• Happy holidays, folks! Thanks to each and every one of you for being part of the Tom's Hardware community!

Question Black Screen Crash & DRAM light when gaming ?

Oct 27, 2024
8
0
10
I built a PC in September and when playing games I have been getting inconsistent random black screen crashes with a DRAM light on the mother board.

After building it I have had no issues when testing with Prime95 and FurMark 2 concurrently. Mundane tasks like watching TV have no issues.

When playing Black Myth Wukong after the initial build it would crash randomly but there were no issues playing COD. I double checked my driver updates and then updated my BIOS. That did not fix the issue. I opened the case, reseated the CPU and cooler, removed and reseated the RAM as well as the graphics card.

The computer continues crashes to a black screen with the DRAM light. The PC Lights and case fans stay on and after about 15 seconds I can hear the windows boot sound, but the screen remains black. The crashes used to be more frequent, but I adjusted the RAM speed to 5200 mt/s and turned off EXPO based on AMD CPU & RAM recommendations, because EXPO is overclocking and can cause instability.

I was able to play Black Myth and a few other games in my library like Dead Cells and Kingdom Come. Once I got my old computer set up for my wife, we started playing It Takes Two and I started experiencing crashes again! I thought it would be tied to heavy workload games, but it is still inconsistent. Now that the new COD is out, I cannot get through one match before it crashes.

I ran the Windows Memory Diagnostic tool, and the report came black clean. After searching the forums, this seems to be a common issue with MSI motherboards but no one has a consistent answer. My next step will be to reset the BIOS to factory settings, test and if it still fails I will run with a single stick of RAM and see where that gets me. This is a frustrating as ii just want to be able to consistently play games with my friends.

I have searched the forums and found answers but have had no luck:
Drivers/BIOS - Updated a few times over the two months

EXPO/XMP Memory profiles - disabled

CPU voltages (auto vs manual) - I adjusted

Undersized PSU – I have a 1000w PSU, shouldn’t be a factor. https://forums.tomshardware.com/thr...dram-error-light-on-what-should-i-do.3707429/

AMD Audio Driver - I switched to HD Audio driver, and it had no effect. https://reddit.com/r/AMDHelp/comments/iehbf9/amd_games_black_screen_fix/

AMD adaptive Sync (free sync) – off

Capacitor overheating / AOI touching capacitor – Corsair AOI not touching and I checked all capacitors. www.youtube.com/watch?v=UHXbQ_kyebw&ab_channel=Zach%27sTechTurf


Some additional info:
I am using a triple monitor set up now, but I was still crashing on a single monitor and dual. I have two monitors on display port and a TV on HDMI plugged into the integrated graphics slot.

The two sticks of RAM are in the correct slots per the MSI manual.

HWInfo does not show any weird thermals or voltages.

I have no popup or error messages that lead me to believe that it is a specific issue.

Specs
Ryzen 5 7600X
Gigabyte Radeon RX 6800XT
MSI MAG B650 Mortar WIFI
Corsair VENGEANCE 32GB (2x16GB) DDR5 DRAM 6000MT/s CL30 - CMH32GX5M2B6000Z30
Corsair r1000e PSU
Crucial P3 Plus 4TB PCIe Gen4 3D NAND NVMe M.2
Windows 11 Pro 23H2
 
Did you use only the cables that came with the installed PSU?

Disk drive(s): make, model, capacity, how full?

Have you checked that all connectors, cards, RAM, jumpers, and case connections are fully and firmly in place?

I would also look for possible shorts via stand-offs and around the I/O panel and other ports.

Are there any error codes, warnings, or even informational events being captured by Reliability History/Monitor and/or Event Viewer?
 
  • Like
Reactions: Itsjusttrevor
Did you use only the cables that came with the installed PSU?

Disk drive(s): make, model, capacity, how full?

Have you checked that all connectors, cards, RAM, jumpers, and case connections are fully and firmly in place?

I would also look for possible shorts via stand-offs and around the I/O panel and other ports.

Are there any error codes, warnings, or even informational events being captured by Reliability History/Monitor and/or Event Viewer?
Thank you for the Quick Reply Ralston18. I just updated the BIOS, reset the BIOS settings* and played COD until it crashed.

All cords are manufacture provided.

2 m.2 SSDs

Crucial P3 Plus 4TB 50% Full
Crucial T500 2TB 10% Full

I have checked all off the connections, I will revisit them again if it continues to persist. I have not removed the motherboard to check for shorts, but i will inspect later.

