Question Yet another... PC crashing to black screen while gaming

[daytondoes]

I purchased a prebuilt PC one year ago. I've been working through this problem ever since. Specs are as follows:

CPU: Intel Core i9-11900K

Motherboard: MSI MPG Z490

Ram: 2 T-Force(?) 16 GB DDR4

SSD: Crucial MX500 1TB

HDD: WDC WD40EZAZ-19SF3B0 4TB

GPU: Nvidia RTX 3070-Ti

PSU: Aorus 850w 80+ Gold

Chassis: Coolman Aurora Gaming Case

OS: Win10

Monitor: 1 really old Acer, 1 YGT as main display, 1 Wacom Cintiq

The crash: The screen will suddenly go dark. Monitors will display no input. Lights in the case stay on, but I can't do anything until I hard reset the system. I've let it stay this way for hours out of curiosity, but the screens will stay dark. Before changing out the PSU/RAM a couple days ago, I would hear the sound of the game/video for a few more seconds before it starts to loop the last half second. Since replacing some parts the crashes have been 100% silent which seems better, but I'd really prefer a system that didn't fall apart in the middle of a game.

The crashes generally happen while gaming, though the game doesn't need to push the limits of the rig... I had a crash a few weeks ago while streaming a Jackbox game.

I've upgraded the PSU and changed out the RAM (despite it passing memtest) hoping that would solve the issue. It hasn't I crashed again today while playing Apex and watching YouTube.

The GPU has passed stress tests with Furmark, and the temps don't get above 85c. Generally idles around 35-45c, but that's normal for the Philippines.

I've seen some suggestions that it might be a conflict with the audio driver. Well, I disable Realtek Audio in hopes that might help since it does seem to happen more often during gaming+YouTube (or Discord call for Jackbox)

There was also a person that fixed their issue by toggling the C-State settings in the BIOS. I tried that and almost thought it worked. Until a few days later. Another crash mid-game.

That doesn't seem to have fixed the issue however. All parts are under warranty for another 2 months or so, but I'm running out of time to get the issue figured out.

I have no way of testing for transient voltage spikes from the GPU that might be knocking the PSU out. I did see that suggested in another thread. I hoped upgrading from 750w to 850w would fix it. But I'm still struggling.

Please bear in mind, I stream full-time for work so I have a LOT of peripherals. StreamDeck XL, Blue Yeti mic, hand tracker, eye tracker, gamepad. Even unplugging the unneeded ones while I'm working doesn't seem to work.

Finally, I did check the event viewer. Event 41- Kernel Power which doesn't really tell me much other than that it crashed.

EventData

BugcheckCode

278

BugcheckParameter1

0xffffa20c078a4010

BugcheckParameter2

0xfffff80327701b20

BugcheckParameter3

0xffffffffc000009a

BugcheckParameter4

0x4

SleepInProgress

0

PowerButtonTimestamp

0

BootAppStatus

0

Checkpoint

0

ConnectedStandbyInProgress

false

SystemSleepTransitionsToOn

9

CsEntryScenarioInstanceId

0

BugcheckInfoFromEFI

false

CheckpointStatus

0

CsEntryScenarioInstanceIdV2

0

LongPowerButtonPressDetected

false

At this point I'd prefer a BSOD or a dead GPU or something... so I could at least tell what the problem is and get it fixed. The folks who built it are very helpful with swapping parts around, but they seem to be having just as much trouble finding the issue as I am. This forum has been very helpful in the past, so I thought now that I'm just about out of options it's finally time to try my luck. Perhaps someone smarter or more experienced than me can get down to the root of the problem and I can have a PC that functions 100% reliably. The crashes aren't constant, but it really hurts during a stream to say "LOL BRB I GUESS" and watch the viewer numbers plummet... just because my rig is having issues. Even outside of the work downsides, it's just plain inconvenient. Any help that you guys can offer would be massively appreciated. Thank you very much in advance!

 

[Ralston18]

Also take a look in Reliability History.

Much more user friendly than Event Viewer plus the time line format can be very revealing.

Look for anything being captured just before or at the time of the crashes. Or some pattern of any sort of entries.

Noted that you upgraded the PSU and changed out RAM.

