PC black screen restarting randomly in lower demanding games

[Guidosaur]

Hey all, I'm trying to do some proper diagnosis before I fire the parts cannon and start throwing money at this problem. I felt bad pestering my IT friend, and didn't want to create a Reddit account so I came here instead lol.

My system:
4000d airflow case (4 120mm fans, 2 140mm fans)
SeaSonic M12II 620w Bronze PSU
MSI X470 Gaming Plus Max motherboard (newest FW: 7B79vHG)
Ryzen 5 3600 with NZXT Z63 AIO
Corsair Vengeance LPX 16GB (2 X 8GB) DDR4 3600 MHz 
2 1TB m.2 SSD's and a 500GB 2.5 PNY SSD
Gigabyte Vision 3070 TI (current FW)
Windows 11 Home
2 1080p monitors
If I missed anything else just ask. 

Alright so the first iteration of my setup I had the Ryzen 5 3600, same MSI motherboard, and an EVGA GTX 770 on Windows 10 Home. Never had any random crashes. It was reliable, just super old and slow, and I finally got my hands on a 3070 TI after the scalping started to slow down. I knew my power supply was pushing its limits, but I barely have any accessories in my system, I only game at 1080p and haven't done any overclocking past stock. I'm able to play Elden Ring, Monster Hunter World, RE2, Black Mesa all maxed out settings and nothing would really go past 55°C. I don't have any RTX games, so these are the most demanding games I have unfortunately lol. Can leave the games running for hours and never have any problems anywhere.

Recently I've been playing older games from my backlog and started running into problems with hitting a black screen and the PC rebooting. I would try to play Dark Souls 3, and after like 10-15 minutes it would just give me the black screen reboot. No mods or anything crazy on the game, just max settings 1080p. Windows Event Viewer shows Event ID 41, but that isn't super conclusive. After the crashes, Windows would show that one of my m.2 drives was missing (the OS is on my 2.5" SSD) until I manually restart the system again myself and then everything would be back to normal. I thought my install of Windows might be corrupted, so I upgraded to Windows 11 on the 2.5" drive, but it still currently has random crashes with the disappearing drive on restart. The drive that disappears isn't the one where any of my games are installed either. All 3 drives always report properly in BIOS. I can play new more demanding games fine, but random other games cause the black screen crashes. Most recently I have been playing Zelda: BOTW on CEMU emulator with unlocked framerate and it will occasionally crash the PC after just a few minutes. However, yesterday I left the game running standing in the open world for hours on end and it was fine. In both cases the other background programs running are all constants. I can completely replicate the problem I've found with Silent Hill 4 on PCSX2 for god knows what reason. I load up my save, and as soon as this area loads the system hard resets. Using OpenGL and rendering the game internally at 1440p. The game played fine for days before, but now it does this every time. 

I downloaded HWiNFO64 and watched my CPU and GPU wattage during Zelda BOTW, and it was averaging around 150w GPU, 50w CPU, and there weren't really any big spikes. Nowhere near 620w. I recorded a datalog of me playing BOTW and then Silent Hill 4 on PCSX2 when it crashed. It was maybe 150w combined on PCSX2 before it crashed. I've attached the data log, and then a more condensed version of the log with colored columns. Maybe you all can see something that I'm not. I get that my PSU is old and doesn't have much headroom, but I'm not really pushing my hardware either. I didn't want to replace it and re-run all my cables until I could afford to switch to AM5. I'm not sure if I have a problem with it, maybe my motherboard or RAM. Sorry for the wall of text, but I was trying to be thorough. Feel free to ask if I missed anything. 

botw then silent hill 4 till crash.CSV edited botw then silent hill 4 till crash.CSV

Edited October 16, 2022 by Guidosaur
forgot the Event Viewer ID

[Guidosaur]

Bump because this happened again today. 

TL:DR- PC crashing to black screen and restarting with a missing drive until manually restarting again. Only pulling about 200w between CPU and GPU when the system has been crashing according to HWiNFO64. 

[Howitz]

i have the same issue, and my motherboard is the msi x570 gaming plus.

