Periodic AMD Driver Crashes at Random Intervals While Playing Games


  1. Posts : 35
    Windows 10 Pro Preview Version 1903 64-bit
       #1

    Periodic AMD Driver Crashes at Random Intervals While Playing Games


    Hello,

    As is written in my computer specifications, I am running an ASUS AMD Radeon R9 380 with 4GB GDDR5, I am running it with no overclock and on stock settings in AMD's Radeon Settings application. In normal use (web browsing, listening to music, programming, etc), my computer is stable, and works great, running perfectly fine for days at a time (during a long download or something).

    However, as soon as I start up certain games like Cities: Skylines or Minecraft (any release version) in particular, I get periodic display driver crashes that the games are unable to recover from, but Windows does recover. The entire screen will freeze for a few seconds, but the mouse still moves, then the mouse will freeze too, and a bit later, everything goes black, then the displays reinitialize and show everything as normal, however, certain programs, such as games will stay white and not recover from the crash. In the case of Cities: Skylines, I get a notification in the corner saying "Display driver stopped responding and has recovered", but with Minecraft, I generally get no notification at all, however, in both cases, if I try to open Radeon Settings, the program immediately freezes and has to be restarted with the "Radeon Settings has stopped working" message box, telling me to either close, restart or troubleshoot or whatever. I also now get a notification saying "General: Default Radeon WattMan settings have been restored due to an unexpected system failure."

    I can't find any meaningful errors in the event viewer, none of the games give meaningful crash messages, as the games don't actually crash, they just don't recover and turn white, leaving me with a perpetually frozen, white box of a game that I have to force close in task manager. I've looked around for a solution to this problem for 2 years now, and I still can't figure it out. Is there anything I can try to solve this? I've tried underclocking my GPU as well, but that didn't help. And I am no longer in any warranty period either, so I can't really replace anything if it is a hardware issue that isn't something cheap and simple.

    Not sure if it matters, but I have 2 displays as listed, an ACER T232HL (HDMI 1920x1080 60Hz) and ACER AL1716 (DVI-I to VGA adapter 1280x1024 60Hz). My ACER AL1716 runs at 75Hz, but I reduced it to 60Hz as I have read it helps with stability to run both displays at the same refresh rate.

    Thank you!

    EDIT: This issue has persisted over 2 years, so it shouldn't really matter, but here's my AMD Radeon software info:
    Code:
    Radeon Settings Version - 2019.0321.1751.32142
    Driver Packaging Version - 19.10-190321a-340873E
    Provider - Advanced Micro Devices, Inc.
    2D Driver Version - 8.1.1.1634
    Direct3D® Version - 9.14.10.01394
    OpenGL® Version - 26.20.11000.13558
    OpenCL™ Version - 10.0.2841.4
    AMD Audio Driver Version - 10.0.1.7
    Vulkan™ Driver Version - 2.0.80
    Vulkan™ API Version - 1.1.101
    Hardware Info:
    Code:
    Graphics Card Manufacturer - Powered by AMD
    Graphics Chipset - AMD Radeon (TM) R9 380 Series
    Device ID - 6939
    Vendor ID - 1002
    SubSystem ID - 04E3
    SubSystem Vendor ID - 1043
    Revision ID - F1
    Bus Type - PCI Express 3.0
    Current Bus Settings - PCI Express 3.0 x16
    BIOS Version - 015.049.000.004
    BIOS Part Number - 115-c766pi20-101
    BIOS Date - 2015/07/14 06:06
    Memory Size - 4096 MB
    Memory Type - GDDR5
    Memory Clock - 1425 MHz
    Core Clock - 990 MHz
    Total Memory Bandwidth - 182 GByte/s
    Memory Bit Rate - 5.70 Gbps
    2D Driver File Path - /REGISTRY/MACHINE/SYSTEM/CurrentControlSet/Control/Class/{4d36e968-e325-11ce-bfc1-08002be10318}/0000
    OpenGL® API Version - 4.6
    OpenCL™ API Version - 2.0
      My Computer


  2. Posts : 19,516
    W11+W11 Developer Insider + Linux
       #2

    Did you try different drivers ?
      My Computers


  3. Posts : 35
    Windows 10 Pro Preview Version 1903 64-bit
    Thread Starter
       #3

    CountMike said:
    Did you try different drivers ?
    I've used a variety of driver versions, both self-installed, and automatically installed (by Windows 10) since like version 16, up to the current version 19. They all seem to be the same, although stability may have increased slightly over the years, I'm not sure.
      My Computer


  4. Posts : 173
    Windows 10 64bit
       #4

    Can you try a stronger PSU? Yours might be failing or just borderline powerful enough to run your system. Starting games might just put it over the edge. I had similar problems a few years ago until I upgraded my PSU.
      My Computer


  5. Posts : 35
    Windows 10 Pro Preview Version 1903 64-bit
    Thread Starter
       #5

    ram1220 said:
    Can you try a stronger PSU? Yours might be failing or just borderline powerful enough to run your system. Starting games might just put it over the edge. I had similar problems a few years ago until I upgraded my PSU.
    I have been considering that. I don't have any other PSUs though. It should be plenty powerful enough to power my computer, as my UPS reports a max draw of about 300W and about 120W of that is my 2 monitors, plus a few random other small things. So my computer should only be drawing about 180W, according to my PSU, which I think is fairly accurate, as it is relatively new and of good quality. I have tried checking AIDA64 and watched the PSU voltages while idle and under load, and they all fell well within acceptable ranges according to the version 2.2 specifications for ATX, hardly deviating at all from the desired values, so I'm not sure if that is an issue either.

