Question Out of ideas, please help. Computer randomly restarting and sometimes various BSOD. Stumped and frustrated.

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Brand new computer, about 6 months old. Last couple weeks i've had issues with my computer randomly restarting and sometimes getting BSOD, various errors.

I've done memtest twice, no errors both times. Just finished a FurMark test and no crashes. Which is why i'm very very stuck and I don't understand what else it could be.

I'm assuming a hardware failure of some sorts. Done some research since the FurMark test and it seems it could be PSU or the PSU isn't giving the right amount of power? But i don't understand why the FurMark passed at high stress.

I am just stumped and i'm getting very frustrated with things. I got a new computer to prevent this fun stuff.

PC - https://pcpartpicker.com/list/KmctrV

I wiped my computer a couple times, I thought maybe software issues, but the problems persisted. I've also switched ram slots, same issues. i did a scv test and it came back with corrupted file and fixed them, problem persisted. (Was recommended after the BSOD)

Edit - Forgot to mention I am getting random FPS drops in various games. Mainly playing PoE atm and I get FPS drops at both intense and seemingly nothing events. With this build it makes no sense I should be getting any severe FPS drops.
 

Colif

Win 11 Master
Moderator
Icue was the victim in both crashes, not the cause. According to Sysnative Icue appears to be having problems with your GPU drivers. In the past I would have blamed the new Nvidia drivers but since getting a new PC myself and using the new Nvidia drivers without conflicts, I am questioning myself. Also running Icue here and no BSOD so far.

I can't see anything in the dumps . Depending on what Logitech device you have, Logitech Gaming Software may have been replaced by G Hub. I doubt its the cause of errors though.

There are ways to force a BSOD but even the program written by Microsoft to do it, can also break WIndows... so I prefer to just wait.
 
Icue was the victim in both crashes, not the cause. According to Sysnative Icue appears to be having problems with your GPU drivers. In the past I would have blamed the new Nvidia drivers but since getting a new PC myself and using the new Nvidia drivers without conflicts, I am questioning myself. Also running Icue here and no BSOD so far.

I can't see anything in the dumps . Depending on what Logitech device you have, Logitech Gaming Software may have been replaced by G Hub. I doubt its the cause of errors though.

There are ways to force a BSOD but even the program written by Microsoft to do it, can also break WIndows... so I prefer to just wait.
So do you think I should just wait for another BSOD and see if it has information?

I also am noticing this "Failed with 0x57 retrieving AppModel Runtime status for package Windows."

When it happens i seem to have frame drops, and sometimes my game CTD. Not sure if related but yeah.


Yesterday i went through all the cables in the computer n made sure they were connected nicely. I reset CMOS.
I gave steam and my game a fresh install (in case it's really the game causing the issues, since that's when it happens most)

I ran Prime95 for an hour or so, no issues. I know it suggests multiple hours but the processor getting to 90c was sorta freaking me out, i know prime95 is meant to stress it but it was very close to the "limit 95" Ryzen Master has.

Think I am going to run a fresh install of my video drivers. Wipe n reinstall. I really hope one of these solutions randomly solves the issue. It's getting exhausting dealing with this lol.

Just wanna say I appreciate everyone's help immensely.
 

tsibiski

Reputable
Jun 23, 2019
95
21
4,565
Here is a thread about this same error (Failed with 0x57 retrieving AppModel Runtime status for package Windows.). It appears to be somewhat widespread with recent Windows updates.

One suggestion was about reinstalling certain Windows updates that were corrupted during install (page 5). The most recent post (page 6) as of now suggests that the issue might be related to a high refresh rate monitor (maybe compounded by other issues), although they imply that changing the refresh rate stopped their issues.
 
Last edited:
Here is a thread about this same error (Failed with 0x57 retrieving AppModel Runtime status for package Windows.). It appears to be somewhat widespread with recent Windows updates.

One suggestion was about reinstalling certain Windows updates that were corrupted during install (page 5). The most recent post (page 6) as of now suggests that the issue might be related to a high refresh rate monitor (maybe compounded by other issues), although they imply that changing the refresh rate stopped their issues.
Ended up solving this one so far it looks like. Seems it's a windows issue so I did Windows Insider and got the beta version and then a had more crashes seemingly set to a setting in game.

