Question Black Screen During Heavy GPU Load

[FishCouch]

Hello,

I'm experiencing regular black screens when gaming. Suddenly my screen will turn black, but the sounds will continue working. Then I just press my power button to reset.

I'm trying to diagnose the problem and have stumbled upon something that I think might be the culprit.

While monitoring my temps using HWinfo during a gaming session my GPU temperature is about 74C, but the "Hot Spot Temperature" is 107C.

Is this normal? Should I send my card in for warranty?

Specs:
Ryzen 9 5900x
TUF Gaming B550-Plus
KFA 2 RTX 3080
64GB RAM
Corsair RM750
Sabrent Rocket-Q 2TB
Windows 11

Best,
F

 

[Roland Of Gilead]

Hey there,

Then I just press my power button to reset.

That's not really the best thing to do, but I understand that you have to, to get back into the system. It will corrupt Windows files, and cause you some serious bugging.

While monitoring my temps using HWinfo during a gaming session my GPU temperature is about 74C, but the "Hot Spot Temperature" is 107C.

It's likely that could be it. Not so concerned about the 74c, although you would expect temps to be about 5-10c lower. Around the 65-70 range is where I'd want it at.

What kind of case are you running? How many fans, and in what location/direction. For example my 3060ti with a 1000mhz mem OC and about 120mhz core OC, the hotspot never gets above 80c or so.

Are all system drivers, and bios up to date?

 

[FishCouch]

Hey there,


That's not really the best thing to do, but I understand that you have to, to get back into the system. It will corrupt Windows files, and cause you some serious bugging.

It's likely that could be it. Not so concerned about the 74c, although you would expect temps to be about 5-10c lower. Around the 65-70 range is where I'd want it at.

What kind of case are you running? How many fans, and in what location/direction. For example my 3060ti with a 1000mhz mem OC and about 120mhz core OC, the hotspot never gets above 80c or so.

Are all system drivers, and bios up to date?

Thanks for the reply.

I have a small case, but I have both side panels removed (It seems like I get more black screens when the side panels are attached, hence why it might be an overheating issue). Im using a Noctua NH-U12A on my CPU. The case is some very cheap one (https://www.amazon.de/gp/product/B07PJZD4QC/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1). It has 2 fans in the front pulling in air, one in the back and top pushing air out.

Systems drivers are not up to date, good catch. Will update those now.

Anything else I should do except press the power button when I get a black screen?

Best,
F

 

[Roland Of Gilead]

Systems drivers are not up to date, good catch. Will update those now.

Let us know how that goes. Bios too?

Anything else I should do except press the power button when I get a black screen?

Not much else. However, it really does nesecitate an OS reinstall.

 

[FishCouch]

Let us know how that goes. Bios too?

Bios has been updated now too

Did some game testing, btu the temps are almost identical. Haven't had enough time to play enough to see if I'm still getting black screens though

Not much else. However, it really does nesecitate an OS reinstall.

fr? How come, everything works fine still after having done this many times.

 

[triplex1]

if you haven't put them for chipset, put them on now from amd
https://www.amd.com/en/support/downloads/drivers.html/chipsets/am4/b550.html

 

[FishCouch]

if you haven't put them for chipset, put them on now from amd
https://www.amd.com/en/support/downloads/drivers.html/chipsets/am4/b550.html

yea, everything is all updated now. Ran some gaming again and got identical temps basically. no black screens yet, but I just tested it for liek 10 min or so. Usualyl Ill get 2-3 black screens after 5 h or something

 

[satrow]

Look in Reliability Monitor at the time you were forced to reboot, there's very likely to be a Bug Check 0x116: VIDEO_TDR_FAILURE recorded.

​

 

[Phaaze88]

74C, but the "Hot Spot Temperature" is 107C.

Too large a gap. Ideally should be around 20C or less.
The core needs a repaste. Send it in for warranty if it's still an option; let them deal with it first.
Now, if they tell you that's normal/expected, that's a load of crap.

By the way, what's the memory junction been like?[Use hwinfo.]

The case is some very cheap one (https://www.amazon.de/gp/product/B07PJZD4QC/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1).

Gpu core load temperature is surprisingly good considering that the front panel looks sealed.

fr? How come, everything works fine still after having done this many times.

Over time, the OS just gets bloated/constipated with older drivers/registry/etc, and we need to give it a laxative every now and then. The boneheads at MS couldn't be bothered to develop a proper digestive system.

 

[FishCouch]

Look in Reliability Monitor at the time you were forced to reboot, there's very likely to be a Bug Check 0x116: VIDEO_TDR_FAILURE recorded.

