hypervisor_error 0x20001 BSOD WHEN USING TWO MONITORS

[kiki4431]

Hello

 

I am trying to set up a sensor panel for my PC, with 7" screen, and AIDA64 running the sensors.

 

The screen connects with HDMI and uses USB power to work, so, I have plugged it in my GPU RTX 5070Ti where I also have my 4k monitor plugged with DP.

It worked fine connected like that, however, it's a 1080p screen, so I don't want to task my main GPU to show the sensor panel, since I want all it's power be allocated to my 4k monitor, for games and stuff.

Now I have a AMD 7600x CPU, which has a built-in GPU as well, not a powerfull one, since it's primary use is diagnostic purposes, but it's powerfull enough to run that sensor panel I belive, so I can make that iGPU take care of sensor panel, and leave the 5070ti to take care of other more demanding things, since it's just sitting there doing nothing, it's active, I mean devmgr sees it, task manager sees it, MSI Afterburner sees it, there is also driver for it on my system, but just the driver, no Radeon software or anything.

I plugged the HDMI to motherboard's output, and it worked... for about 5 secs, and than, hypervisor_error 20001 BSOD appeared, and it just froze at 0% forcing me to manually shut down the PC with a power button, so minidumps are not available.

Anyway, restarted the PC, leaving everything as it is, and than it worked normally, no issues, I actually spent about 3 hours designing and modifying gauges and stuff for my sensor panel, using the screen for preview, as i worked, than I got tired and went to sleep, leaving the PC running all night, in the morning, everything was the way I left it, no problems, shut down the PC, went to work, returned to it about 6 hours later to continue with designs, and after powering on, the moment desktop appears, I get the BSOD again, and again freezes at 0% forcing me to manually turn the PC off, restart, and it happens again, and again, five times in row, and than I just decided to unplug the screen, and now it works fine, so something about that screen or iGPU is making that BSOD, Also the display, that's supposed to be my sensor panel, is showing the background image of my desktop, even when the main monitor is displaying the BSOD. Never experienced that particular BSOD before, and also, for months my PC ran just fine, no issues, well appart from "low virtual memory" errors I got a few months back, but that was solved with reinstall of windows 11.

 

I also never used the iGPU on my PC, ever since I bought the 7600x, I've been using only descrete GPUs, first 3070Ti and now 5070Ti.

 

I would try to install full driver package, along with Radeon software, but Im afraid it might conflict with nVidia drivers, I mean, I used to work with multi-GPU PCs, but never like that (AMD+NVidia), it was always all AMD or all Nvidia.

 

Did anyone experience the issue described above, specificcally the dual monitor, one running of iGPU, and the other of dGPU?

if so, How did you solve the issue?, any advice is much appreciated

Thanks

 

specs:

AMD RYZEN 7600X (no OC appart from Advanced PBO, with just all limits (TDC/EDC/PWR) lifted, so no voltage/frequency tunning, no curve optimizer, nothing, I just put some high values there to make it "unlimited", still even under the most demanding load, it doesn't take more than 95W or so)

Corsair Vengeance RGB 2x16 GB 6000 Mhz (EXPO/XMP enabled)

Corsair H150i AIO with BeQuiet Pure wings 4 fans installed instead of original corsair ones, The rest of cooling in the case is fitted with 7 LianLi AL 120 fans connected to a AL controller

Case: 011D EVO LianLI

GPU: Gigabyte RTX5070Ti Gaming OC

PSU: LianLI Edge 1000W

MB: MSI B650 Tomahawk Wi-fi, latest BIOS

 

 

[Sawa Takahashi]

This is usually a hardware error related to the AMD power management driver. 

One easy thing to test is to disable PBO in the BIOS. If it doesn't help, you can always re-enable PBO afterwards.

Also, try to stress test a bit your system RAM and your CPU. Since the iGPU uses system memory, any instability between the two may cause the issue.

Lastly, running a small display on your graphics card will not impact performance unless your are running graphics and animation heavy stuff of the small display. You should benchmark performance on your mainscreen with and without that secondary monitor installed to see if it makes a difference at all.

Good luck !

[Bjoolz]

Go to C:\Windows\Minidump and check if you have any minidump files. If you do, go back to the Windows folder and copy the Minidump folder itself to the Downloads folder (You can use the desktop if you don't have OneDrive syncing files). Zip the copied folder and attach it to a post.

