New AMD PC blackscreen crashes while gaming

The AMD Ryzen Master app may not have been installed at the time of the Driver Verifier BSODs, but a driver for it most certainly was, because that's what Driver Verifier failed!

You might try using the AMD Cleanup Tool to remove any AMD drivers from your system and prepare it for new driver installs. Note that this doesn't uninstall chipset drivers. I would then suggest you download and install the AMD graphics driver from the ASRock website for your motherboard, but I would not at this stage install any other AMD software.

In addition, the ASRock website for your motherboard contains a very recent BIOS update (3/6/2024, version 2.08) containing updated AGESA microcode - actually there have been two BIOS updates containing AGESA microcode since the 2.02 BIOS version that you have (dated 11/17/2023). With AMD processors it's usually important to keep up to date with AGESA microcode updates. I would update to the latest BIOS version.

See whether that solves this problem, which does seem to be related to partially installed/uninstalled AMD software/drivers.
 
Last edited:
ubuysa said:
The AMD Ryzen Master app may not have been installed at the time of the Driver Verifier BSODs, but a driver for it most certainly was, because that's what Driver Verifier failed!

You might try using the AMD Cleanup Tool to remove any AMD drivers from your system and prepare it for new driver installs. Note that this doesn't uninstall chipset drivers. I would then suggest you download and install the AMD graphics driver from the ASRock website for your motherboard, but I would not at this stage install any other AMD software.

In addition, the ASRock website for your motherboard contains a very recent BIOS update (3/6/2024, version 2.08) containing updated AGESA microcode - actually there have been two BIOS updates containing AGESA microcode since the 2.02 BIOS version that you have (dated 11/17/2023). With AMD processors it's usually important to keep up to date with AGESA microcode updates. I would update to the latest BIOS version.

See whether that solves this problem, which does seem to be related to partially installed/uninstalled AMD software/drivers.
Click to expand...

Yes, seems odd that drivers for it was installed but not the application, but what do I know :D Yesterday I tried removing GPU drivers and installed PRO version of drivers from AMD (aimed more towards 3d modeling I think). Ran quite long before crash, 1,5 hours or something. And before this I also managed to uninstall the Ryzen Master app, looked in registry and it was also gone from there. After crash it was back in the registry. Attaching a screen of that one.

I will try your suggestions with removing old gpu drivers with amd software and installing drivers from Asrock, they are a lot older though? Should I also download chipsetdrivers from Asrock and uninstall my current? Current drivers are newest from AMD. I will also update BIOS.
 

Attachments

  • ryzen.jpg
    ryzen.jpg
    35.5 KB · Views: 3
One would hope that the AMD Cleanup Tool will cleanup the registry entries too. If you're in any doubt then export that registry entry (and any similar) and then delete it/them. Reboot and see how things are. According to what I'm reading the AMD Cleanup Tool should also remove Adrenalin but I see no mention of Ryzen Master. If the tool doesn't remove Ryzen Master then you might try the instructions here (which I've not tried nor tested). I think this is a problem with AMD tools that you don't really need except in extreme cases, like Adrenalin and Ryzen Master, so completely removing all traces of them will clearly show whether they really were the root cause.

When you're dealing with BSODs that you suspect may be driver related it's always wise to go back to the motherboard vendor's versions, because you know for sure that they are compatible with your board. That's an important level of confidence. In addition, you don't always need the very latest driver version - unless it contains fixes and/or features that you need.
 
ubuysa said:
One would hope that the AMD Cleanup Tool will cleanup the registry entries too. If you're in any doubt then export that registry entry (and any similar) and then delete it/them. Reboot and see how things are. According to what I'm reading the AMD Cleanup Tool should also remove Adrenalin but I see no mention of Ryzen Master. If the tool doesn't remove Ryzen Master then you might try the instructions here (which I've not tried nor tested). I think this is a problem with AMD tools that you don't really need except in extreme cases, like Adrenalin and Ryzen Master, so completely removing all traces of them will clearly show whether they really were the root cause.

When you're dealing with BSODs that you suspect may be driver related it's always wise to go back to the motherboard vendor's versions, because you know for sure that they are compatible with your board. That's an important level of confidence. In addition, you don't always need the very latest driver version - unless it contains fixes and/or features that you need.
Click to expand...

Well it seems like DDU cleaned up the registry since it was gone but it got created again after restart/while I was ingame perhaps? I have tried deleting it from registry but after reboot its back. Will try the uninstall process from your link. And change chipset and gpu drivers to the one on motherboard support page. And flash to newest bios. I noticed i got a new livekernel minidump from the latest crash wich is pointing towards USB3 drivers, will attach it in post.


A USB device failed enumeration. This bugcheck suggests that there is a problem with your hardware system configuration.
 

Attachments

I also noticed while looking over drivers in Device Manager that alot of the drivers are dated to 2006 / 2009? Is this normal? Seems very strange?
 

Attachments

  • cpu.jpg
    cpu.jpg
    43 KB · Views: 2
  • m2.jpg
    m2.jpg
    44.8 KB · Views: 2
  • monitor.jpg
    monitor.jpg
    45.4 KB · Views: 2
It is normal for Microsoft drivers, you'll find a few with install dates in the future as well.

The live kernel dump isn't anything to worry about. These are failures from which Windows recovered and this just looks like a USB3 device that failed and was recovered. We wouldn't normally look at live kernel reports, all they do is give us a feel for devices/drivers that may be becoming flaky.

BTW. It wasn't DDU I suggested you run it was the AMD Cleanup Tool...
 
