There is a rare cold boot GPU error sound and a green display before starting Windows.
There is a rare cold boot GPU error sound and a green display before starting Windows.
OS: Windows 10
Mobo: ASRock AM5 B650M PG Riptide Wi-Fi
GPU: RTX 2060 Super Founders Edition
CPU: Ryzen 5 7600
RAM: 2 x 8GB DDR5
I experienced this boot problem twice recently. On the first day, I heard five quick beeps, likely indicating a GPU error on ASRock AM5 boards, followed by a brief full green display before the Windows login screen appeared. The POST screen and pre-login loading screen never showed up. After restarting, everything booted normally without any beeps. This happened after replacing the GPU a few weeks prior, though it didn’t occur with my previous GPU (GTX 970). It seems to be a recurring issue that might be fixable.
If you restart and try again, the system loads correctly without any sounds. The problem appears to be isolated to this GPU model. Could be something I can resolve?
I didn't specify the details of your PSU, make, model, or age. The BIOS version on your ASRock motherboard is unknown. Did you just acquire the GPU a few weeks ago? It wasn't showing this issue until recently. Before the RTX 2060 Super FE, what were you using? Was it typical? This seems unusual. Could a corrupted BIOS, a faulty motherboard, or a failing PSU be the cause?
This unit appears to be a high-quality PSU purchase. The BIOS version is reported as 3.15 from late 2024. The latest BIOS is P3.50, so ensure your motherboard drivers are updated and flash the BIOS to version 3.50. Clear CMOS and enable X.M.P/E.X.P.O in BIOS to check for improvements. Previously, you had a GTX 970. Did you use DDU in Safe Mode to remove all GPU drivers (Intel, AMD, Nvidia) and then reboot? Installed the latest driver from Nvidia's support site via an elevated command, such as right-clicking the installer and selecting Run as Administrator?
I updated the BIOS, but the driver installs automatically as admin. This shouldn't matter much since the issue is mainly about POST. I ran DDU in safe mode and installed the latest version.
It might be useful: I adjust my RAM settings and keep Memory Context Restore off to ensure full memory training on every boot. Sometimes, if I power on the PC when the room temperature is slightly different, I notice a RAM beep code during the first training attempt (since the beep speaker is installed). The BIOS then restarts to retrain with new timing or impedance values. I think this could be due to temperature or humidity affecting the electrical properties of my DDR5 RAM, combined with my aggressive tuning, causing it to need re-training. Despite passing memtest86 multiple times, it still manages to pass the second training attempt if a beep code appears from the RAM.
For the GPU, it doesn't reboot on a five-beep signal; it just continues into Windows but with poor video performance until the user login screen appears. It would help if the BIOS failed the first training attempt and then restarted to try again with new settings. I also have a USB-powered light that only activates after a cold boot or a failed training restart, not during a Windows reboot.
I bought the beep speaker because I sometimes saw the USB light turn on twice after a cold boot, suggesting the system was restarting by itself. This made me think the speaker could assist in diagnosing the problem. I also suspect that even with default RAM settings, many users encounter similar issues with DDR5 RAM training, but they don't realize it until they lack proper indicators to detect BIOS reboots after a failed attempt.
I’m curious whether temporary temperature-related electrical changes might be impacting the GPU or PCIe pins on the CPU, as seems to happen with the RAM. Or perhaps there’s capacitor aging in the GPU.
I’ll monitor the situation with the new BIOS and provide an update later.
No GPU beep codes have appeared since the BIOS update, and the weather has been more consistent, reducing drastic temperature changes.