[kiki4431]

Thanks, I will try disabling PBO, I don't think it's doing much anyway, again even under most load it's pulling less than 95W even though limits are lifted, temps are howering about 88-90C (less than 95), so it's not slowing itself down to prevent overheating.

As for stress  testing CPU and RAM are both fine, I've tested them several times over the months, esspecially since I had other type of issue, which led me to belive, something might be wrong, but it wasn't.

 

update:, in the meantime, it worked again for hours, and today, I've got the BSOD again. (I've just experienced the BSOD, before writting this) and this time, it didn't freeze, it went to 100% and restarted on it's own, so now I have that minidump , so I can attach it.

Minidump.zip

[Smalahovet]

That error is rare, but can happen with cheap chinese displays using a cheap USB to HDMI adapter. 
This is the reason i got the USB-C version of the display for monitoring my temps. Cost 12 dollars at aliexpress but works. 

Or it can be firmware/driver issue. Like if it try hook up to the graphics at a kernel level. That will trip hypervisor checks. 

 

[Bjoolz]

3 hours ago, kiki4431 said:

Thanks, I will try disabling PBO, I don't think it's doing much anyway, again even under most load it's pulling less than 95W even though limits are lifted, temps are howering about 88-90C (less than 95), so it's not slowing itself down to prevent overheating.

As for stress  testing CPU and RAM are both fine, I've tested them several times over the months, esspecially since I had other type of issue, which led me to belive, something might be wrong, but it wasn't.

 

update:, in the meantime, it worked again for hours, and today, I've got the BSOD again. (I've just experienced the BSOD, before writting this) and this time, it didn't freeze, it went to 100% and restarted on it's own, so now I have that minidump , so I can attach it.

Minidump.zip 1.34 MB · 0 downloads

This one shows an error with the Nvidia driver. The Nvidia driver triggering a Hypervisor_Error crash is unusual, but let's just try DDU.

 

If you have/get more dump files, please post them. Just having one to work from is a bit light. 

[kiki4431]

37 minutes ago, Smalahovet said:

That error is rare, but can happen with cheap chinese displays using a cheap USB to HDMI adapter. 
This is the reason i got the USB-C version of the display for monitoring my temps. Cost 12 dollars at aliexpress but works. 

Or it can be firmware/driver issue. Like if it try hook up to the graphics at a kernel level. That will trip hypervisor checks. 

 

I'm not using an adapter though, it's HDMI to HDMI cable for signal, and USB-A to USB-C (the screen has USB-C connection) for power.

 

got mine from aliexpress too, and if you are refering to those Torzx, or whatever the name is, the ones that don't use AIDA64, but instead have their own software, and use only USB, I did consider those as well, the reason I went HDMI route, or better AIDA64 route, is becouse I wanted to custumize it down to the tiniest detail, and also I had no guarantees, that software will provide the options, and readouts (sensors) I wanted.

 

could be a driver/firmware issue as you said, but I don't know at this point how to fix that.

[kiki4431]

1 minute ago, Bjoolz said:

This one shows an error with the Nvidia driver. The Nvidia driver triggering a Hypervisor_Error crash is unusual, but let's just try DDU.

 

If you have/get more dump files, please post them. Just having one to work from is a bit light. 

will do that as well, actually just for the sake of removing one unlikelly possibillity, I had installed Malwarebytes, just to scan the system, found some stuff, not sure if that had anything to do with it, but I crossed that one, from the list.

 

for now it's just this one, should more BSOD happen, and if it does produce a minidump, I will post it here.

[kiki4431]

Well, it's been a few days without issues, until now.

 

So, I was playing a game, and had an issue where my main monitor (connected to RTX 5070ti), would just go dark, like it has lost input from GPU, actually this started happening yesterday, and I tried looking at event Viewer, but nothing at all was reported there.

This would happen exclusivelly during gameplay, and seeminglly only when some complex GPU-intensive thing would run. 

This would happen every minute or so, and "blackout" would last for about 5 seconds.

 

so just to make sure that DP cable didn't somehow loosen, I disconnected and reconnected it on the monitor's side, and that triggered BSOD with the same hypervisor error.

The PC froze again forcing me to manually shut it down, unfortunettly no minidump was created.

