Question PC black screen restarting/crashing even with new parts

[flsanji]

For a while now I've got an issue with my PC, it keeps crashing when gaming but now it restarted during browsing chrome.

Specs:

CPU: Ryzen 7 7700
CPU Fan: be quiet! Pure Rock 2
Case: be quiet! Pure Base 500
GPU: AMD RX 7800 XT
RAM: Kingston Fury KF560C36BBE 16gb 1x1 6000mhz (set at 4800mhz in bios at default)
SSD: Samsung 980 1TB
MOBO: ASUS TUF Gaming B650-PLUS
PSU: be quiet! Pure Power 12 M 750W
OS: Windows 11 Pro

This is my current setup. However I did change some parts over these past months to see if the issue would be resolved with different parts.


Old parts that I changed:


GPU: PNY RTX 4080 16GB VERTO
RAM: Kingston Fury Renegade 16gb 1x1 6000mhz
SSD: Kingston NV2 1TB
MOBO: ASUS PRIME B650-PLUS

I also sent my PSU for RMA.
And changed my monitor from dual setup to a complete new single monitor, different cables.


It started with PC suddenly getting this blackscreen and audio crash. Event viewer showed nvidia driver not responding so I swapped it for an AMD 7800XT, fresh windows reinstall. After a month or so I then got the same exact issue, but this time it said amd ryzen master drive failed to start system cannot find file. I started sending in different PC parts over couple months. But each time it would take a week or so for the same issue to reappear. The temperatures also seem fine.


What I tried:

Using older gpu drivers
Updating chipset driver
Updating bios driver
Updating Windows 11
Used different power outlet
Used different cables for monitor and psu(from rma)
Used different monitors
Each new hardware, fresh windows 11 pro installation.
Used different parts
Used different memory slots(i got 1 stick of 16gb)
Played around with driver settings(only started playing with settings after the issue appeared)
Tried different power plans on W11
Disabled automatic driver update on windows 11
No antivirus except windows defender

I would also not get any dump files on my PRIME B650-PLUS but with my new mobo I found the dumpfile, using WhoCrashed.

Dump type Kernel memory dump
BugCheck name VIDEO_DXGKRNL_LIVEDUMP


Could anyone take a look at this?

Edit: I forgot to mention that with each new part it would take at least a week to show the said issue, with time the crashes would happen more frequent.

 

[Lutfij]

Each new hardware, fresh windows 11 pro installation
Where did you source the installer for the OS?

Updating bios driver
+
MOBO: ASUS TUF Gaming B650-PLUS
What BIOS version are you on at this moment of time?

RAM: Kingston Fury KF560C36BBE 16gb 1x1 6000mhz (set at 4800mhz in bios at default)
Did you enable X.M.P in BIOS?

If you're using DDU, remove all GPU drivers, Intel, Nvidia and AMD, not just the GPU you have installed on your platform. When reinstalling your GPU drivers, use the latest and do so in an elevated command, i.e, Right click installer>Run as Administrator.

 

[flsanji]

Each new hardware, fresh windows 11 pro installation
Where did you source the installer for the OS?

Updating bios driver
+
MOBO: ASUS TUF Gaming B650-PLUS
What BIOS version are you on at this moment of time?

RAM: Kingston Fury KF560C36BBE 16gb 1x1 6000mhz (set at 4800mhz in bios at default)
Did you enable X.M.P in BIOS?

If you're using DDU, remove all GPU drivers, Intel, Nvidia and AMD, not just the GPU you have installed on your platform. When reinstalling your GPU drivers, use the latest and do so in an elevated command, i.e, Right click installer>Run as Administrator.

Hi, thanks for replying!

I used the installation media for USB from microsoft's website. I didnt use the same file over and over.
I haven't updated the new MOBO (yet) it's on the version 1811 x64. I did update it on my old MOBO which didn't solve/change anything before getting this current MOBO.

XMP/EXPO is by default disabled in BIOS.

I forgot to mention that I also tried DDU.

 

[flsanji]

Just now playing Dragon's Dogma 2, I got the black screen crash again. It didn't restart immediately, I got a sudden black screen at first while I could hear the sound of the game it quickly got distorted as if it was stuck in a loop but very distorted after 5 to 10 seconds of that my PC restarted. No dumpfile this time.

When I look at event viewer I get the following critical message:
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Source being Kernel-Power.

I'm on 24.3.1 adrenaline. Used DDU in safe boot and ran the file as administrator.

Could someone take a look at this please? Thanks in advance!

 

[flsanji]

Update:

The crash only happens once a day at the moment, it gradually increases and goes back to once a day when I install a new part(either rma or totally new product) or reinstall W11(maybe its just a coincidence).

I have a feeling it's my PSU just not being compatible with my GPU. It is a multi-rail PSU.
Also I was thinking of having 2 separate pci cables going from psu to gpu. But my PSU doesn't have enough slots for that.

This is the be quiet! pure power 12M 750w PSU that I've been using. The blue highlighted area is the PCIe slot I'm using it for the 7800XT. The 7800xt needs 2x 8 pin connectors. I'm using the PCI cable that was included with the PSU.