Still experience crashes after clean graphics install. Not really sure where to go from here until i get maybe a BSOD. =\
 
So you are on Windows Insider now? The beta version? On the cutting edge of windows. Drivers are even less guaranteed of working there.
I just went on it last night after crashing multiple times. Last kinda effort. I literally don't know what to do now. Should i send it into a computer shop? Return my video card for another? I really don't wanna just keep living with this cause i'm sure it's not good in the long run for the computer lol.

I might try another wipe / fresh install and only install my game n see where that leads me. But i basically already did that a week or so ago. I did download chrome and discord and multiple games though.


Is everyone else out of ideas? I'll do whatever i need to to have a normal ass computer again. Preferably not buying an entirely new one? lol.
 
Last edited:
Well I thought wiping my computer again and fresh installing and virtually downloading nothing else was working quite well, lasted 8ish hours. Just got another crash. Again 2 WHEA-Logger Errors. APIC ID:0 and APIC ID:4

I stumbled upon this View: https://www.reddit.com/r/buildapc/comments/cgdbkw/pc_restarts_randomly_always_the_same_two_errors/eugd6h9/


This is exactly whats happening to me as well. This guy said it ended up being a MOBO and another person with also the same issues said it was his processor.

Also a few others say it's likely the PSU. Not sure where I should start about replacing, or if there's a way to pinpoint which one it actually is. Still not getting BSOD though.
 

tsibiski

Reputable
Jun 23, 2019
95
21
4,565
Well I thought wiping my computer again and fresh installing and virtually downloading nothing else was working quite well, lasted 8ish hours. Just got another crash. Again 2 WHEA-Logger Errors. APIC ID:0 and APIC ID:4

I stumbled upon this View: https://www.reddit.com/r/buildapc/comments/cgdbkw/pc_restarts_randomly_always_the_same_two_errors/eugd6h9/


This is exactly whats happening to me as well. This guy said it ended up being a MOBO and another person with also the same issues said it was his processor.

Also a few others say it's likely the PSU. Not sure where I should start about replacing, or if there's a way to pinpoint which one it actually is. Still not getting BSOD though.

Pinpointing the part effected would require you to do what I mentioned earlier in this thread. You will need to take parts out, one by one. First, remove one of the sticks of RAM, and use only one at a time in the A2 channel. Use that for a day or two, or until you hit the next BSOD. Then insert the other stick of RAM, and have that be the only RAM, also in A2.

Next borrow a graphics card from a friend, or buy a really cheap older graphics card on ebay, and use that for a day or so, until it BSOD's.

Next, try to borrow or buy a backup PSU and use that until the next BSOD.

Finally, try this with another CPU. You could get a really cheap used Ryzen 2700 or whatever on ebay.

If it is still BSOD-ing after all of this, it is probable that your motherboard is FUBAR.

(Each time you move to the next part, add the other one(s) back in)

This is the only way for you to find out which part is problematic given the information on what has been happening. You could also pay a computer repair shop for this... but I doubt that would be a reasonable price because they'd have to be using the machine a ton to reproduce the issue, and you'd be paying by the hour.

So the way I see it, you can:
A) Do what I suggested
B) Buy a new computer, and sell that one on the cheap to someone who has the patience/time to diagnose the issue part by part
 
Pinpointing the part effected would require you to do what I mentioned earlier in this thread. You will need to take parts out, one by one. First, remove one of the sticks of RAM, and use only one at a time in the A2 channel. Use that for a day or two, or until you hit the next BSOD. Then insert the other stick of RAM, and have that be the only RAM, also in A2.

Next borrow a graphics card from a friend, or buy a really cheap older graphics card on ebay, and use that for a day or so, until it BSOD's.

Next, try to borrow or buy a backup PSU and use that until the next BSOD.

Finally, try this with another CPU. You could get a really cheap used Ryzen 2700 or whatever on ebay.

If it is still BSOD-ing after all of this, it is probable that your motherboard is FUBAR.

(Each time you move to the next part, add the other one(s) back in)

This is the only way for you to find out which part is problematic given the information on what has been happening. You could also pay a computer repair shop for this... but I doubt that would be a reasonable price because they'd have to be using the machine a ton to reproduce the issue, and you'd be paying by the hour.

So the way I see it, you can:
A) Do what I suggested
B) Buy a new computer, and sell that one on the cheap to someone who has the patience/time to diagnose the issue part by part
Okay thanks so much. I do have my previous PC still, do i have to worry about compatibility with things? Like can i use my old video card / psu.