 

After that, I went into safe mode and DDU everything AMD and nVidia, downloaded new drivers a few days back, I just didn't yet do the reinstall, after reboot, I thought, maybe if I install the newest AMD driver first, and than nVidia, since I am pretty sure before nVidia driver was first installed, and than windows update secretlly pulled AMD one for iGPU, which I never used. But before the install finished I got the hypervisor BSOD again, PC rebooted on it's own, but I could no longer get into windows, just before log-in screen, I would get another hypervisor error again, now I know there is some conflict between these two GPUs, after two failed attempts to start the windows decided to "fix itself" unsuccessfully ofcourse. But it did allow me to enter safe mode through advanced recovery options.

 

DDU the AMD driver, and prevent windows from updating drivers automatically, did that, and than install just nVidia driver, for now things are good, but my plan to use iGPU didn't work, and without driver, it's no longer even detected, so now I have that sensor panel connected back to my RTX 5070ti GPU, I just hope it won't create another one of these BSODs, or interfere performance-wise.

 

I don't think these new minidumps are any different from the one I already posted, but it might reveal what happened, now that nVidia driver didn't exist when the error happened, which might give clue to what exactlly conflicted.

 

I will post these new ones, and if they hold any new informations, I would like to know, since I can't really see their contents, or don't know how.

 

BTW: I disabled PBO, it didn't help, since I got that BSOD the same evening, but it froze, and no minidump was created, so I didn't post anything about that, especially since the following 3 days had no issues.

Minidump.zip

Edited September 5, 2025 by kiki4431
Additional information

[Bjoolz]

1 hour ago, kiki4431 said:

Well, it's been a few days without issues, until now.

 

So, I was playing a game, and had an issue where my main monitor (connected to RTX 5070ti), would just go dark, like it has lost input from GPU, actually this started happening yesterday, and I tried looking at event Viewer, but nothing at all was reported there.

This would happen exclusivelly during gameplay, and seeminglly only when some complex GPU-intensive thing would run. 

This would happen every minute or so, and "blackout" would last for about 5 seconds.

 

so just to make sure that DP cable didn't somehow loosen, I disconnected and reconnected it on the monitor's side, and that triggered BSOD with the same hypervisor error.

The PC froze again forcing me to manually shut it down, unfortunettly no minidump was created.

 

After that, I went into safe mode and DDU everything AMD and nVidia, downloaded new drivers a few days back, I just didn't yet do the reinstall, after reboot, I thought, maybe if I install the newest AMD driver first, and than nVidia, since I am pretty sure before nVidia driver was first installed, and than windows update secretlly pulled AMD one for iGPU, which I never used. But before the install finished I got the hypervisor BSOD again, PC rebooted on it's own, but I could no longer get into windows, just before log-in screen, I would get another hypervisor error again, now I know there is some conflict between these two GPUs, after two failed attempts to start the windows decided to "fix itself" unsuccessfully ofcourse. But it did allow me to enter safe mode through advanced recovery options.

 

DDU the AMD driver, and prevent windows from updating drivers automatically, did that, and than install just nVidia driver, for now things are good, but my plan to use iGPU didn't work, and without driver, it's no longer even detected, so now I have that sensor panel connected back to my RTX 5070ti GPU, I just hope it won't create another one of these BSODs, or interfere performance-wise.

 

I don't think these new minidumps are any different from the one I already posted, but it might reveal what happened, now that nVidia driver didn't exist when the error happened, which might give clue to what exactlly conflicted.

 

I will post these new ones, and if they hold any new informations, I would like to know, since I can't really see their contents, or don't know how.

 

BTW: I disabled PBO, it didn't help, since I got that BSOD the same evening, but it froze, and no minidump was created, so I didn't post anything about that, especially since the following 3 days had no issues.

Minidump.zip 1.77 MB · 0 downloads

Both the two most recent crashes show an NMI being sent to the CPU. NMI (Non-Maskable Interrupt) is a type of interrupt where the CPU has to drop everything it's doing and handle it immediately. It skips the execution queue commands usually have to wait in. So it's reserved for more serious issues like hardware errors. Almost anything can send an NMI, but on consumer systems it's almost always the CPU itself.

 

That the first crash pointed to the Nvidia driver changes things here though. Like I said, having the Nvidia driver involved in a Hypervisor_Error crash is unusual so I'm not quite sure what to make of it. 

