- Sep 20, 2022
- 2
- 0
- 10
Hello,
I've had this problem since I finished my build last week where the graphics card driver crashes if the graphics card is used for too long, or at least it seems so.
My system:
Ryzen 9 5950x max 74c
MSI 3060ti Ventus 2x max 75c
Corsair 3600Mhz ddr4 no thermostat but it does get quite warm to the touch
x570i motherboard mini-itx, one NVMe drive gets to 66c
1x 1080p60, 1x 1080p144 screens, both on the graphics card
850w SFX Gold Cooler Master
I have tried reinstalling Windows 10 from scratch, reinstalling graphics drivers 100x, updating BIOS, changing out RAM, checking temps, running HeavyLoad and Furmark benchmarks to test CPU, GPU and RAM, changing RAM freq and graphics card clock (under and over), jury-rigging the case fan so it blows on the RAM, changing Nvidia control panel settings, staring hopelessly at the jumble of parts and the only problem that actually seems to exist is this message that gets recorded in the Windows Event Viewer after my screens go blank and Windows makes the 'disconnected' and 'reconnected' device sounds.
[
The description for Event ID 14 from source nvlddmkm 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:
SPI read failure at address 272! (0x00000065)
The message resource is present but the message was not found in the message table
]
The SPI read line could be different, like
"\Device\Video3
Error occurred on GPUID: 900"
or
"\Device\Video3
Reset TDR occurred on GPUID:900"
The progression of symptoms goes like this:
I open a game, absolutely any game but not the benchmarks above, and I play it just fine for a few minutes, then small graphics problems start to occur such as flickering HUD elements, (and even audio has cut out) and frame rate starts to drop. The game can still (but not always) be responsive to ie mouse movements, as if it's able to know what it's supposed to do but has so few frames to show it. Usually, lagging games are unresponsive, so when you move the mouse the camera in-game takes a few frame updates to respond.
At this point, more than likely, the game will crash and my computer may or may not blue screen and the error in the Event Viewer will show up.
As soon as I open the games, the graphics card runs at 100%, on menus or in-game, the performance is exactly the same. Soon after, it crashes.
The problem has also gotten worse over the last week, I have played a game for a full day and never seen any drop in performance, with a 1080p twitch stream on the 60hz monitor all the while. Randomly and rarely, even, the case fan will start blasting while browsing firefox, for example while writing this post with nothing in the background.
A hopefully useful tidbit is that the GPU monitoring software on FurMark, called GPU Shark, stops reading the current state of the GPU. Usually it tells the current clock speed, usage, current power, limiting policies, and more, but after I try to play a game it stops outputting any of that info.
What in the sam heck is this thing's deal?
I've had this problem since I finished my build last week where the graphics card driver crashes if the graphics card is used for too long, or at least it seems so.
My system:
Ryzen 9 5950x max 74c
MSI 3060ti Ventus 2x max 75c
Corsair 3600Mhz ddr4 no thermostat but it does get quite warm to the touch
x570i motherboard mini-itx, one NVMe drive gets to 66c
1x 1080p60, 1x 1080p144 screens, both on the graphics card
850w SFX Gold Cooler Master
I have tried reinstalling Windows 10 from scratch, reinstalling graphics drivers 100x, updating BIOS, changing out RAM, checking temps, running HeavyLoad and Furmark benchmarks to test CPU, GPU and RAM, changing RAM freq and graphics card clock (under and over), jury-rigging the case fan so it blows on the RAM, changing Nvidia control panel settings, staring hopelessly at the jumble of parts and the only problem that actually seems to exist is this message that gets recorded in the Windows Event Viewer after my screens go blank and Windows makes the 'disconnected' and 'reconnected' device sounds.
[
The description for Event ID 14 from source nvlddmkm 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:
SPI read failure at address 272! (0x00000065)
The message resource is present but the message was not found in the message table
]
The SPI read line could be different, like
"\Device\Video3
Error occurred on GPUID: 900"
or
"\Device\Video3
Reset TDR occurred on GPUID:900"
The progression of symptoms goes like this:
I open a game, absolutely any game but not the benchmarks above, and I play it just fine for a few minutes, then small graphics problems start to occur such as flickering HUD elements, (and even audio has cut out) and frame rate starts to drop. The game can still (but not always) be responsive to ie mouse movements, as if it's able to know what it's supposed to do but has so few frames to show it. Usually, lagging games are unresponsive, so when you move the mouse the camera in-game takes a few frame updates to respond.
At this point, more than likely, the game will crash and my computer may or may not blue screen and the error in the Event Viewer will show up.
As soon as I open the games, the graphics card runs at 100%, on menus or in-game, the performance is exactly the same. Soon after, it crashes.
The problem has also gotten worse over the last week, I have played a game for a full day and never seen any drop in performance, with a 1080p twitch stream on the 60hz monitor all the while. Randomly and rarely, even, the case fan will start blasting while browsing firefox, for example while writing this post with nothing in the background.
A hopefully useful tidbit is that the GPU monitoring software on FurMark, called GPU Shark, stops reading the current state of the GPU. Usually it tells the current clock speed, usage, current power, limiting policies, and more, but after I try to play a game it stops outputting any of that info.
What in the sam heck is this thing's deal?
Facebook
Twitter
Instagram