Also, since I am not getting BSOD even currently, and haven't in 3 days now. Should i just behave, within your steps, of a random computer reboot ( what its been doing constantly) as a BSOD error. As in "oh guess it's not the video card, it restarted randomly again".
 

tsibiski

Reputable
Jun 23, 2019
95
21
4,565
Yeah. That's a fair approach. Consider the restart to be the same as a BSOD in the context of my debugging steps.

RAM is definitely the place to start. When I overclocked my RAM, it sometimes was stable enough to boot, but the entire PC would shut off and restart in the middle of a benchmark. A sudden restart, in my own limited experience, is either A) RAM issue, or B) PSU issue where the power draw exceeded the PSU's capabilities during high load. However, your PSU should be provided plenty of power, so if it was a PSU issue, it would be due to some damage.

Again, I want to stress that I am very far from an expert, but to my knowledge, PSU's don't often "fail just a little bit". When they fail, it is extremely noticeable. If the PSU works at all after damage, it will only supply enough power to light up LEDs on the board, and maybe spin some fans - but not likely boot. If that is indeed true, it seems like it's probably not the culprit here. So I'd definitely start with the RAM test.

Compatibility is definitely important. Your B450 should be able to handle any Ryzen 1 or 2 gen CPU. Your old graphics card, unless it is insanely old, like over a decade old, is probably going to be fine being plopped into most PC's on most any motherboard. Your old PSU, as long as it provides enough watts, will be fine.
 
Yeah. That's a fair approach. Consider the restart to be the same as a BSOD in the context of my debugging steps.

RAM is definitely the place to start. When I overclocked my RAM, it sometimes was stable enough to boot, but the entire PC would shut off and restart in the middle of a benchmark. A sudden restart, in my own limited experience, is either A) RAM issue, or B) PSU issue where the power draw exceeded the PSU's capabilities during high load. However, your PSU should be provided plenty of power, so if it was a PSU issue, it would be due to some damage.

Again, I want to stress that I am very far from an expert, but to my knowledge, PSU's don't often "fail just a little bit". When they fail, it is extremely noticeable. If the PSU works at all after damage, it will only supply enough power to light up LEDs on the board, and maybe spin some fans - but not likely boot. If that is indeed true, it seems like it's probably not the culprit here. So I'd definitely start with the RAM test.

Compatibility is definitely important. Your B450 should be able to handle any Ryzen 1 or 2 gen CPU. Your old graphics card, unless it is insanely old, like over a decade old, is probably going to be fine being plopped into most PC's on most any motherboard. Your old PSU, as long as it provides enough watts, will be fine.
Thanks so much. My other PC is about 8 years old i think. So hopefully it's fine. i know it's a radeon graphics card of some sorts and an intel processor. totally unsure of the rest of it though.
 
So, so far i've been rocking with my other video card (the older one from the other computer) and i've been a few hours now without a crash. That isn't to say it's the issue yet, as i've learned not to get my hopes up here. But can the GPU be causing the issues that I am having? I never saw any solutions to similar problems be because of the video card.
 

Colif

Win 11 Master
Moderator
WHEA errors can be anything. So it could be the GPU, I could guess the RTX 2060 won't work in old PC so maybe you need to get outside help to figure out if it is cause or not, A repair store can put it in a working PC to isolate it.

CPU & motherboard almost impossible to test. I mean, AMD CPU you have Prime 95 and thats it. Motherboards have even less tests. If you suspect motherboard you have to test everything else and make sure it works as a process of elimination. If all else works, it has to be MB. Not very scientific. Or you give PC to a repair store and ask them to work it out.

The dump info you gave us isn't enough for me to say if its gpu or not. Now i remember, the Icue was having issues with gpu driver functions so maybe.
 
WHEA errors can be anything. So it could be the GPU, I could guess the RTX 2060 won't work in old PC so maybe you need to get outside help to figure out if it is cause or not, A repair store can put it in a working PC to isolate it.

CPU & motherboard almost impossible to test. I mean, AMD CPU you have Prime 95 and thats it. Motherboards have even less tests. If you suspect motherboard you have to test everything else and make sure it works as a process of elimination. If all else works, it has to be MB. Not very scientific. Or you give PC to a repair store and ask them to work it out.