[Smalahovet]

On 8/31/2025 at 5:11 PM, kiki4431 said:

I'm not using an adapter though, it's HDMI to HDMI cable for signal, and USB-A to USB-C (the screen has USB-C connection) for power.

 

got mine from aliexpress too, and if you are refering to those Torzx, or whatever the name is, the ones that don't use AIDA64, but instead have their own software, and use only USB, I did consider those as well, the reason I went HDMI route, or better AIDA64 route, is becouse I wanted to custumize it down to the tiniest detail, and also I had no guarantees, that software will provide the options, and readouts (sensors) I wanted.

 

could be a driver/firmware issue as you said, but I don't know at this point how to fix that.

No, you are not using an adapter. But the device is internally. As they are mass produced for USB. So its cheaper to slap on an USB to HDMI converter than actually developing a HDMI version. Thats how it goes with cheap stuff from China often. 

[kiki4431]

15 hours ago, Bjoolz said:

Both the two most recent crashes show an NMI being sent to the CPU. NMI (Non-Maskable Interrupt) is a type of interrupt where the CPU has to drop everything it's doing and handle it immediately. It skips the execution queue commands usually have to wait in. So it's reserved for more serious issues like hardware errors. Almost anything can send an NMI, but on consumer systems it's almost always the CPU itself.

 

That the first crash pointed to the Nvidia driver changes things here though. Like I said, having the Nvidia driver involved in a Hypervisor_Error crash is unusual so I'm not quite sure what to make of it. 

Perhaps for whatever reason the iGPU (which is part of CPU) sent it.

 

just checked, the iGPU is detectable in task manager as well as device manager again, checked windows update, and well the damn thing was installing the AMD driver behind my back, regardless of me configuring it not to download device-specific software.

 

my sensor panel is still hooked to HDMI on RTX 5070ti, I will now see if the issues persist, but as far as I can tell iGPU might be causing problems, because yesterday the BSOD happened the moment I installed AMD driver, and had to DDU it afterwards, my guess the two GPUs are fighting each other and the CPU doesn't know how to make them cooperate, so the result is that error.

 

14 hours ago, Smalahovet said:

No, you are not using an adapter. But the device is internally. As they are mass produced for USB. So its cheaper to slap on an USB to HDMI converter than actually developing a HDMI version. Thats how it goes with cheap stuff from China often. 

I see, that could be the reason I am able to sent it 1080p signal, even though it's not advertised as 1080p, but windows detected it that way, and it works, the OSD menu on screen itself also reports that it's recieving 1080p signal, and nothing seems to be wrong with it specifically, wierd, but maybe there is some compression happening on the screen, or it's outright falsing the input and report it as 1080p capable, to both the PC and the user. I also configured AIDA64 to make a sensor panel 1920x1080, and no problems there either, bassically everything works regarding the screen, but something is happening elsewhere in my PC to cause these errors.

 

For now I will leave it plugged in RTX5070ti, (because these errors first started happening the moment I decided to use iGPU for it), and see what happens

[Bjoolz]

35 minutes ago, kiki4431 said:

Perhaps for whatever reason the iGPU (which is part of CPU) sent it.

 

just checked, the iGPU is detectable in task manager as well as device manager again, checked windows update, and well the damn thing was installing the AMD driver behind my back, regardless of me configuring it not to download device-specific software.

 

my sensor panel is still hooked to HDMI on RTX 5070ti, I will now see if the issues persist, but as far as I can tell iGPU might be causing problems, because yesterday the BSOD happened the moment I installed AMD driver, and had to DDU it afterwards, my guess the two GPUs are fighting each other and the CPU doesn't know how to make them cooperate, so the result is that error.

You could just disable the iGPU in the BIOS for testing if you are no longer using it. 

[kiki4431]

OK, it's been 10-11 days since last crash, everything, so far is working fine, I guess the issue is solved by connecting the screen to the main GPU, instead of using iGPU.

 

performance-wise, I haven't notticed too much of a difference, so I will just leave it that way, as for iGPU, well It's still enabled, but at this point, unless the BSOD happens again, I can disable it, but for now I will just leave it enabled, it's not doing anything anyway, I think.

 

Thank you all, for helping me sort this.