[Howitz]

 here is my thread, same issue and just swapped to a 1000 watt power supply because we were suspecting it ran out of power. same thing happened within 5 minutes of playing apex. black screen for 2-3 seconds and reboot.

[Guidosaur]

21 minutes ago, Howitz said:

 here is my thread, same issue and just swapped to a 1000 watt power supply because we were suspecting it ran out of power. same thing happened within 5 minutes of playing apex. black screen for 2-3 seconds and reboot.

I was scared I was going to get roasted for using an old lower power PSU  I'll check out your thread. I really didn't want to spend $100, re-run all my cables, and then still have the same crashes. I really wish I didn't have any products from Gigabyte or MSI in my system, but you live and learn I guess. Now that I'm on Win11 I'm scared that I can't go too far back in BIOS drivers either. 

[Howitz]

I tried both windows 10 and 11 and it restarts on both. Right now i'm trying to install win10 os on my old 1tb ssd and play apex from there to single out my 2tb ssd. If it doesnt crash from running everything from the hdd instead of the ssd then it means it's either the motherbpard or the ssd since the ssd is plugged in the mobo slot.

[Guidosaur]

58 minutes ago, Howitz said:

I tried both windows 10 and 11 and it restarts on both. Right now i'm trying to install win10 os on my old 1tb ssd and play apex from there to single out my 2tb ssd. If it doesnt crash from running everything from the hdd instead of the ssd then it means it's either the motherbpard or the ssd since the ssd is plugged in the mobo slot.

I think all I've got left to test is my RAM. Not sure how I can confirm if its my motherboard or not. I want to say it can't be my SSD that vanishes because the games that have been crashing my system aren't even on said drive. It's been doing this to me across multple BIOS, GPU, and even Ryzen drivers. I've tried updating them all and still get the random crashes.

[Howitz]

21 minutes ago, Guidosaur said:

I think all I've got left to test is my RAM. Not sure how I can confirm if its my motherboard or not. I want to say it can't be my SSD that vanishes because the games that have been crashing my system aren't even on said drive. It's been doing this to me across multple BIOS, GPU, and even Ryzen drivers. I've tried updating them all and still get the random crashes.

Pretty sure its not my ram since ive been using those 2 sticks for months without issues after having replaced a faulty ram kit that was causing a crash issue. So my ram is good at least.

[BlackManINC]

 

@Guidosaur Did your crashes only occur after updating to Windows 11? And yeah, I'm not a power supply guru, but I'd say you are indeed pushing your luck with it just considering your GPU's manufacturer recommends no less than a 750 Watt power supply. 

[Guidosaur]

46 minutes ago, BlackManINC said:

 

@Guidosaur Did your crashes only occur after updating to Windows 11? And yeah, I'm not a power supply guru, but I'd say you are indeed pushing your luck with it just considering your GPU's manufacturer recommends no less than a 750 Watt power supply. 

yeah, had the problem on Win 10 as well. I know I'm pushing the limit of my PSU on paper, but I'm only gaming at 1080p, and have like no peripherals besides my AIO. 2 of my case fans I set to not even turn on unless things get to 60C. Looking at my data log, and watching HwINFO64, the most I've ever been pulling is like 250w tops. Howitz having similar issues with a 1000w PSU is making me think it might be Mobo related. 

[BlackManINC]

1 hour ago, Guidosaur said:

yeah, had the problem on Win 10 as well. I know I'm pushing the limit of my PSU on paper, but I'm only gaming at 1080p, and have like no peripherals besides my AIO. 2 of my case fans I set to not even turn on unless things get to 60C. Looking at my data log, and watching HwINFO64, the most I've ever been pulling is like 250w tops. Howitz having similar issues with a 1000w PSU is making me think it might be Mobo related. 

Have you tried using any older CPU's or GPU? Maybe the new GPU is a dud and it might have to be RMA'ed like I had to do mines. How long have you had the power supply? I'd also try a new surge protecting power strip and power cable to be sure. 

