Question Crash to black screen but only with some games

[Akhronik]

Hello everyone, first time poster but long time lurker!

I apologise in advance for the lengthy post, I'll try to give as many details as I can without adding unnecessary stuff. So, as the title says, my pc crashes to a black screen, with fans spinning up and it being still turned on, but only with some games. This is what happens in sequence:

1. I start a game (Veilguard and FFXVI are the major culprits) and, usually in about 10/20 minutes, pc crashes;
2. Goes to black screen, fans spin up to max. LEDs are still on (both in case and peripherals);
3. Stays like that for about 3/4 seconds, then the Debug monitor on my mobo (Asrock b550 Steel Legend, I'll post my full specs at the bottom) starts giving codes. In sequence, they are: 36 (PEI_CPU_SMM_INIT), 98 (DXE_CON_IN_CONNECT), 99 (DXE_SIO_INIT) and finally d6 (DXE_NO_CON_OUT).
4. At the same time the beeper on my motherboard makes 5 short beeps followed by 1 short beep. I couldn't find it in the manual, but by looking around online it looks like this means that GPU is not being recognised (for any reason);
5. If I press the Reset button, the it does exactly the same thing, starting from step 3. If I force the shut down with the Power button it instead starts normally;

As I said, this only happens with some games which are usually quite GPU intensive, however other games, which are just as demanding (like Modern Warfare 3 and Black Ops 6) work with *mostly* no issues. I say mostly because the issue has presented itself with these games, although extremely rarely.

Below are my specs:

CPU: AMD Ryzen 7 5800X 3.8 GHz 8-Core Processor
CPU Cooler: Thermalright Phantom Spirit 120 SE 66.17 CFM CPU Cooler
Motherboard: ASRock B550 Steel Legend ATX AM4 Motherboard
Memory: Corsair Vengeance RGB Pro 32 GB (2 x 16 GB) DDR4-3600 CL18 Memory
Storage: Kingston A400 240 GB 2.5" Solid State Drive (this is where the OS is installed and most default apps/software)
Storage: Western Digital Blue SN580 1 TB M.2-2280 PCIe 4.0 X4 NVME Solid State Drive
Storage: Western Digital Black 2 TB 3.5" 7200 RPM Internal Hard Drive
Video Card: XFX Speedster SWFT 319 Radeon RX 6800 16 GB Video Card
Case: Thermaltake CTE C700 Air Chassis
Power Supply: Corsair RM650 (2023) 650 W 80+ Gold Certified Fully Modular ATX Power Supply
Operating System: Microsoft Windows 10 Pro OEM - DVD 64-bit


Now for the lengthy part where I tell you what I've already tried:

• Every single part in this computer, Storage aside, is new. I've been having this issue for a while, but only recently it has started getting worse and worse. Mobo, CPU, CPU Cooler, GPU, RAM, Case and PSU have all been replaced in 2024. This issue was present on my system, though not as serious, long before that (started at least in 2023 that I can remember);
• I've already checked the connections from PSU to all the other parts and they look and feel solid. I've redone them just in case;
• Tried running with just one stick of RAM;
• I've tried running my RAM both with and without XMP enabled;
• I've kept an eye on the temps (I have two monitors) while gaming and the temperatures are well within normal measurements;
• I've tried creating minidumps after each crash, and yet they are never generated. I've checked that the settings are correct after each crash and they are;
• I've replaced the paste on the CPU cooler more than once, both when I bought the new cooler (which happened first) and when I replaced the CPU itself;
• I've considered the possibility that the power draw might be too much as in theory this system should have a Wattage of ~500. My previous PSU was a 750 and I still had the same issue (also, before replacing the parts, the wattage was lower, around 450);
• I've performed a Windows reset but not yet a clean install with USB (which in hindsight I probably should have done anyway);
• I've performed a clean driver install with DDU when I swapped the GPU (from Nvidia RTX 2060 to what I have now);
• It has very rarely happened that, after a crash, the GPU drivers would get messed up. The GPU would show up in Speccy, but if I tried to open any game it would tell me that there's no graphics card (which seems weird since, AFAIK, my CPU does not have an integrated GPU and shouldn't even let the system boot without one). If I tried opening AMD Adrenalin it would then tell me that the software version is incompatible with the installed drivers. This usually requires a full reinstall as simply rebooting wouldn't fix it;

I'm sure I've tried other stuff too, but at this point it's hard to remember. I don't even know how to pinpoint the issue and I'm frankly feeling quite defeated. I'll accept any suggestions and thanks in advance to anyone who's taken the time to read my wall of text!

 

[Roland Of Gilead]

Hey there,

Are you using the cables that came with PSU replacement, or the older cables already in place?

What bios are you currently running?

Have you tried sfc scannow/DISM? : https://support.microsoft.com/en-us...em-files-79aa86cb-ca52-166a-92a3-966e85d4094e - run both.

650w is a bit on the light side. Are you overclocking the CPU or GPU? Min for that GPU is 650w.

 

[Akhronik]

Hey Roland !

I'm only using the cables that came with the new PSU, lead cable included (I replaced it just for good measure, you never know);

Currently running BIOS 2.40 (could do with an update, I imagine) This made me notice that I didn't list the mobo as one of the parts that I've already replaced (as the crash happened with the old mobo as well), I've edited the original post to reflect this, thanks!

I'm not overclocking anything as I feel it's way outside my comfort zone. I am running RAMs with XMP, but that's supposed to be "manufacturer approved", right?

I've just ran DISM and sfc scannow. DISM just said "The restore operation completed successfully" and sfc scannow had "Windows Resource Protection did not find any integrity violations."

 

[Roland Of Gilead]

Hey Roland !

I'm only using the cables that came with the new PSU, lead cable included (I replaced it just for good measure, you never know);

Currently running BIOS 2.40 (could do with an update, I imagine) This made me notice that I didn't list the mobo as one of the parts that I've already replaced (as the crash happened with the old mobo as well), I've edited the original post to reflect this, thanks!

I'm not overclocking anything as I feel it's way outside my comfort zone. I am running RAMs with XMP, but that's supposed to be "manufacturer approved", right?

I've just ran DISM and sfc scannow. DISM just said "The restore operation completed successfully" and sfc scannow had "Windows Resource Protection did not find any integrity violations."

Okay, so 2.40 is quite old. From Nov 2022. This might be a good time to update fully to rule out any bugs or fixes that need applying on the hardware level. When updating the bios you will need to clear CMOS after the update. : https://www.asrockrack.com/support/faq.de.asp?id=46

XMP is fine. Are you running PBO on the CPU? PBO is the Precision Boost Override functionality. It's like built in overclocking for the CPU, but it does so within spec.

 

[Akhronik]

Okay, so 2.40 is quite old. From Nov 2022. This might be a good time to update fully to rule out any bugs or fixes that need applying on the hardware level. When updating the bios you will need to clear CMOS after the update. : https://www.asrockrack.com/support/faq.de.asp?id=46

XMP is fine. Are you running PBO on the CPU? PBO is the Precision Boost Override functionality. It's like built in overclocking for the CPU, but it does so within spec.

I'll definitely update as soon as I have some time and energy to do it, probably tomorrow morning.

As for PBO, I noticed in the BIOS that it's set to Auto. Should it be enabled or disabled? Or should I leave it as it is?