Possible display driver issue, input appreciated
Possible display driver issue, input appreciated
Hi there,
I recently started experiencing a weird issue with my old rig. I cannot pinpoint the exact moment when it started, but I assume it was after a Windows update (Win10 24H2). It started recently when I woke the PC from sleep (I always put it to sleep, I
never
shut it down). By the looks of it, I believe it’s a display driver failing and recovering. I used to do a hard reset a couple of times at first, since I taught the PC hanged, but then I decided to wait and see if it will recover and it does. No issues like overheating or when the PC is under stress (like playing games or doing other CPU or GPU-intensive tasks; all temps are within limits no matter how long I tested it and this does not happen when the PC is under load or used for browsing, working, playing, etc., just when waking it up from sleep or booting it up). I don’t get the message that the DD failed and recovered and there are no logs of any problems anywhere to be found. I updated the driver (did not use DDU this time), but the issue persists.
I don’t think this is particularly useful info, but the OS was installed back in 2017 (Win7) and was in-place upgraded to Win10 a couple of years after that, so it’s already 8 years old, and I wanted to do another in-place upgrade to Win11, but started experiencing this issue. I am tempted now to do a clean install of Win11, but I am dreading it for obvious reasons (having to reinstall and reconfigure all the software I regularly use).
Any other suggestions where to look or what to check (other than using the DDU, since this is my next step) or how to resolve this issue?
Welcome to the forums, newcomer!
First thing I'd ask you to do is open up Device Manager, under Display Adapter see if it's flagged with a yellow exclamation mark and an error code 43. Following that, use GPU-Z and see if one or more fields on the first tab shows up with zero(0)'s. If the answer is yes to both, your GPU's conked out.
I don’t think this is particularly useful info, but the OS was installed back in 2017 (Win7) and was in-place upgraded to Win10 a couple of years after that, so it’s already 8 years old, and I wanted to do another in-place upgrade to Win11, but started experiencing this issue.
Users who have upgraded to Windows 10 using the internal upgrade path were advised to reinstall Windows 10 to root out any OS corruptions and subsequent driver conflicts. I doubt you'd be able to migrate to Windows 11 if your system was from the Windows 7 era(assuming it's Haswell).
Please list the specs to your build like so:
CPU:
CPU cooler:
Motherboard:
Ram:
SSD/HDD:
GPU:
PSU:
Chassis:
OS:
Monitor:
include the age of the PSU apart from it's make and model. BIOS version for your motherboard at this moment of time.
Just a few notes to share......
I think when you say hard reset means pressing and holding the power button, that’s definitely not okay... eventually something will get damaged and you might not be able to reach the recovery option at startup, which could cause trouble. Before attempting a hard reset, try opening Task Manager and using the circle in the lower right corner to restart your system.
If you think a Windows update is the issue, go to Add/Remove Programs and look for a link to uninstall updates. The challenge with updates is that since Windows 10 was released, hiding updates isn’t an option anymore—third-party tools claim to do this, but they often fail because Windows doesn’t recognize your changes. Unless you’re sure, avoid updating the driver.
Also, I don’t know what GPU you have. If you’ve upgraded to a recent 40 or 50 series card, the driver might be the problem since NVidia is focusing heavily on AI features and hasn’t fixed many issues. Many forums suggest not updating the driver for these cards.
Damaged graphics card. Likely cracked solder joints in VRAM chips. When the graphics card warms up after turning on the PC, these broken connections re-form.
Thank you for your response. The welcome and your comments were appreciated.
No yellow exclamation marks appear in the Device Manager, and GPU-Z displays accurate readings.
During a hard reset, I simply pressed the reset button on the chassis. It would take some time to restart, as if performing a power cycle, and the problem appeared resolved once I reached the login screen. The concern seems to relate to what SkyNetRising is recommending, but it might lead to complications when the PC is actively used. Everything appears normal after logging into Windows.
This is my older system, yet I still perform a significant amount of work on it. Here are the details you asked for:
CPU: i7-7820
GPU cooler: CoolerMaster 240 AIO
RAM: Kingston KHX2400C15/16G (2 sticks, total 32GB)
Storage: Samsung 980 Pro 1TB
GPU: 1080Ti Posseidon (air-cooled)
Power supply: CoolerMaster V850 (RS-850-AFBA-G1, model from 2017)
Case: Gigabyte (exact model unknown)
Operating system: Win10 24H2
BIOS version: v3701
Display: LG 27GL850
I plan to perform a clean installation of Windows 11 over the weekend to check if it fixes the issue. I suspect only certain files may have become corrupted over time. The board supports TPM 2.0, which should allow Windows 11 installation.
Edit: I also intended to upgrade the BIOS to the latest release to resolve the recent Intel ME FW problem. That could be another option.
Edit 2: I added the precise PSU model to the description.