[Guidosaur]

Alright I found another way to reproduce the problem. Tried to let my sister play Poppy's Playtime and the system crashed immediately. Tried to play with and without Auto OC turned on in Ryzen Master and it crashed both times. I took a datalog in HwInfo64 of the crash. Highest CPU wattage was like 50w and the highest on GPU was about 100w. 12v, 3.3v, and 5v rail didn't have any noticeable deviation in voltage. There's no way this game should be tanking my system, but every time I click settings in the menu it crashes. In Silent Hill 4 on emulator it's always when I open a save and go to the overworld that I get a crash. Windows Event Viewer just says ID 41 so it isn't a ton of help. The HwInfo64 log is less than a minute and a half long, maybe some fresh eyes could help me see something wrong?

Thanks

ppp.CSV

[BlackManINC]

On 10/20/2022 at 6:50 PM, Guidosaur said:

Alright I found another way to reproduce the problem. Tried to let my sister play Poppy's Playtime and the system crashed immediately. Tried to play with and without Auto OC turned on in Ryzen Master and it crashed both times. I took a datalog in HwInfo64 of the crash. Highest CPU wattage was like 50w and the highest on GPU was about 100w. 12v, 3.3v, and 5v rail didn't have any noticeable deviation in voltage. There's no way this game should be tanking my system, but every time I click settings in the menu it crashes. In Silent Hill 4 on emulator it's always when I open a save and go to the overworld that I get a crash. Windows Event Viewer just says ID 41 so it isn't a ton of help. The HwInfo64 log is less than a minute and a half long, maybe some fresh eyes could help me see something wrong?

Thanks

ppp.CSV 60.14 kB · 0 downloads

And none of this occurs when just web surfing for instance? What apps are you running in the background? Anti virus apps?

[Guidosaur]

On 10/21/2022 at 10:28 PM, BlackManINC said:

And none of this occurs when just web surfing for instance? What apps are you running in the background? Anti virus apps?

Nope, it's all fine for just basic productivity stuff. Only background apps i have while gaming are some tabs in Brave Browser and that github FanControl program. The other day I tried running User Benchmark (so I could crosspost to reddit: it's a requirement there) and the system crashed when it got to GPU. Tried Furmark and as soon as I hit start it would crash immediately. Then I realized that RE2 Remake has RTX, so I ran that game on max settings, RTX on, 1080p with an unlocked framerate. Between the CPU and GPU I was pulling around 400w, which is way more than the 200ish that I normally pull on emulators, and everything was under 70c stable. I'm thinking maybe some weird shit is going on with drivers but I'm at a loss. I might be able to steal my girlfriend's 5700XT for a day and see if the system is still unstable. 

[BlackManINC]

1 hour ago, Guidosaur said:

Nope, it's all fine for just basic productivity stuff. Only background apps i have while gaming are some tabs in Brave Browser and that github FanControl program. The other day I tried running User Benchmark (so I could crosspost to reddit: it's a requirement there) and the system crashed when it got to GPU. Tried Furmark and as soon as I hit start it would crash immediately. Then I realized that RE2 Remake has RTX, so I ran that game on max settings, RTX on, 1080p with an unlocked framerate. Between the CPU and GPU I was pulling around 400w, which is way more than the 200ish that I normally pull on emulators, and everything was under 70c stable. I'm thinking maybe some weird shit is going on with drivers but I'm at a loss. I might be able to steal my girlfriend's 5700XT for a day and see if the system is still unstable. 

I would try it without the Brave Browser and that Fan Control program running, especially the latter. I would also make sure Brave Browser is up to date as well. And, you're not running any anti virus app besides Defender? 

[Guidosaur]

On 10/23/2022 at 3:13 PM, BlackManINC said:

I would try it without the Brave Browser and that Fan Control program running, especially the latter. I would also make sure Brave Browser is up to date as well. And, you're not running any anti virus app besides Defender? 

Alright, just confirmed the system still crashes with no fan control program or browser running. Windows was running for maybe 30 seconds, ran fur mark and insta crash. Should I just DDU or something?