Okay I see. Yes, I used DDU prior to your comment about AMD software. I have now done those steps (amd cleanup), uninstalled chipset and installed drivers from Motherboard page (GPU only drivers, no software and chipset) and flashed to newest BIOS version. So far so good 1,5h into rdr2 running, will report back later tonight if it crashes or if that indeed has fixed things. Cheers
 
It blackscreen restart crashed about 2h in, unfortunenately..Seems very strange it crashes after so long, could it stiill be temp issues somehow? Is there a program for logging temps so i can see last temps before crashing? Unsure about direction now, should I try with driver verifier again and get dumps? Or remove dGPU and try only iGPU?
 
For temps use HWMonitor (free). Expand the CPU temps section so we can see each core and then post a screenshot of that display at idle and another screenshot of that display when at the biggest load you can throw at it.

Then remove the graphics card and try it on just the iGPU.
 
ubuysa said:
For temps use HWMonitor (free). Expand the CPU temps section so we can see each core and then post a screenshot of that display at idle and another screenshot of that display when at the biggest load you can throw at it.

Then remove the graphics card and try it on just the iGPU.
Click to expand...
Logged the whole gaming session with HWinfo (didnt find way to log with Hwmonitor) until crash, found a good viewer "HwInfo Log Viewer" if you have that you can check everything. Otherwise I can provide prints but maybe easier with logviewer? I noticed some things, temp seems fine, but VRAM usage keeps increasing untill almost maxed out and thats when I get blackscreen crash+restart. Also, virtual memory seems to be decreasing until going very low right when it crashes also. Could this point to something specific software related or hardware faults? Will try with iGPU tomorrow unless this new find is of significance. Thanks so much for helping!!
 

Attachments

I don't really want to have to trawl through all that data, a simple screenshot of the temp display will do - but I do want to see a CPU temp display at idle AND at maximum load.

When you talk about it crashing, what do you mean?....
  • Does it BSOD (if so run the SysnativeBSOD CollectyionApp again)?
  • Does it crash to the desktop?
  • Does the whole system crash and restart with no BSOD?
I'd really like to see how it performs without the 6950 graphics card installed.
 
ubuysa said:
I don't really want to have to trawl through all that data, a simple screenshot of the temp display will do - but I do want to see a CPU temp display at idle AND at maximum load.

When you talk about it crashing, what do you mean?....
  • Does it BSOD (if so run the SysnativeBSOD CollectyionApp again)?
  • Does it crash to the desktop?
  • Does the whole system crash and restart with no BSOD?
I'd really like to see how it performs without the 6950 graphics card installed.
Click to expand...

Its 99% of the time screen turns black and it reboots.

Something very strange now when Im running OCCT to get CPU temps under load, it used to boost up to around 85C almost instantly and stay there when I was running stresstest before, now its at 44C while running OCCT "CPU large extreme" test..with 100% use of cpu..and I feel the air is cool from cpufans. Ran a linpack test and it got to around 60C..Only running with one RAM, could this be from that? Otherwise I have no clue to why it suddenly is such low temps..Trying to run "Power" stress test from occt now seems like temp slowly is increasing from that one so will post ss after 1h of that test.
Attaching ss of hwmonitor from earlier tests
 

Attachments

Here is ss, got up to around 80C but nothing like before when I was stresstesting. And I ran benchmark now in OCCT and got very low score, much lower than when I ran same test when newly built. And CPU only draws around 80W max when it should be 168.5 W.. so something is up, question is what hehe..attaching OCCT score also. Seems like CPU or Motherboard is broken? Or PSU?
 

Attachments

So I tested some more, in OCCT power test when I turn on XMP in BIOS , CPU power maxing at 130W and temps go up to 85C almost instantly and stays there, running CPU test gives errors almost immediately (was like this before also with XMP on). But why would turning XMP on make the CPU draw much more power? 4800mhz vs 6000mhz in memory speed cant change that much?...Seems like with XMP on it behaves more "like it should", actually drawing some power and delivering better results in benches. With it off it was basically close to idle temps on many benchmarks..anyway, attaching screenshot from Hwmonitor with XMP on
 

Attachments

Well, after turning XMP off again it still behaves like with XMP on now temp wise, goes up to 85C instantly and draws a little less power but still more than the 80W before (100W now) really dont know what was the deal before. Most random events ever, i guess it had something to do with only having one RAM or something. No clue. I will test tomorrow with only iGPU and see what info that can give.
 
ubuysa said:
I don't really want to have to trawl through all that data, a simple screenshot of the temp display will do - but I do want to see a CPU temp display at idle AND at maximum load.

When you talk about it crashing, what do you mean?....
  • Does it BSOD (if so run the SysnativeBSOD CollectyionApp again)?
  • Does it crash to the desktop?
  • Does the whole system crash and restart with no BSOD?
I'd really like to see how it performs without the 6950 graphics card installed.
Click to expand...
I have taken out the GPU now, what testing should I do with the iGPU?
 
Running Driver Verifier and Rdr2 right now, I think that gave BSOD crash last time with the 6950 pretty quickly and it has been running for 20 min now, ofc very bad fps and quality so RAM and CPU is not hot at all..But seems like could be faulty GPU? Also looked on teeth of GPU and saw that 2 are shorter, got scared but after googling this is normal?
 
Yeah after testing without the GPU it seems like it doesnt crash, but Im guessing this doesnt necessarily mean its the GPU thats faulty? Could be PSU, CPU, motherboard also? or RAM? SO annoying dont know if I should RMA the GPU or not..
 
It's not going to be RAM or CPU, it would crash on the iGPU too if it were. It could conceivably be a PSU or motherboard problem, but it's massively more likely to be a faulty graphics card. The only way to be sure is to swap test it.
 

Has Sysnative Forums helped you? Please consider donating to help us support the site!

Back
Top