- Jan 10, 2024
- 3
- 0
- 10
Hello guys,
Ryzen9 3900X
Gigabyte B450M
A-Data 16GB DDR4
MSI 1080TI Gaming X 11GB
Corsair CS 850M
Samsung SSD 2TB
With the introduction of Holiday OPS 2024 in World of Tanks, I've started experiencing a bunch of weird game crashes to destkop. The first thing I've noticed after WoT introduced the December patch was that after 3-5 minutes while navigating through the WoT garage (a main menu where you can see your tanks etc), the game is freezing for 2-3 seconds then it crashes to destkop saying 'ERROR - A videocard error has occured. This error is usually caused by the graphics drivers crashing; try updating the drivers, check the videocard or restart the game'.
My first action was to remove existing nvidia drivers using DDU in + safemode + disabled network then installing latest driver. It didn't work.
Creating a ticket in WoT didnt help much as I've been told nobody else reported the issue ( I also confirmed with my friends playing the game ).
Creating a ticket on Nvidia forums got me up to a certain point and helped a bit with testing - they suggested using an older driver for testing 517.48 which I currently use.
.
Going in depth with the testing of this particular game resulted in the following: Playing the battles maxedout on ULTRA settings doesn't create any issues. The moment when you return to garage after a battle or log into the garage the above crash happens. Next I have discovered a particular video graphical setting in game called 'Lighting' makes or brakes the garage crash. Setting this value to Medium or below removes the crash completly. Setting this value to High, Max or Ultra induces the crash in a matter of 2-3 minutes while in staying in garage. Leaving the game IDLE in garage with this setting on a higher value also produces the crash.
... since yesterday they removed the Winter Holiday patch , no more gamecrash in World of Tanks
The next weird thing noticed recently is with Dota2 ( again weird coincidence they updated a Winter special theme for the main menu). It happened while in main menu during finding matches, or while in game and navigating through the main menu ( you can do that in Dota ), the application itself becomes black, leaving a red outline which you can alt+tab out of it but you cannot go back to the game anymore, not even kill the process. In oder to close the game you need to signoff from windows and start Dota2 over hoping for the best.
It is super weird that I had a couple of days without any problems in Dota2, but today, for the first time the game crashed (technical hang actually ) even while playing the game ( it didnt happen before).
Witcher3 which I've played 200 hours during the last years (ULTRA details) using this System config is also behaving funny recently. Last month when the first issues came, I did a Witcher3 test using Ultra details in order to load my GPU to the max , all went fine no problems. After repeating some tests 2-3 days ago, I ve noticed this weird behavior. Setting the graphic details to ULTRA and then loading a savegame , progresses the savegame load 100% up to the point when sounds are coming and the video is expected but then after 3 seconds the game is crashing to desktop. Setting details to Max or High , then loading the game seemed to do the trick. Not anymore, now I can load any savegame with Medium details in order to avoid crash. Once loaded, I can set everyhing to ULTRA and play without problems ( only tried 5 minutes of gameplay so far ).
Running 3D Mark Firestrike doesnt crash the test.
Under load, HWInfo tool shows low values for 8pin voltage connectors, such as 11.4v or even 11.3v for the GPU 8 pin connector no1 & 2. As I am writing this I've checked the recent minimum and it dropped at some point to 11.096v. Using Windows normally keeps it at 12v. Please note I've got a Ponytail 8pin cable which connects PSU with GPU ( I know it's not ideal and I should use 2 different 8 pin cables for my GPU , but no issues in 3 years using this approach)
Took my GPU out today and went to a friend's house. First test under load revealed a constant 12v for the 2 8pin input voltage connectors, with a minimum value registered as 11.880v during my 1 hour testing. I then hoped my PSU is the reason for the recent issues. Then we ran Cyberpunk2077 on his rig using my GPU. What we've seen was weird enough to confuse me more. Setting graphics to ULTRA in main-menu before loading a savegame, then jumping into a savegame works fine. Jumping into a savegame with Medium details AND then while playing setting details to ULTRA crashes the game with send report option. No need to say that setting details to Medium then loading savegame doesnt crash the game.
Other games didnt crash (Control, Doom Eternal, Dota2), but I had limited time to test 4-5 games at his place. (we also used a ponytail cable here)
Both PSU & GPU have been used for mining by the previous owner for around 1 year in 2019. No issues until now whatsoever (yes, i know mining damages components ).
What do I need to replace as broken component, PSU or GPU? Hopefully not both as a good PSU is 150 Eur while a 4070(TI) is almost 1000 Eur in my country.
Whould would you guys do in my place ?
edit: can a faulty PSU providing undervoltage for GPU for a 1-2 years result in a damaged GPU in order for it to start behaving funny on a sudden ?
Thank you for reading this long post!!
