Ryzen 9 5900x
GPU: Powercolor Radeon RX 6700XT
Motherboard: Asus ROG Crosshair VIII Dark Hero Wifi
RAM: G. Skill Trident Z Royal 32gb
SSD: Corsair MP600 PRO LPX 1TB
PSU: Thermaltake Toughpower GF A3 (1200W)
Cooler: Thermalright Peerless Assassin 120 SE
Case: Be Quiet Pure Base 500DX
BIOS and GPU drivers are all up to date.
PC experiences crashes from WHEA 18 error.
Error details: Machine Check Exception, Cache Hierarchy Error, APIC ID changes each crash, automatic restart after failure.
Issue began after playing Ghost of Tsushima a few weeks prior; frequent crashes occurred. Even with other games like My Time at Sandrock, Monster Hunter Rise, and Monster Hunter Wilds, crashes happened before the intro screen loaded.
Stardew Valley runs without problems despite over 100 mods installed (per SMAPI). This suggests a GPU rather than CPU issue.
I purchased a Ryzen 9 5900x from Amazon to replace the 5900x. Problems remained identical, though WHEA 18 errors were replaced by Kernel 41 errors temporarily. Still, crashes occurred when attempting to launch demanding games.
BIOS voltage was adjusted +0.1, possibly causing a transition to Kernel 41. I also modified settings such as enabling optimized mode, turning on DOCP, undervolting the GPU, disabling Global C states, using or removing AMD Adrenalin, reseating components, and securing cables.
Despite these efforts, the same WHEA 18 error persisted with changing APIC IDs.
I ran CPU stress tests via OCCT and RAM diagnostics with Windows Memory Diagnostic, both passed without issues.
However, GPU testing with OCCT caused an immediate crash upon starting.
This leads me to suspect a GPU problem, though I lack deep technical knowledge. Should I proceed with replacing the GPU and returning my 5950X, or consider upgrading to something like the 9060XT?

M
medbabe
Member
70
12-09-2025, 11:33 PM
#2
PSU:
Thermaltake Toughpower GF A3
Wattage, age, condition (original to build, new, used, refurbished)?
M
medbabe
12-09-2025, 11:33 PM #2

PSU:
Thermaltake Toughpower GF A3
Wattage, age, condition (original to build, new, used, refurbished)?

R
roberto5235
Junior Member
10
12-09-2025, 11:33 PM
#3
1200W, recently updated, assembled alongside the existing PC 2.5 years prior. Revised version.
R
roberto5235
12-09-2025, 11:33 PM #3

1200W, recently updated, assembled alongside the existing PC 2.5 years prior. Revised version.

I
IzNix
Junior Member
33
12-09-2025, 11:33 PM
#4
Is there any way to fix the issue or should I just go ahead and get a new gpu?
I
IzNix
12-09-2025, 11:33 PM #4

Is there any way to fix the issue or should I just go ahead and get a new gpu?

R
RM123
Member
195
12-09-2025, 11:33 PM
#5
The situation remains unclear as the issue hasn't been pinpointed yet. Additional information is necessary.

When reviewing the past, it's worth checking the Reliability History/Monitor and Event Viewer for any error codes, warnings, or events recorded just before or during the occurrence of these problems. The Reliability History tool offers a more user-friendly interface and can help identify patterns, such as issues emerging after modifications. Event Viewer demands more time and effort but serves a similar purpose; it involves examining logs and relevant folders. It might be helpful to start with hardware-related logs.

In summary, I recommend using the built-in Windows troubleshooters—dism and sfc /scannow—and refer to the provided links for guidance. Share your findings accordingly.
R
RM123
12-09-2025, 11:33 PM #5

The situation remains unclear as the issue hasn't been pinpointed yet. Additional information is necessary.

When reviewing the past, it's worth checking the Reliability History/Monitor and Event Viewer for any error codes, warnings, or events recorded just before or during the occurrence of these problems. The Reliability History tool offers a more user-friendly interface and can help identify patterns, such as issues emerging after modifications. Event Viewer demands more time and effort but serves a similar purpose; it involves examining logs and relevant folders. It might be helpful to start with hardware-related logs.

In summary, I recommend using the built-in Windows troubleshooters—dism and sfc /scannow—and refer to the provided links for guidance. Share your findings accordingly.