    AIDA64 at medium load (game running, but not stress testing)
    Code:
    Temperatures	
    Motherboard					35 °C  (95 °F)
    CPU							36 °C  (97 °F)
    CPU Diode					46 °C  (115 °F)
    Chipset						47 °C  (117 °F)
    VRM							60 °C  (140 °F)
    Aux							62 °C  (144 °F)
    PCI-E						44 °C  (111 °F)
    GPU Diode					65 °C  (149 °F)
    GPU VRM1					58 °C  (136 °F)
    GPU VRM2					79 °C  (174 °F)
    Samsung SSD 860 EVO 1TB		29 °C  (84 °F)
    WDC WD2003FZEX-00Z4SA0		33 °C  (91 °F)
    Samsung SSD 960 EVO 250GB	46 °C  (115 °F)
    	
    Cooling Fans	
    CPU							1239 RPM
    GPU							753 RPM  (30%)
    	
    Voltage Values	
    CPU Core					1.416 V
    CPU VID						0.887 V
    CPU VDD						1.400 V
    CPU VDDNB					1.087 V
    +3.3 V						3.305 V
    +5 V						5.010 V
    +12 V						12.168 V
    +3.3 V Standby				3.288 V
    VBAT Battery				3.216 V
    VDDP						0.900 V
    SoC							1.236 V
    DIMM						1.356 V
    Battery Input				115.000 V
    Battery Output				115.000 V
    GPU Core					1.156 V
    GPU Vcc						3.242 V
    GPU Memory					1.156 V
    GPU +12V					12.156 V
    GPU VRM						1.164 V
    	
    Current Values	
    CPU VDD						32.22 A
    CPU VDDNB					17.32 A
    GPU Memory					4.00 A
    GPU VRM						46.00 A
    	
    Power Values	
    CPU Package					43.28 W
    CPU VDD						45.10 W
    CPU VDDNB					18.83 W
    Battery						243.80 W
    GPU							61.88 W
    GPU Memory					4.63 W
    GPU VRM						53.19 W
      My Computer


  6. Posts : 35
    Windows 10 Pro Preview Version 1903 64-bit
    Thread Starter
       #6

    Sorry to bring up an old thread, but I wanted to mention for in case somebody ever sees this for some reason, I think the issue here was my RAM. I tried doing a run of Memtest, and within 1 pass, it found 164 errors. Not as bad as some other memory I've seen, but likely still bad enough to cause subtle issues that are hard to pin down. I have contacted Corsair and should get a replacement set, but I am fairly certain now that unstable RAM was causing my problems.
      My Computer


  7. Posts : 19,516
    W11+W11 Developer Insider + Linux
       #7

    WolfLeader116 said:
    Sorry to bring up an old thread, but I wanted to mention for in case somebody ever sees this for some reason, I think the issue here was my RAM. I tried doing a run of Memtest, and within 1 pass, it found 164 errors. Not as bad as some other memory I've seen, but likely still bad enough to cause subtle issues that are hard to pin down. I have contacted Corsair and should get a replacement set, but I am fairly certain now that unstable RAM was causing my problems.
    There's no bad or good number of memory errors, even one is enough to be of concern. Same with bad sectors on disks.
      My Computers


  8. Posts : 35
    Windows 10 Pro Preview Version 1903 64-bit
    Thread Starter
       #8

    CountMike said:
    There's no bad or good number of memory errors, even one is enough to be of concern. Same with bad sectors on disks.
    Yeah, but I meant that I have gotten RAM in the past where it was so broken, the Windows couldn't even boot. I just meant it was a relatively minor error that made it more difficult to diagnose, as it was not as obvious that it was a memory problem, especially due to the lack of complete crashes of the OS and total lack of diagnostic information to go off of. I mean, I don't have as much experience as others obviously, as this is just my personal computer, and not my job or anything, but normally, when I have a problem, it is a bit more obvious as to what is causing it. There is usually some diagnostic information to read too that points in the right direction, but this had nothing. Thank you for the advice though! I appreciate your help.

    EDIT: Forgot to mention, I did get new RAM from Corsair, and it now works perfectly. I was skeptical about the Vengeance RAM as I've had 3 dead sets of it in the past, but they kindly provided me with Dominator RAM, which hopefully is less prone to errors due to the higher quality... But it had 0 errors on Memtest and everything works perfectly now!
      My Computer


  9. Posts : 19,516
    W11+W11 Developer Insider + Linux
       #9

    That's great, shouldn't have to tolerate bad hardware, specially when new.
      My Computers


 

  Related Discussions
Our Sites
Site Links
About Us
Windows 10 Forums is an independent web site and has not been authorized, sponsored, or otherwise approved by Microsoft Corporation. "Windows 10" and related materials are trademarks of Microsoft Corp.

© Designer Media Ltd
All times are GMT -5. The time now is 13:47.
Find Us




Windows 10 Forums