Were you also able to check that all other connections, cards, jumpers, and case connections are full and firmly in place?

Do you have another known working (no crashes) TV or monitor to test with? Just as a matter of elimination?

 

[daytondoes]

Also take a look in Reliability History.

Much more user friendly than Event Viewer plus the time line format can be very revealing.

Look for anything being captured just before or at the time of the crashes. Or some pattern of any sort of entries.

I was unaware that existed. I'm not exceedingly savvy, but I will go through it now and see if I can spot anything unusual.

Were you also able to check that all other connections, cards, jumpers, and case connections are full and firmly in place?

Can confirm that everything is locked firmly in place. Even got a GPU stand in case the card was imperceptibly sagging. No dice there. This was one of the first suspicions I had. Checked again while installing the PSU. I feel confident that this isn't the issue.

Do you have another known working (no crashes) TV or monitor to test with? Just as a matter of elimination?

I can certainly try to hook the rig up to my office TV. My wife also has a build coming in a couple days. I can borrow her monitor.

Will update on how it goes with Reliability History and monitor swap ASAP. Thanks!

 

[daytondoes]

Reliability history is a really cool tool. I have pasted the "Windows stopped working" errors, and I will attach the images. I can attach the DMP file if needed but it is pretty big. Unsure how to trim that down to size before sending it... Or if it's even needed. It's all greek to me, but I'm sure there are plenty of folks here that can interpret them properly.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xffff9503ba87f010, 0xfffff80129ab1b20, 0xffffffffc000009a, 0x0000000000000004). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 00000000-0000-0000-0000-000000000000.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xffff84094a98b010, 0xfffff8012fdb1b20, 0xffffffffc000009a, 0x0000000000000004). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 00000000-0000-0000-0000-000000000000.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xffff9b8a63772010, 0xfffff80674751b20, 0xffffffffc000009a, 0x0000000000000004). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 00000000-0000-0000-0000-000000000000.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xffffc609f8608010, 0xfffff8068f691b20, 0xffffffffc000009a, 0x0000000000000004). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 00000000-0000-0000-0000-000000000000.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xffffba0cb964b460, 0xfffff80237031b20, 0xffffffffc000009a, 0x0000000000000004). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 00000000-0000-0000-0000-000000000000.

Screenshots are here:

https://imgur.com/a/BUk71IU

View: https://imgur.com/a/BUk71IU


I do notice that barraider pops up multiple times. Those are addons for the StreamDeck, but the StreamDeck wasn't added to the computer until 2 months after it was purchased so I'm fairly sure that alone isn't the issue. It does seem to be related to audio again however, which I do find very interesting. Maybe I'm suffering from confirmation bias...

The builders suggested a fresh install of Windows, so I'll likely allow that. It might at least rule out software being the issue.

My wife's new monitor will arrive in 2 days. The YGT monitor has been paired with the PC 100% of the time so I'm hopeful on that front.

Will continue to update this thread as new information comes to light. Thank you!

 

[Ralston18]

Will barraider uninstall? Or can it at least be disabled?

Another thing you can do is to run "sfc /scannow" and "dism".

References:

https://www.lifewire.com/how-to-use-sfc-scannow-to-repair-windows-system-files-2626161

How to use DISM command tool to repair Windows 10 image | Windows Central

Could find and fix problems created by the crashes/"improper shutdowns".

 

[daytondoes]

Will barraider uninstall? Or can it at least be disabled?

Barraider was uninstalled. Seemed to work for a few days, but had another crash recently.

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000116 (0xffffd309d81dd010, 0xfffff80180a33d00, 0xffffffffc000009a, 0x0000000000000004). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 2a30e219-20bd-44cd-8c14-33d260205dec.

Another thing you can do is to run "sfc /scannow" and "dism".

References:

https://www.lifewire.com/how-to-use-sfc-scannow-to-repair-windows-system-files-2626161

How to use DISM command tool to repair Windows 10 image | Windows Central

Could find and fix problems created by the crashes/"improper shutdowns".

SFC and DISM came back without any problems. HD Sentinel said my SSD's health was only 94%. I changed that out, but still having crashes.

 

[Ralston18]

I also see errors with respect to Civilization.

And it appears that you have picked up on that interesting pattern of 3 types of critical errors every 6 - 7 days.