The Reliability Monitor says "Windows was not shut down properly" across the 7 crashes today. Additionally I am getting fails with Razer "Game Manager Services.exe". I have disabled it and have yet to retry.

Event Viewer in the last hour has 1 Critical event, 19 errors and 45 warnings in the last hour.
C. Event 41 Kernel-Power - The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
E. Kernel-EventTracing - Error setting traits on Provider {8444a4fb-d8d3-4f38-84f8-89960a1ef12f}. Error: 0xC0000001
E. Application Error - Faulting application name: GameManagerService3.exe, version: 3.5.0.370, time stamp: 0x622f7973
E. Service Control Manager - The AMDRyzenMasterDriverV27 service failed to start due to the following error: Cannot create a file when that file already exists.
W.DistributedCOM - DCOM got error "1115" attempting to start the service SecurityHealthService with arguments "Unavailable" in order to run the server: {470B9B9B-0E95-4963-B265-5D58E5808C3D}
W. LSA (LsaSrv) - LSA package is not signed as expected. This can cause unexpected behavior with Credential Guard. PackageName: msv1_0

I would post screenshots if i knew how too.
 
Screenshots can be posted here via imgur (www.imgur.com).

Look for Reliability History/Monitor and Event Viewer entries that precede the "Windows was not shut down properly" entries.

An unexpected shutdown can and does corrupt files. Then that corruption causes even more problems.

Run "dism" and "sfc /scannow" to find and fix any corrupted files.

https://www.windowscentral.com/how-use-dism-command-line-utility-repair-windows-10-image

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

Also look in Update History for any failed or problem updates.
 
Screenshots can be posted here via imgur (www.imgur.com).

Look for Reliability History/Monitor and Event Viewer entries that precede the "Windows was not shut down properly" entries.

An unexpected shutdown can and does corrupt files. Then that corruption causes even more problems.

Run "dism" and "sfc /scannow" to find and fix any corrupted files.

https://www.windowscentral.com/how-use-dism-command-line-utility-repair-windows-10-image

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

Also look in Update History for any failed or problem updates.
SFC came back clean. I have ran it a few times

I'll read the article and figure out how to run DISM properly
 
RAM is not compatible with your motherboard or one or more RAM are defective. The memory timings or voltages in BIOS/UEFI settings are not configured correctly for your RAM modules
Compatibility may be the issue here. After running DDU and reinstalling drivers, call of duty ran for about three solid hours, but then the computer eventually crashed. That is with a BIOS reset and factory JDEC timings . I'm not sure what else to do, other than replace the DDR5 sticks with Kingston or another AMD recommended manufacture. Its just strange that a the Prime95 memory test nor the memory diagnostic showed any issues.

And as an update, SFC is still clean Reliability History is showing the same errors.
 
I purchased a new RAM kit that AMD says is compatible. Details below.
Kingston Fury 2x 16gb - KF556C36BBEAK2-32 (4800 / 40-39-39--5600 / 36-38-38) https://www.amd.com/en/products/processors/ryzen-compatible-memory.html

After a power off swap I had to reset the BIOS by jumping the clear CMOS pins, because it was showing CPU/DRAM fail LEDs. After the memory training, it booted normally. SFC came back clean. After 1 COD Match, the computer reset again.

Event Viewer and Reliability Monitor still read the same. I did check all of the cables and the capacitors. but I have not had the chance to remove the motherboard and finely comb though everything. I wondering if this issue is software and not hardware, but there is nothing on this computer besides steam and some generic apps. The only thing nonstandard was the windows install. The install was a unattended install of windows. Link and info below.
(https://schneegans.de/windows/unattend-generator/)
View: https://www.youtube.com/watch?v=VapEEf4k7Eo&ab_channel=CyberCPUTech


I plan to buy more thermal paste in order to remove the CPU to check the pins and reseat the CPU.

ADD. note:
I am American but currently live in Japan and the language barrier makes it hard to get help. Additionally, as previously stated, the other people online do not seem to have a good resolution to this problem, even with manufacture support. This is why I am reaching out to those much smarter than me in this forum.
 
Last edited:
Hey dudes,

It took a while but I have resolved the issue.

I purchased a new mother board, swapped only the minimum required parts and it still crashed. So I bit the bullet and bought another newer graphics card. Since its been installed (4 days) I have not had a single issues. I have since swapped the GPU into a friends system and it crashed there too.

All to say something is wrong with my GPU. It is still under warranty so I am reaching out to Gigabyte to RMA the hardware.

I really appreciate all of you heroes that have tried to help me figure out this problem.