Is it possible that the PSU can't handle the power spikes of the 7800xt? (or the 4080 before that when it was using PCIe 5.0)

I'm just brainstorming here, hopefully someone can share some of their wisdom. Thanks!

 

[flsanji]

Another update:

I had zero issues for like a week , I was playing Dragon's Dogma 2. Max settings on 1440p except for Tree quality and shadow. After I turned these two settings to max the system became very unstable, keep getting crashes. I am not getting any dump files sadly.

However when I look at Event Viewer I am seeing Critical > The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly. Event 41, Kernel-Power.

That's all the critical I can find at the time the PC crashed.

Can anyone look at this? Thanks

 

[flsanji]

Sadly my PC restarting again while gaming even after setting ram to 5200mhz (matching the ryzen 7700). And yet no dump file whatsoever.

Any ideas?

 

[kel_sceptic]

Reading your journey scares the hell out of me.
I'm exactly in the same spot with my system:
MB: X570 Aorus Pro v1.0
CPU: Ryzen 7 3700x
RAM: Corsair Vengeance CMW16GX4M2C3200C16 4x8G
GPU: MSI RTX 2060 Super Armor OC
PSU: Seasonic X-Series 850W (14 years old, trust worthy).


Since last summer since I've migrated to BIOS Update + Win11 + m2 ssd Samsung 980Pro + 2 more RAM sticks (had only 2x8GBs before), my pc started crashing once per month or so. I've tried everything, bios reset, bios downgrade, stock DDR 2133Mgz (I don't OC the CPU, nor the RAM beside XMP speeds customized).

Phase1 I have tried replacing the PSU with a Quiet Pure Power 12M 1000W. The day I've replaced it, after 10 mins of playing WoW. BTW: playing fornite, it crashed once in like ~10 moths.

Now I'm in Phase2 CPU+MB+RAM (7800x3d) ordered, waiting to replace them and see if the issue is still there.

Then Phase3, the GPU replaced.


At the same time, after upgrading to F38 in ~20 of april, the PC run without any crash for 3 weeks. But after reading your story with R7 7700, having the same issue ... I am really really dissapointed. If the crashes continue, I will never touch AMD until the price/performance difference is 40-50% in favor of AMD.


My only hopes for a stable system are that the GSKILL ram I have purchased are EXPO recommended by AMD EXPO page themselfs, and the rams are also present in the compatibility list of the Asus Prime x670-P MB I got.


What's your current status ?

After re-reading the face that you changed PSU+MB+RAM+GPU, except CPU ...I start to think its either SSD m2 Samsung 980Pro or the Windows 11. Or the combination of both since Windows 11 uses DirectStorage with that SSD.

GL HF,

TSO

 

[kel_sceptic]

Hi again,
In the past 2 days I've tried even removing the 2 RAM sticks I've bought last summer, since they were same SERIAL but different versions 4.31 (first set bought) and 4.32 (second set bought). But the issue still occured.

In the end, since I am still waiting on the DDR5 rams for the new platform, I ended up entering the community-support channel in the Gigabyte discord server.
There, after not even 2minutes, a great guy named @Sythen opened my eyes regarding the issues I'm facing.
The manifestation of "both monitors going to black screen, but sound and functionality continuing to work" mean that the video card goes into protection mode.

While I've checked the GPU temperatures and they were always lower then 80degrees C, it seems that I've not taken into consideration another indicator in HW Info, called "GPU Hot Spot Temperature", and that was reaching 106degrees C.

At @Sythen's recommandation, I have imediately made a thermal paste replace on the GPU, and after that, MIRACLE, the "GPU Hot Spot Temperature" didn't raised to more then 86degress in MAXIMUM LOAD.

The existent thermal paste was all dried up and was braking in tiny pieces.

After that, at least for the time being, there were no more crashes. Also, before, the GPU was throttleing badly and the maximum Power usage was ~130W, but after changing the paste it went up to 180W with lower temps and fan speeds.

I need to mention that in 26 years of experience and multiple PCs owned, I have never got the need to replace the thermal paste on the GPU, and so rarely did on the CPU. But that is mostly due to replacing the GPU mostly at every 2-3 years. But the current MSI 2060 Super Armor OC I got in 2019, so .... it still does good for what I use it at.


GL HF

Br,
TSO

 

[flsanji]

Hi kel_sceptic,

I just saw the notifications.

I was using a Kingston NV2 SSD (1b) before the Samsung, but it had no difference. So I don't think that's the culprit. I wanna say it's definitely the GPU but I had a PNY RTX 4080 before and it gave me weird green screen crash/restarts with Nvidia driver being the main issue in event viewer.

My PC has been stable ever since I changed CPU voltage from auto to 1,4 (weird right?). The only time it still occasionally restarts is when I turn on Instant Replay in AMD's driver. Right now I think it's either CPU or GPU ( bad luck on both cards maybe ).

Right now I'm just waiting on the new Nvidia cards either 5080 or 5090.

Also on your second post,

I also have never replaced the thermal paste on the GPU. But I'm glad it's working out for you! Please keep us updated to solve this mystery crash