Do you have any backups or AV scans or some other routine actions executing with a "weekly pattern"?

Take a look in Task Scheduler also. There could be something being launched with that sort of pattern.

Capture a couple more Reliability History screens that are up to date.

[Note: As for the memory dumps I will defer to other posters regarding interpretation and so forth.]

Reading back a bit there is still another troubleshooting path you can take.

Try using Task Manager and Resource Monitor to observe system performance. Use both tools but only one tool at a time.

The objective being to watch what the system is doing or trying do prior to any given crash.

Specifically what resources are being used, to what extent (%), and what is using any given resource.

First observe the system after boot up and stable, then while doing light work or browsing, then while gaming. May take some trial and error to work through the process. There is no need to hurry or rush the observations. Just make small changes as you add more "work" to the system.

Even though the PSU was upgraded (750 to 850 watts) that does not mean that the PSU is fully capable of meeting game demands.

Key is to discover some prelude condition to the crashes.

 

[RodroX]

How are you colling your i9 11900k ?, Did you check the temps while gaming/streaming?

Are you runing stock intel settings, or did you removed the power limits ?

This could very well be a motherboard/cpu random issue. Very hard to point out unless there are some on site testing involved with extra components.

On the other hand if you are going to reinstall Windows (which I hope is Windows 10 64 bits), remember to:

1. Unplug the 4TB driver and leave only the boot drive (which I believe is your 1TB SSD). You can plug the the WD HDD later on.
2. Unplug every other peripheral, but for the keyboard, mouse and speakers.
3. install driver starting with the chipset ones, get the newest ones from intel website.
4. Let Windows install the "sound" driver.
5. Install nvidia drivers, download them from nvidia website.
6. Once Windows is installed, with allthe main drivers and all updates from windows updates, then strat to add peripherals one by one, and restart after each install.
7. Aboid all motherboard apps, you need to install the chipset drivers from intel, and let Windows do the rest. Motherboard apps can conflict with other software, as the next one (8) do to.
8. Avoid by all means RGB app controllers. Don't use any of this software crap until you are 100% sure the issue is gone.
9. For checking temps (and a lot of other info), since it can keep the MAX temps on its log I suggest hwinfo64 portable, "Sensors Only" option. Theres no need to install this app and it can monitor your CPU, RAM, Mobo, GPU, SSD, HDD and more. It has 4 columns (Current, Min, Max and AVG) so you can check Max value, write down the number, and then go ahead play for 30 mins. Close the game and go check the Max number again.

Finally, what the full model number of your Gigabyte PSU?, I think you are missing one or more letters.

 

[daytondoes]

I also see errors with respect to Civilization.

And it appears that you have picked up on that interesting pattern of 3 types of critical errors every 6 - 7 days.

Do you have any backups or AV scans or some other routine actions executing with a "weekly pattern"?

Take a look in Task Scheduler also. There could be something being launched with that sort of pattern.

For some reason Task Scheduler has the history disabled. I've enabled that. A short glance doesn't show anything running weekly, but I have yet to really pick it apart. Should be much easier to find the task that was running during the time of the crash now that this history is enabled.

Capture a couple more Reliability History screens that are up to date.

[Note: As for the memory dumps I will defer to other posters regarding interpretation and so forth.]

I'll refrain from spamming more error codes. Here are the most recent crashes.

One interesting thing to note is that the most recent crash today occurred while doing nothing but watching YouTube and cruising for a hotel on two separate monitors. I think Discord was open on the third screen.

https://imgur.com/a/eFpqvBa

View: https://imgur.com/a/eFpqvBa

Reading back a bit there is still another troubleshooting path you can take.

Try using Task Manager and Resource Monitor to observe system performance. Use both tools but only one tool at a time.

The objective being to watch what the system is doing or trying do prior to any given crash.

Specifically what resources are being used, to what extent (%), and what is using any given resource.

First observe the system after boot up and stable, then while doing light work or browsing, then while gaming. May take some trial and error to work through the process. There is no need to hurry or rush the observations. Just make small changes as you add more "work" to the system.

Even though the PSU was upgraded (750 to 850 watts) that does not mean that the PSU is fully capable of meeting game demands.

