Discord screensharing contributing to PC freeze/crash

[shovenose]

I am having an intermittent lockup on my PC that requires me to do a hard reboot.

It has been an intermittent problem for a while, and since I started having the problem I've done many hardware and software changes.

Initial configuration when I started having the problem was Win11 on a 1TB SSD. Switched to Win10 on a 2TB SSD. Now I'm back to Win11 on a new 4TB SSD. So that's three different SSDs with a clean install each time so I don't think it's an OS issue or an SSD issue.

32GB to 64GB RAM, completely different brand of RAM. Memory tests were fine on old and new RAM anyway.

The computer typically works fine under all sorts of load so I don't believe the issue is power supply or temperature related. Prior to the most recent lockup, I had been using my computer for several days without trouble. I mainly use it for web development, web browsing, and occasionally gaming, and at night I leave it running. It never crashes while I'm not using it. I can play intensive games like Cities Skylines 2 just fine.

But then this morning I get on a Discord call and start screensharing and within a minute it locks up and I have to hold the power button down and start up again.

I do notice Discord screensharing has a possibly related issue where if I have it on monitor 2, and try to share from monitor 1, it doesn't work. It basically reboots Discord. If I want to screenshare from monitor 1 I have to move the Discord window onto monitor 1 and then screenshare monitor 1 and it works fine. My setup is slightly unconventional, I have two monitors, one vertical, one horizontal, not that it should matter...

I will say I never had this issue with my Radeon 7900XT but I downgraded to a Radeon RX6950XT about 6 months ago and I think that's when I think I started having this issue. But I don't remember as it's been a while and like I said I've reinstalled the OS multiple times so I don't have the logs. I do have the latest AMD drivers.

Do I have an underlying hardware issue or is the issue with Discord? I can't say for certain it only happens during Discord calls with screensharing but it seems to be a trigger.

Event viewer gives me nothing useful beyond "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."

Thanks for any input. I'm getting frustrated.

 

[ShrimpBrime]

It's such a broad spectrum issue....

So, memory timings menu. Disable power-down mode. This sometimes helps with event 41 errors. This setting prevents the memory from going into low power states. You can also experiment with windows Power plans.

The system only suddenly restarts? Event 41? From the past, 2000 and 3000 series AMD processors had this issue. Half the time the cpu was just bad. The other half the system didn't like the memory installed.

 

[shovenose]

It's such a broad spectrum issue....

So, memory timings menu. Disable power-down mode. This sometimes helps with event 41 errors. This setting prevents the memory from going into low power states. You can also experiment with windows Power plans.

The system only suddenly restarts? Event 41? From the past, 2000 and 3000 series AMD processors had this issue. Half the time the cpu was just bad. The other half the system didn't like the memory installed.

I didn't see anything power related in the memory menu in the UEFI BIOS. Maybe worth noting I did recently update the BIOS to the latest version from the Asus website as the one one I was running before was quite old. It made no difference.

I just changed my power plan from Balanced to Performance. Will see if it makes any difference.

I would sure hope the CPU is not defective, that would be weird. I can't say I've ever had a bad CPU in my entire history with computers!

 

[ShrimpBrime]

I didn't see anything power related in the memory menu in the UEFI BIOS. Maybe worth noting I did recently update the BIOS to the latest version from the Asus website as the one one I was running before was quite old. It made no difference.

I just changed my power plan from Balanced to Performance. Will see if it makes any difference.

I would sure hope the CPU is not defective, that would be weird. I can't say I've ever had a bad CPU in my entire history with computers!

It was more an AMD problem.

But event 41 could literally be anything plugged into the motherboard that can cause it. Typically not a software issue that I'm aware of. Obviously you searched and found all kinds of the same issue with very few resolves without replacement of some piece of hardware.

One of the moderators at LTT forums had this issue. With the above mentioned generation of AMD cpu. His resolve was to replace the entire system because 1 piece at a time led nowhere.

 

[Dr. Dro]

Downgrade your graphics driver to 23.9.3 (so before their latest changes to the encoder) and see if the issue is resolved. Probably the AMD driver being the AMD driver again. I know it says abrupt shutdown/power loss, but could just be the whole thing resetting the computer...

 

[RJARRRPCGP]

Nothing but the message that Windows wasn't shut down properly->Sounds like the infamous Windows 11 "blank screen with backlight still lit" crashes. (The screen turns blank, with the backlight still lit and you have to hold down the power button)

 

[Calenhad]

I am having an intermittent lockup on my PC that requires me to do a hard reboot.

It has been an intermittent problem for a while, and since I started having the problem I've done many hardware and software changes.

Initial configuration when I started having the problem was Win11 on a 1TB SSD. Switched to Win10 on a 2TB SSD. Now I'm back to Win11 on a new 4TB SSD. So that's three different SSDs with a clean install each time so I don't think it's an OS issue or an SSD issue.

32GB to 64GB RAM, completely different brand of RAM. Memory tests were fine on old and new RAM anyway.

The computer typically works fine under all sorts of load so I don't believe the issue is power supply or temperature related. Prior to the most recent lockup, I had been using my computer for several days without trouble. I mainly use it for web development, web browsing, and occasionally gaming, and at night I leave it running. It never crashes while I'm not using it. I can play intensive games like Cities Skylines 2 just fine.

But then this morning I get on a Discord call and start screensharing and within a minute it locks up and I have to hold the power button down and start up again.