​

I looked into the realiblity monitor (this is something I've never heard about before). I wasn't able to paste an image here, but here's a link to the screenshot of what I believe are the black screen events. "hardware error" and the after that "Windows Stopped working" error, and then "Windows was not properly shut down" error.

in the Technical details of the "Windows Stopped Working" error it says "The bugcheck was: 0x00000119 (0x0000000000000002, 0xffffffffc000000d, 0xffffd00d664a78e0, 0xffffdf0e6563b860)". Does that mean anything to you?

https://ibb.co/ZKZLDnD

 

[FishCouch]

By the way, what's the memory junction been like?[Use hwinfo.]

Its very similar to the core temp, within 1 degree difference

Gpu core load temperature is surprisingly good considering that the front panel looks sealed.

I keep both side panels removed, so quite a lot of airflow I suppose (or that was my reasoning)

Over time, the OS just gets bloated/constipated with older drivers/registry/etc, and we need to give it a laxative every now and then. The boneheads at MS couldn't be bothered to develop a proper digestive system.

I suppose it serves their purpose in getting people to buy new products once its bloated enough.
Any easy simple way to perform a reset without messing with my files?

 

[Phaaze88]

Its very similar to the core temp, within 1 degree difference

I keep both side panels removed, so quite a lot of airflow I suppose (or that was my reasoning)


I suppose it serves their purpose in getting people to buy new products once its bloated enough.
Any easy simple way to perform a reset without messing with my files?

That's great!

OH. I thought that was a momentary thing. Sorry.

Backup the important files before doing so.

 

[FishCouch]

OH. I thought that was a momentary thing. Sorry.

I figured that it will be cooler that way idk haha

Backup the important files before doing so.

Oh that's a massive job of reinstalling terabytes of software and backing up more terabytes of files (big sad)

 

[satrow]

0x00000119

https://learn.microsoft.com/en-us/w...-check-0x119---video-scheduler-internal-error

Bug Check 0x119: VIDEO_SCHEDULER_INTERNAL_ERROR

The VIDEO_SCHEDULER_INTERNAL_ERROR bug check has a value of 0x00000119. This bug check indicates that the video scheduler has detected a fatal violation.

Parameter 1 is the only parameter of interest. It identifies the exact violation.
Parameter 1 Cause of error :-
0x2 Driver failed upon the submission of a command. (DRIVER_FAILED_SUBMIT_COMMAND)

I'm not 100% sure of what 'DRIVER_FAILED_SUBMIT_COMMAND' means, the video scheduler detected that a (graphics) driver failed to ?respond to?, or to ?give? a command?

So, might it be 'something' interfered/blocked it? Or there's a hardware error, as suggested in the Reliability Monitor logs? Can you try to access the details from the Hardware Errors, perhaps they'll help confirm *something* - ?heat? as the cause?

 

[FishCouch]

https://learn.microsoft.com/en-us/w...-check-0x119---video-scheduler-internal-error

Bug Check 0x119: VIDEO_SCHEDULER_INTERNAL_ERROR

The VIDEO_SCHEDULER_INTERNAL_ERROR bug check has a value of 0x00000119. This bug check indicates that the video scheduler has detected a fatal violation.

Parameter 1 is the only parameter of interest. It identifies the exact violation.
Parameter 1 Cause of error :-
0x2 Driver failed upon the submission of a command. (DRIVER_FAILED_SUBMIT_COMMAND)

I'm not 100% sure of what 'DRIVER_FAILED_SUBMIT_COMMAND' means, the video scheduler detected that a (graphics) driver failed to ?respond to?, or to ?give? a command?

So, might it be 'something' interfered/blocked it? Or there's a hardware error, as suggested in the Reliability Monitor logs? Can you try to access the details from the Hardware Errors, perhaps they'll help confirm *something* - ?heat? as the cause?

This is the infor given in the hardware error:
"Description
A problem with your hardware caused Windows to stop working correctly.

Problem signature
Problem Event Name: LiveKernelEvent
Code: 141
Parameter 1: ffffdf0e727e3010
Parameter 2: fffff80563c05390
Parameter 3: 0
Parameter 4: 68e0
OS version: 10_0_22631
Service Pack: 0_0
Product: 768_1
OS Version: 10.0.22631.2.0.0.768.101
Locale ID: 8192"

Nothing about heat

But it seems clear its from the graphics card, right?

 

[satrow]

But it seems clear its from the graphics card, right?

Related to definitely, the GFX driver is in the running, not sure if anything outside the graphics card/drivers was involved.