PC Shutting down

[Korgen]

I saw one other post here about someone having their entire computer shutdown in starfield. People were saying thermal or voltage related but for me it is very specific and can be reproduced 100% of the time. In the Akila City GalBank after getting the note for sam in the quest "The Empty Nest", leaving the bank will always cause my PC to power off unexpectedly. A temporary fix for me was to load a save before getting the note and leaving the bank by fast traveling instead of exiting through the door. I can then enter Akila City without any issue and have not ran into any unexpected shutdowns. This doesn't seem like a heat or power issue to me.

Edit: To the people just seeing this post now, no one has been able to figure out the issue yet, but we have confirmed that it is not a PSU or thermal issue. Several people have indicated they have these problems on AMD systems, one person even replaced their PSU with a more powerful one and is still encountering the issue. This is an issue with Starfield and only Starfield. My best guess right now is that it's something with AMD drivers.

[enderbeta]

Originally posted by Caduryn:

Originally posted by Carlos Spicywiener:

evga 850
Half the fun of owning a pc is the problem solving. Its one of the main reasons we have one

But.... evga 850 what? bronce, gold, platin....?

I see what this person is getting at. You have to really research power supplies. Some of these power supplies really cannot do what they claim especially the cheap ones.

I have Corsair AX1600i units in my 10900K and 5950X system because it is among the best. Even with power flickering in a bad storm this model psu keeps right on going. My HTPC with a RM850X will turn off in the same scenario.

[[☥] - CJ -]

Originally posted by enderbeta:

Originally posted by Caduryn:

But.... evga 850 what? bronce, gold, platin....?

I see what this person is getting at. You have to really research power supplies. Some of these power supplies really cannot do what they claim especially the cheap ones.

I have Corsair AX1600i units in my 10900K and 5950X system because it is among the best. Even with power flickering in a bad storm this model psu keeps right on going. My HTPC with a RM850X will turn off in the same scenario.

high quality caps possibly helps with that

[juskom95]

Originally posted by enderbeta:

Originally posted by juskom95:

Which means it would be a BSOD with an error code about the driver.

Not if the issue is the machine overheating. If it is overheating it will simply shutdown.

Correct, which I already addressed further up in the thread.

[=CrimsoN=]

Originally posted by Carlos Spicywiener:

Originally posted by =CrimsoN=:

It's just weird though because I've done benchmarks in Cinebench no problem. No BSOD. I did take a look at Windows Event Viewer after my PC restarted though. Seconds before the crash, there are several universal print issues reported:

The description for Event ID 1 from source Universal Print cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

Failed to get auth header with 0x8086000c
mcpmanagementservice.dll

The locale specific resource for the desired message is not present

starting to sound like a driver issue

That's definitely my thought as well. Like an actual full stress test isn't gonna give me hard restarts and the wattage usage from that far exceeds Starfield. Yet, it's the only game giving me crashing like this. I don't have the latest Nvidia drivers yet, but I've heard that they seem to cause more harm than good at this point. I'm on the previous drivers (537.34).

[juskom95]

Originally posted by =CrimsoN=:

Originally posted by juskom95:

1.
If you're seeing a hard power off, it is %100 the hardware. Starfield may be the only game that it is happening during, but what means is that Starfield is pushing a piece of hardware into a 'broken,' area and revealing something that is already broken.

Let's say your 1000w PSU has a faulty cable, or bad connection, but for most of your games the GPU doesn't pull enough wattage to cause the fault. Well, here comes Starfield and it pulls that %1 higher on the GPU, and causes the fault, which causes a hardware shutdown.

2.
If you're seeing a BSOD, then there is an error code and some other diagnostic data that will be on the screen. That data will help to identify where the issue is coming from.

It's just weird though because I've done benchmarks in Cinebench no problem. No BSOD. I did take a look at Windows Event Viewer after my PC restarted though. Seconds before the crash, there are several universal print issues reported:

Yeah, not that unusual for a benchmark to not overly stress a system but a game to do that.

The description for Event ID 1 from source Universal Print cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

Failed to get auth header with 0x8086000c
mcpmanagementservice.dll

The locale specific resource for the desired message is not present

That is a very strange error to be getting, as that .dll is related to printing?

Try a DISM repair:

https://www.windowscentral.com/how-use-dism-command-line-utility-repair-windows-10-image#:~:text=use%20these%20steps%3A-,Open%20Start.,Online%20%2FCleanup-Image%20%2FRestoreHealth

[=CrimsoN=]

Originally posted by juskom95:

Originally posted by =CrimsoN=:

It's just weird though because I've done benchmarks in Cinebench no problem. No BSOD. I did take a look at Windows Event Viewer after my PC restarted though. Seconds before the crash, there are several universal print issues reported:

Yeah, not that unusual for a benchmark to not overly stress a system but a game to do that.

The description for Event ID 1 from source Universal Print cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

Failed to get auth header with 0x8086000c
mcpmanagementservice.dll

The locale specific resource for the desired message is not present

That is a very strange error to be getting, as that .dll is related to printing?

Try a DISM repair:

https://www.windowscentral.com/how-use-dism-command-line-utility-repair-windows-10-image#:~:text=use%20these%20steps%3A-,Open%20Start.,Online%20%2FCleanup-Image%20%2FRestoreHealth

Yeah, what's weird is that I did that same command a few weeks ago! Also used chkdsk.

Idk man. Maybe I'm just cursed lol

[juskom95]

Originally posted by =CrimsoN=:

Originally posted by juskom95:

Yeah, not that unusual for a benchmark to not overly stress a system but a game to do that.


That is a very strange error to be getting, as that .dll is related to printing?