I do notice Discord screensharing has a possibly related issue where if I have it on monitor 2, and try to share from monitor 1, it doesn't work. It basically reboots Discord. If I want to screenshare from monitor 1 I have to move the Discord window onto monitor 1 and then screenshare monitor 1 and it works fine. My setup is slightly unconventional, I have two monitors, one vertical, one horizontal, not that it should matter...

I will say I never had this issue with my Radeon 7900XT but I downgraded to a Radeon RX6950XT about 6 months ago and I think that's when I think I started having this issue. But I don't remember as it's been a while and like I said I've reinstalled the OS multiple times so I don't have the logs. I do have the latest AMD drivers.

Do I have an underlying hardware issue or is the issue with Discord? I can't say for certain it only happens during Discord calls with screensharing but it seems to be a trigger.

Event viewer gives me nothing useful beyond "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."

Thanks for any input. I'm getting frustrated.

What memory test did you use? They are not built the same. But I have rarely seen unstable memory only cause problems in a singular program, such as Discord. So I would not put that as one of the top three suspects for now.

One vertical and one horisontal monitor in multimonitor is more common than you think. I am using two horisontal monitors these days. But I have had one + one before and even one vertical and two horisontal. I am on Nvidia graphics so that could absolutely be a factor, I am able to share monitor one while having Discord on monitor two. Then again I can't remember the last time my computer crashed either.

Event 41 is 100% useless for finding a cause. It literally only says that your computer did not shut down properly. Why? No one knows. Could be a power outage in your area based on that log entry.

And I would definitely try what Dro suggests:

Downgrade your graphics driver to 23.9.3 (so before their latest changes to the encoder) and see if the issue is resolved.

 

[shovenose]

Nothing but the message that Windows wasn't shut down properly->Sounds like the infamous Windows 11 "blank screen with backlight still lit" crashes. (The screen turns blank, with the backlight still lit and you have to hold down the power button)

Yes, the screen backlit still. Sometimes blank screen, sometimes gray.

What memory test did you use? They are not built the same. But I have rarely seen unstable memory only cause problems in a singular program, such as Discord. So I would not put that as one of the top three suspects for now.

One vertical and one horisontal monitor in multimonitor is more common than you think. I am using two horisontal monitors these days. But I have had one + one before and even one vertical and two horisontal. I am on Nvidia graphics so that could absolutely be a factor, I am able to share monitor one while having Discord on monitor two. Then again I can't remember the last time my computer crashed either.

Event 41 is 100% useless for finding a cause. It literally only says that your computer did not shut down properly. Why? No one knows. Could be a power outage in your area based on that log entry.

And I would definitely try what Dro suggests:

I will try this! AMD drivers grr.

OK, I am going to try something different... I went to the "Pro" drivers latest version from 22Q4. I'll update in a few days if it makes a difference.

 

[RJARRRPCGP]

Yes, the screen backlit still. Sometimes blank screen, sometimes gray.

Chromium is the most likely to cause this issue! I never got it in games. Most likely would get it in Chromium-based Edge. But I haven't saw this in months now, IIRC.
It's most likely to be a combination of the following:

Running anything using Chromium and your monitor is a late-model Samsung. (not TVs)

Windows 11, additionally has an issue where it may randomly crash with the same symptom when trying to change the refresh rate, IIRC!

There is no error being reported, when this happens. Instead, it crashes in a more-Windows 95-like manner, where a simple software problem easily takes down Windows.

 

[shovenose]

Chromium is the most likely to cause this issue! I never got it in games. Most likely would get it in Chromium-based Edge. But I haven't saw this in months now, IIRC.
It's most likely to be a combination of the following:

Running anything using Chromium and your monitor is a late-model Samsung. (not TVs)

Windows 11, additionally has an issue where it may randomly crash with the same symptom when trying to change the refresh rate, IIRC!

Well I did a test this morning since downgrading my video card driver. Just one Discord call with screenshare and it worked out OK. I will update once I have done more testing to see if the issue is actually resolved.

I did have Edge open this morning but I don't always.

My monitors are Dell not Samsung but I can't imagine that has anything to do with it?

 

[RJARRRPCGP]

The system only suddenly restarts? From the past, 2000 and 3000 series AMD processors had this issue. Half the time the cpu was just bad. The other half the system didn't like the memory installed.

Sounds like the dreaded WHEA "Cache Hierarchy Error" issue, which is more likely to happen randomly with high temps, even at stock.

Symptom would look like this to many users:

You double-click to start a game and it black-screen-reboots. And you only see the WHEA error reported in the event log afterwards.

 

[ShrimpBrime]

Sounds like the dreaded WHEA "Cache Hierarchy Error" issue, which is more likely to happen randomly with high temps, even at stock.

Symptom would look like this to many users:

You double-click to start a game and it black-screen-reboots. And you only see the WHEA error reported in the event log afterwards.

Most of this issue was a complaint that the system was just idle and it would just random restart or the person left it idle to come back to that black screen.

But like I was saying, it was more common on AMD platforms 2000 and 3000 series. I chalked it up as memory compatibility issues while I experienced the same problem, but resolved it with a decent set of Samsung B-die memory, while my corsair Vengeance made the issue prominent and clear. My son uses this memory, 3200mhz, but I have it set to 2666mhz to prevent any issues with that set of memory.

But I can't stand trying to help diagnose error41. It literally could be anything plugged into the motherboard up to and also including the board it's self.

 

[ir_cow]

Straight up I assume it's memory related on AMD systems. Disable EXPO/XMP and see if the crashes go away. That will narrow it down