Ryzen9 3900X
Gigabyte B450M
A-Data 16GB DDR4
MSI 1080TI Gaming X 11GB
Corsair CS 850M
Samsung SSD 2TB
With the introduction of Holiday OPS 2024 in World of Tanks, I've started experiencing a bunch of weird game crashes to destkop. The first thing I've noticed after WoT introduced the December patch was that after 3-5 minutes while navigating through the WoT garage (a main menu where you can see your tanks etc), the game is freezing for 2-3 seconds then it crashes to destkop saying 'ERROR - A videocard error has occured. This error is usually caused by the graphics drivers crashing; try updating the drivers, check the videocard or restart the game'.
My first action was to remove existing nvidia drivers using DDU in + safemode + disabled network then installing latest driver. It didn't work.
Creating a ticket in WoT didnt help much as I've been told nobody else reported the issue ( I also confirmed with my friends playing the game ).
Creating a ticket on Nvidia forums got me up to a certain point and helped a bit with testing - they suggested using an older driver for testing 517.48 which I currently use.
.
Going in depth with the testing of this particular game resulted in the following: Playing the battles maxedout on ULTRA settings doesn't create any issues. The moment when you return to garage after a battle or log into the garage the above crash happens. Next I have discovered a particular video graphical setting in game called 'Lighting' makes or brakes the garage crash. Setting this value to Medium or below removes the crash completly. Setting this value to High, Max or Ultra induces the crash in a matter of 2-3 minutes while in staying in garage. Leaving the game IDLE in garage with this setting on a higher value also produces the crash.
... since yesterday they removed the Winter Holiday patch , no more gamecrash in World of Tanks
The next weird thing noticed recently is with Dota2 ( again weird coincidence they updated a Winter special theme for the main menu). It happened while in main menu during finding matches, or while in game and navigating through the main menu ( you can do that in Dota ), the application itself becomes black, leaving a red outline which you can alt+tab out of it but you cannot go back to the game anymore, not even kill the process. In oder to close the game you need to signoff from windows and start Dota2 over hoping for the best.
It is super weird that I had a couple of days without any problems in Dota2, but today, for the first time the game crashed (technical hang actually ) even while playing the game ( it didnt happen before).
Witcher3 which I've played 200 hours during the last years (ULTRA details) using this System config is also behaving funny recently. Last month when the first issues came, I did a Witcher3 test using Ultra details in order to load my GPU to the max , all went fine no problems. After repeating some tests 2-3 days ago, I ve noticed this weird behavior. Setting the graphic details to ULTRA and then loading a savegame , progresses the savegame load 100% up to the point when sounds are coming and the video is expected but then after 3 seconds the game is crashing to desktop. Setting details to Max or High , then loading the game seemed to do the trick. Not anymore, now I can load any savegame with Medium details in order to avoid crash. Once loaded, I can set everyhing to ULTRA and play without problems ( only tried 5 minutes of gameplay so far ).
Running 3D Mark Firestrike doesnt crash the test.
Under load, HWInfo tool shows low values for 8pin voltage connectors, such as 11.4v or even 11.3v for the GPU 8 pin connector no1 & 2. As I am writing this I've checked the recent minimum and it dropped at some point to 11.096v. Using Windows normally keeps it at 12v. Please note I've got a Ponytail 8pin cable which connects PSU with GPU ( I know it's not ideal and I should use 2 different 8 pin cables for my GPU , but no issues in 3 years using this approach)
Took my GPU out today and went to a friend's house. First test under load revealed a constant 12v for the 2 8pin input voltage connectors, with a minimum value registered as 11.880v during my 1 hour testing. I then hoped my PSU is the reason for the recent issues. Then we ran Cyberpunk2077 on his rig using my GPU. What we've seen was weird enough to confuse me more. Setting graphics to ULTRA in main-menu before loading a savegame, then jumping into a savegame works fine. Jumping into a savegame with Medium details AND then while playing setting details to ULTRA crashes the game with send report option. No need to say that setting details to Medium then loading savegame doesnt crash the game.
Other games didnt crash (Control, Doom Eternal, Dota2), but I had limited time to test 4-5 games at his place. (we also used a ponytail cable here)
Both PSU & GPU have been used for mining by the previous owner for around 1 year in 2019. No issues until now whatsoever (yes, i know mining damages components ).
What do I need to replace as broken component, PSU or GPU? Hopefully not both as a good PSU is 150 Eur while a 4070(TI) is almost 1000 Eur in my country.
Whould would you guys do in my place ?
edit: can a faulty PSU providing undervoltage for GPU for a 1-2 years result in a damaged GPU in order for it to start behaving funny on a sudden ?
Thank you for reading this long post!!
Last edited:
Will continue to test other games for weird crashes in the next few days.
Facebook
Twitter
Instagram