Try a DISM repair:

https://www.windowscentral.com/how-use-dism-command-line-utility-repair-windows-10-image#:~:text=use%20these%20steps%3A-,Open%20Start.,Online%20%2FCleanup-Image%20%2FRestoreHealth

Yeah, what's weird is that I did that same command a few weeks ago! Also used chkdsk.

Idk man. Maybe I'm just cursed lol

So, I had this same issue on my system about three months ago. I was getting two or three different BSOD's, and I went through and fixed the supposed issue behind them all.

I finally got fed up and bought a new NVME and rebuilt my system. I pulled it apart, gave it a good and thorough cleaning and did a clean and fresh install of Win10.

I did find some connectors that were working their way loose, which happens, and I had to re-seat my GPU after cleaning.

After that, I've been running without a problem, no BSOD's.
----------------
I believe it was a corrupt OS file that was refusing to repair with DISM, this can happen and it is frustrating, but starting over fresh eliminates that issue. Some of the dust/dirt was probably contributing to overheating as well, I just wasn't seeing that error code.

[=CrimsoN=]

Originally posted by juskom95:

Originally posted by =CrimsoN=:

Yeah, what's weird is that I did that same command a few weeks ago! Also used chkdsk.

Idk man. Maybe I'm just cursed lol

So, I had this same issue on my system about three months ago. I was getting two or three different BSOD's, and I went through and fixed the supposed issue behind them all.

I finally got fed up and bought a new NVME and rebuilt my system. I pulled it apart, gave it a good and thorough cleaning and did a clean and fresh install of Win10.

I did find some connectors that were working their way loose, which happens, and I had to re-seat my GPU after cleaning.

After that, I've been running without a problem, no BSOD's.
----------------
I believe it was a corrupt OS file that was refusing to repair with DISM, this can happen and it is frustrating, but starting over fresh eliminates that issue. Some of the dust/dirt was probably contributing to overheating as well, I just wasn't seeing that error code.

I might just try that. I am on Windows 11, which might not have been the brightest idea when I built this PC. Not the upgrade either. Like Windows 11 off a fresh disc.

[juskom95]

Originally posted by =CrimsoN=:

Originally posted by juskom95:

So, I had this same issue on my system about three months ago. I was getting two or three different BSOD's, and I went through and fixed the supposed issue behind them all.

I finally got fed up and bought a new NVME and rebuilt my system. I pulled it apart, gave it a good and thorough cleaning and did a clean and fresh install of Win10.

I did find some connectors that were working their way loose, which happens, and I had to re-seat my GPU after cleaning.

After that, I've been running without a problem, no BSOD's.
----------------
I believe it was a corrupt OS file that was refusing to repair with DISM, this can happen and it is frustrating, but starting over fresh eliminates that issue. Some of the dust/dirt was probably contributing to overheating as well, I just wasn't seeing that error code.

I might just try that. I am on Windows 11, which might not have been the brightest idea when I built this PC. Not the upgrade either. Like Windows 11 off a fresh disc.

In my experience, in-place OS upgrades are always trash. I would rather rebuild a whole server from scratch than bring over trash from an old OS and HOPE it works.

My BSOD frustrated me for a while, but when I got it running Word I realized it wasn't a GPU or CPU related issue.

I did use it as an excuse to upgrade a few other parts of my rig though, since it was opened.

[Carlos Spicywiener]

Originally posted by juskom95:

Originally posted by =CrimsoN=:

I might just try that. I am on Windows 11, which might not have been the brightest idea when I built this PC. Not the upgrade either. Like Windows 11 off a fresh disc.

In my experience, in-place OS upgrades are always trash. I would rather rebuild a whole server from scratch than bring over trash from an old OS and HOPE it works.

My BSOD frustrated me for a while, but when I got it running Word I realized it wasn't a GPU or CPU related issue.

I did use it as an excuse to upgrade a few other parts of my rig though, since it was opened.

I discovered this upgrading from 8 to 10. So many things wouldn't work right, like the windows key lol

[juskom95]

Originally posted by Carlos Spicywiener:

Originally posted by juskom95:

In my experience, in-place OS upgrades are always trash. I would rather rebuild a whole server from scratch than bring over trash from an old OS and HOPE it works.

My BSOD frustrated me for a while, but when I got it running Word I realized it wasn't a GPU or CPU related issue.

I did use it as an excuse to upgrade a few other parts of my rig though, since it was opened.

I discovered this upgrading from 8 to 10. So many things wouldn't work right, like the windows key lol

I was working a contract with a customer who had gone from 2012R2 -> 2016 -> 2019 . . . as I dug through the OS I found so much garbage, old DLL's etc . . . I finally said we are just going to rebuild this from scratch, that will be cheaper for you.

[Legion]

^^Looks over at rig stuffed with RGB fans and lights and a ancient i7 8700K and 3080 powered by a 750W PSU...don't you get any f****** ideas!

[Neozoon]

Wtf is wrong with you guys, this thread is like 99% arguing whether it's a software issue or a hardware issue. Only like 1% is actually troubleshooting.

[juskom95]

Originally posted by Neozoon:

Wtf is wrong with you guys, this thread is like 99% arguing whether it's a software issue or a hardware issue. Only like 1% is actually troubleshooting.

I provided several steps to try, repeatedly only to have someone just state: "No, that's not it, it can't be!"

Part of troubleshooting is ruling things out, and determining a most likely cause based on the information given.

Just because someone dislikes the answer, does not make it wrong.

[[☥] - CJ -]

Originally posted by Legion:

^^Looks over at rig stuffed with RGB fans and lights and a ancient i7 8700K and 3080 powered by a 750W PSU...don't you get any f****** ideas!

a 95W CPU from 2017 isnt ancient