Occasionally shutting down with VGA debug indicator on
Occasionally shutting down with VGA debug indicator on
The system experiences intermittent problems, switching between crashes and stable operation. There are no overheating concerns during gaming, though temperatures never exceeded 75°C with monitoring tools active. The GPU occasionally overruns, causing random shutdowns, and sometimes restarts itself. VGA issues appear on the motherboard after reboots. Most crashes occur unpredictably, lasting minutes or weeks apart. Replacing the GPU usually resolves the problem temporarily. Firmware and drivers are up to date. No temperature spikes were recorded. This behavior is new compared to previous GPUs, and I’m seeking advice to understand and fix it.
You seem to have changed from using Intel to AMD. A complete reset of your operating system might be beneficial. Ensure your BIOS is current and install the latest chipset drivers for your processor. If that fails, consider adding a small support for the GPU—this can help prevent random crashes. Check for any sag in the setup, even minor issues can cause problems. When you try it, gently tap the GPU; if it crashes, it likely points to a loose connection due to sag. I found an issue with a 2070ti that was too much sag, leading to a loose connection. If this doesn’t resolve it, you might need to look at the AMD support page for your motherboard.
It's funny how you bring up those things. Switching to AMD brought a new M.2 drive and a fresh Windows install. I had similar concerns with the GPU sag, so I bought a cheap support piece that worked well. It arrived about 1.5 months ago without any problems until today when another crash happened. I’ll try the little tap on the GPU, though—I might have damaged the PCIe slot if I hadn’t had the support. From what I saw, the support was still functioning properly. I’ll double-check my chipset drivers later. Good to know you have the latest M.2 drivers from Asus. Thanks!
Another approach could be disabling GPU support completely. Allow it to hang without any GPU functionality. If it starts up, gently move the GPU slightly. If your system freezes, it seems you've found the solution. It might just be a loose connection. Experiment with adjusting the GPU support settings—maybe a tiny tweak will make it fit better. Or it could be outdated. I'm currently using this setup to hold my 3080 graphics card in place. A screwgun drill bit works great, thankfully. It's quite short, around a millimeter or so, which keeps the GPU extremely secure and tight with no movement. The cardboard wind chute also has its own unique story. Edited December 20, 2023 by Madmaximus01
I attempted to move the GPU without any support, but nothing changed. It’s been stable so far, though it’s likely to fail again soon. When it crashes, all fans spin at maximum speed across the whole system. I don’t remember seeing temperatures exceed 75°C. This makes me think the card’s base clock might be too high, and reducing power limits could help.
The system may show signs of dissatisfaction with power delivery. Look for loose connections or cables—gently move the main power cable on the motherboard and the CPU cable. Ensure the motherboard’s power cable is firmly attached. Test the power connection to your GPU by wiggling it. Verify that all PSU cables are securely inserted and not slipping. Check your HDD or SSD connectors for solid SATA power connections. Confirm the PSU power cable is fully connected to both the PSU and the wall; sometimes minor movements can shift it. After these checks, download FurMark and run a test on the GPU. If it crashes during testing, it could point to a power-related problem with the GPU. Ask someone else with a PC if they experience similar issues—they might help confirm the cause. Review your BIOS settings for compatibility mode (CSM) status and whether the resizable bar is enabled or disabled. If CSM is active, ensure the NVIDIA settings reflect that setting. A black screen error saying “VGA card not supported” during a reboot often indicates a firmware mismatch. Make sure you’ve cleared the CMOS before proceeding.
I verified all my connections, they’re functioning properly. Executed commands and resolved several problems. Disabled resizable BAR since the 1080ti isn’t compatible, kept CSM enabled. Discovered a chipset update I missed, which stopped the random black screen crashes. However, I’m now experiencing a DirectX crash every 45 minutes to two hours into gaming (error 0x887A0005). I captured a screenshot of the pop-up but it didn’t save. Since then I’ve researched this error and it looks like a new GPU is the only solution, which is disappointing. Most advice I’ve tried matches what I’ve already attempted to fix.