Key is to discover some prelude condition to the crashes.

I have a HWinfo plugin for my StreamDeck so temps and CPU usage are monitors all the time, but I do think that pairing that with Task manager might tell me what is causing a spike in GPU power (which is still what I suspect is causing these issues)

How are you colling your i9 11900k ?, Did you check the temps while gaming/streaming?

First off, welcome to my living nightmare. Thanks for joining us. Ummm... The CPU is water-cooled with this thing. Temps are relatively stable. As mentioned previously, I do monitor it via HWinfo display on my StreamDeck and it never really gets above 70. Maybe touches 80 after a multiple-hour video render.

Are you runing stock intel settings, or did you removed the power limits ?

It is prebuilt... I haven't played with anything. I'd assume its stock, unless the builders overclocked it. But considering one of their suggested solutions was using MSI Afterburner to slightly underclock it? I'd presume they're the stock settings. I'd need to Google some stuff in order to find out how to check in order to verify 100% though.

This could very well be a motherboard/cpu random issue. Very hard to point out unless there are some on site testing involved with extra components.

Considering these are the only parts that haven't been replaced... It seems fairly likely. Day one busted components are sad. But it happens, I guess.

On the other hand if you are going to reinstall Windows (which I hope is Windows 10 64 bits), remember to:

1. Unplug the 4TB driver and leave only the boot drive (which I believe is your 1TB SSD). You can plug the the WD HDD later on.
2. Unplug every other peripheral, but for the keyboard, mouse and speakers.
3. install driver starting with the chipset ones, get the newest ones from intel website.
4. Let Windows install the "sound" driver.
5. Install nvidia drivers, download them from nvidia website.
6. Once Windows is installed, with allthe main drivers and all updates from windows updates, then strat to add peripherals one by one, and restart after each install.
7. Aboid all motherboard apps, you need to install the chipset drivers from intel, and let Windows do the rest. Motherboard apps can conflict with other software, as the next one (8) do to.
8. Avoid by all means RGB app controllers. Don't use any of this software crap until you are 100% sure the issue is gone.
9. For checking temps (and a lot of other info), since it can keep the MAX temps on its log I suggest hwinfo64 portable, "Sensors Only" option. Theres no need to install this app and it can monitor your CPU, RAM, Mobo, GPU, SSD, HDD and more. It has 4 columns (Current, Min, Max and AVG) so you can check Max value, write down the number, and then go ahead play for 30 mins. Close the game and go check the Max number again

It is Win 10 64 bit. This reinstall guide is golden. I hoped that it wouldn't come to all that, but again... It is what it is. I can't work with a PC that has a score of 4 for reliability on the high end...

Finally, what the full model number of your Gigabyte PSU?, I think you are missing one or more letters.

You're right. Box says: Model GP-AP850GM.

Again: "most recent crash today occurred while doing nothing but watching YouTube" It seems really bad... But it might help to narrow down the issue. It isn't because of temps or the gaming... There is definitely a deeper issue happening here.

Lastly, thank you both for throwing your hats into the ring. This has been a headache, but I really do appreciate you spending your valuable time and helping me out.



ANOTHER EDIT:

Big news. My wife is a genius. She said to throw all the parts into a PC builder which... I guess the build-team doesnt? Some basic hardware incompatibilities were found. I will share them with you now so we can all ponder this together.

https://imgur.com/a/JfoVfm6

View: https://imgur.com/a/JfoVfm6

https://imgur.com/a/UKRbKn6

View: https://imgur.com/a/UKRbKn6

https://imgur.com/a/Vm81dze

View: https://imgur.com/a/Vm81dze

https://imgur.com/a/QZcBpMQ

View: https://imgur.com/a/QZcBpMQ


Could it be so simple? Could the PC boot at all? Am I looking at this all wrong? Has hubris been my downfall for an entire YEAR??? /drama

Lemme know what ya think. Thanks!

 

[RodroX]

The BIOS date showed in that picture is really close to the BIOS you need to run 11th gen intel CPU.

One easy way to check the BIOS version in Windows is using CPU-Z and looking at the mainboard tab. You should see the BIOS version pretty clear there.