F
FUMMBLELORD
Junior Member
17
12-09-2025, 11:33 PM
#6
Thanks for the advice, here's what I did:
Ran dism on admin cmd
Dism stuck, closed cmd
Restart pc, windows update
Update stuck on underway for 30min
Restart pc
Ran dism again, stuck 62.3, waited until complete
Ran sfc scannow
Found and repaired corrupt files btha2dp bthhfenum bt hmodem
Afterwards, I ran Monster Hunter World and ended up with another WHEA 18 error. Preceding it in event viewer were the following:
WLAN-AutoConfig Event ID 10001
HttpService Event ID 114
HttpService Event ID 111
Warning--e1express Event ID 27, Intel ® I211 Gigabit Network Connection Network link is disconnected
Related critical events in Reliability Monitor just showed that Windows was not properly shut down, with no other details.
I also just did a clean reinstall of the newest GPU driver, but I'm still getting the same error and crashes.
F
FUMMBLELORD
12-09-2025, 11:33 PM #6

Thanks for the advice, here's what I did:
Ran dism on admin cmd
Dism stuck, closed cmd
Restart pc, windows update
Update stuck on underway for 30min
Restart pc
Ran dism again, stuck 62.3, waited until complete
Ran sfc scannow
Found and repaired corrupt files btha2dp bthhfenum bt hmodem
Afterwards, I ran Monster Hunter World and ended up with another WHEA 18 error. Preceding it in event viewer were the following:
WLAN-AutoConfig Event ID 10001
HttpService Event ID 114
HttpService Event ID 111
Warning--e1express Event ID 27, Intel ® I211 Gigabit Network Connection Network link is disconnected
Related critical events in Reliability Monitor just showed that Windows was not properly shut down, with no other details.
I also just did a clean reinstall of the newest GPU driver, but I'm still getting the same error and crashes.

B
Bezerk_Legacy
Junior Member
19
12-09-2025, 11:33 PM
#7
I also ran tests on OCCT for power, VRAM, and 3DAdaptive. It immediately crashed during the power test, failing to run even a full second, though it managed a few seconds on the GPU tests before stopping. All results showed the WHEA 18 error.
B
Bezerk_Legacy
12-09-2025, 11:33 PM #7

I also ran tests on OCCT for power, VRAM, and 3DAdaptive. It immediately crashed during the power test, failing to run even a full second, though it managed a few seconds on the GPU tests before stopping. All results showed the WHEA 18 error.

M
Maliwan99
Senior Member
346
12-09-2025, 11:33 PM
#8
Visit the web and look up using the specified service event IDs. The main aim is to gain more knowledge. The next goal is to apply what you read to understand your system better. The third aim is to explore possible solutions. For instance:
https://www.elevenforum.com/t/event-id-1...vice.9808/
Reference Post #5.
I’m not sure if the suggested fix applies to the issue or if it will help you. You might find other useful resources to identify the cause and resolve the problem. Avoid any software fixes that require downloading files. Also, keep away from Registry changes—they should be a last option.
M
Maliwan99
12-09-2025, 11:33 PM #8

Visit the web and look up using the specified service event IDs. The main aim is to gain more knowledge. The next goal is to apply what you read to understand your system better. The third aim is to explore possible solutions. For instance:
https://www.elevenforum.com/t/event-id-1...vice.9808/
Reference Post #5.
I’m not sure if the suggested fix applies to the issue or if it will help you. You might find other useful resources to identify the cause and resolve the problem. Avoid any software fixes that require downloading files. Also, keep away from Registry changes—they should be a last option.

M
Migun
Member
67
12-09-2025, 11:33 PM
#9
I'll remember this. Given the immediate crashes during PSU and GPU tests, it seems possible a faulty power supply could be delivering incorrect voltage to the motherboard and CPU. A defective wall outlet might also be contributing.
M
Migun
12-09-2025, 11:33 PM #9

I'll remember this. Given the immediate crashes during PSU and GPU tests, it seems possible a faulty power supply could be delivering incorrect voltage to the motherboard and CPU. A defective wall outlet might also be contributing.

P
PixelChick
Junior Member
5
12-09-2025, 11:33 PM
#10
Someone with the suitable skills and knowledge should check the wall socket. And maybe the electrical circuit connected to that outlet. PSU testing guide: https://www.lifewire.com/how-to-manually...er-2626158
P
PixelChick
12-09-2025, 11:33 PM #10

Someone with the suitable skills and knowledge should check the wall socket. And maybe the electrical circuit connected to that outlet. PSU testing guide: https://www.lifewire.com/how-to-manually...er-2626158

Pages (2): 1 2 Next