On the other hand, Im almost sure that "1.60" in system info, is probably equal to the 7C79v16. (Never used an MSI mobo, but I think thats how they are shown in Win 10 system info)

 

[daytondoes]

The BIOS date showed in that picture is really close to the BIOS you need to run 11th gen intel CPU.

One easy way to check the BIOS version in Windows is using CPU-Z and looking at the mainboard tab. You should see the BIOS version pretty clear there.

On the other hand, Im almost sure that "1.60" in system info, is probably equal to the 7C79v16. (Never used an MSI mobo, but I think thats how they are shown in Win 10 system info)

Hmm... CPU-Z also lists it as 1.60, but your logic makes sense. I really thought this was my big break. What else would the v stand for? I'll have check out BIOS when I get a chance. If that matches? Well, I'm rolling past the PC build shop so I'll probably just hand it off to them. They have more resources and know-how than I do, that's for sure.

 

[Ralston18]

There may be a mix of problems.

Reliability History is showing Critical Errors regarding improper shutdowns. The system may be doing the shutdowns and/or you are being forced to shutdown the system via the PSU switch.

What are the warnings? Noted a few of them and they may or may not be relevant.

Click to view the Technical details on a few of most recent Critical errors and Warnings. What error codes are listed? The error codes may or may not prove useful but worth taking a look at.

= = = =

Power down, unplug, open the case.

Clean out dust and debris.

Verify by sight and feel that all connections, cards, RAM, jumpers, and case connections are fully and firmly in place. Components and connections do wiggle loose over time due to heat related expansion/contraction and vibrations.

Use a bright flashlight to look for signs of damage: bare conductor showing, melted insulation, kinked or pinched wires, browned or blackened areas/components, loose or missing screws. swollen components, signs of liquids.

Objective simply being to eliminate, or at least reduce, the chance that the underlying cause is a bad connection somewhere. Something physical.

= = = =

Also: power losses and improper shutdowns may corrupt files. So the next step is to run "sfc /scannow" and "dism" via the Command Prompt.

FYI:

https://www.lifewire.com/how-to-use-sfc-scannow-to-repair-windows-system-files-2626161

How to use DISM command tool to repair Windows 10 image | Windows Central

 

[RodroX]

Hmm... CPU-Z also lists it as 1.60, but your logic makes sense. I really thought this was my big break. What else would the v stand for? I'll have check out BIOS when I get a chance. If that matches? Well, I'm rolling past the PC build shop so I'll probably just hand it off to them. They have more resources and know-how than I do, that's for sure.

Besides what Ralston18 wrote up there, and considering how "old" the BIOS is, it wont hurt to install a newer one. But before you do it or if you send it to the PC shop, make sure you do a backup. If you have no clue whats going on with your system is always a good idea to backup the important data.

And Im not saying its faulty at all but, next time you "build" a PC, try to go with high-end PSUs models of brands like Corsair, Seasonic or EVGA. And try to avoid "new players" like Gigabyte.

 

[daytondoes]

There may be a mix of problems.

There definitely were a mix of problems... But the shop seems to have gotten them sorted to an acceptable degree.

Try to avoid "new players" like Gigabyte.

They probably have some deal with the builders that I used. I stopped into the shop on the way to Manila. Seems like all they have.

--------------

In summary, I took the rig to the PC shop on our way to Manila. Told them to keep it for months. Told them I didn't want it back until it stopped crashing. 2 days later they send a message saying it's ready to be picked up. They were right, the crashes have finally stopped. GPU still hiccups once in a while, but I can handle that.

It seems the GPU was faulty from day one. Very anti-climatic. It's also the part that they were most loathe to replace for obvious reasons ($$$) but they did it. They also reinstalled Windows and replaced the SSD with an M2... All of this at no cost, which I'm super grateful for.

So there really was no great mystery, but I will consider this solved for now. I'll also keep the things that I've learned in this thread in mind in case I experience something similar in the future. Thank you guys so much for your help.

 

[Ralston18]

You are welcome.

Do make it a bit of a habit to check in on Task Manager, Resource Monitor, Update History, Reliability History, and Event Viewer every so often.

Even Process Explorer...

Not that doing so will prevent anything from happening but just having some overall sense about how the system is performing and what errors etc. are happening goes a long ways towards knowing more about your computer.

Could save you some time and money in the future.