The dump info you gave us isn't enough for me to say if its gpu or not. Now i remember, the Icue was having issues with gpu driver functions so maybe.
Well as i was reading this I got a blue screen of death. Although it isn't an error code i've gotten before.

https://we.tl/t-DPVIXV6ejp

Also just seen in event viewer it made a memory.dmp file. It is 1.2 gb though so not sure if I should share it / its worthwhile. I am uploading it but it'll take a bit.
 
Last edited:

Colif

Win 11 Master
Moderator
Gardenman should convert that soon if he is around.

Event viewer isn't an ideal way to track errors, often you get errors in there that only happen once and the next time the action happens, no errors. A perfectly normal PC will still have errors in event viewer. I try to ignore it and reliability monitor, as they can make you worry about nothing.

Obviously there is something now but I was just saying.
 
Gardenman should convert that soon if he is around.

Event viewer isn't an ideal way to track errors, often you get errors in there that only happen once and the next time the action happens, no errors. A perfectly normal PC will still have errors in event viewer. I try to ignore it and reliability monitor, as they can make you worry about nothing.

Obviously there is something now but I was just saying.
Yeah I've kinda gotten that feeling about event viewer from other people. Do you think uploading / posting the 1.2 gb file is worth a check?
 
is that your memory.dmp file as Gardenmans internet would die thinking about downloading that. Minidumps are way smaller and contain most of the info we need.

if its something from event viewer, probably not worth it.
Okay awesome. Yeah it says "A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 56c078c3-c0a3-4446-a463-7c9863707ad5."


I did the minidump thing and got the file that i just linked previously, after the dump. As i wiped my computer I didn't have that option again. It seemed to still keep record of the minidump though so I hope it still works / is okay?

Also I have a feeling the new dump might be because of the video card I have in it atm, since i am in the process of elimination. I don't know anything about that stuff though but I hope it has something to say about wtf is going on =\.
 

gardenman

Splendid
Moderator
I ran the dump file through the debugger and got the following information: https://jsfiddle.net/4oet320q/show This link is for anyone wanting to help. You do not have to view it. It is safe to "run the fiddle" as the page asks.

File information:092920-4468-01.dmp (Sep 30 2020 - 00:15:40)
Bugcheck:PAGE_FAULT_IN_NONPAGED_AREA (50)
Driver warnings:*** WARNING: Unable to verify timestamp for ntoskrnl.exe
Probably caused by:memory_corruption (Process: ?)
Uptime:0 Day(s), 6 Hour(s), 25 Min(s), and 54 Sec(s)

Comment: I don't think I've ever seen so many symbol errors in a single minidump. I edited most of them out.

This information can be used by others to help you. Someone else will post with more information. Please wait for additional answers. Good luck.
 
I ran the dump file through the debugger and got the following information: https://jsfiddle.net/4oet320q/show This link is for anyone wanting to help. You do not have to view it. It is safe to "run the fiddle" as the page asks.

File information:092920-4468-01.dmp (Sep 30 2020 - 00:15:40)
Bugcheck:PAGE_FAULT_IN_NONPAGED_AREA (50)
Driver warnings:*** WARNING: Unable to verify timestamp for ntoskrnl.exe
Probably caused by:memory_corruption (Process: ?)
Uptime:0 Day(s), 6 Hour(s), 25 Min(s), and 54 Sec(s)

Comment: I don't think I've ever seen so many symbol errors in a single minidump. I edited most of them out.

This information can be used by others to help you. Someone else will post with more information. Please wait for additional answers. Good luck.
Thank you so much Gardenman. Hopefully somebody can take a look at this soon n give me a "low down" :)

All i see atm is ram failure but i'm dumb when it comes to in depth comp stuff. I just know basics =\
 

Colif

Win 11 Master
Moderator
*** WARNING: Unable to verify timestamp for ntoskrnl.exe
this is the windows kernel. If this was corrupt windows wouldn't be running right now.

but if you look at the big picture...

fffff80576cf418e-fffff80576cf418f 2 bytes - dxgmms1!VidSchiCheckPreemptionPolicy+11e
[ 48 ff:4c 8b ]
fffff80576cf4195-fffff80576cf4198 4 bytes - dxgmms1!VidSchiCheckPreemptionPolicy+125 (+0x07)
[ 0f 1f 44 00:e8 26 b7 59 ]
fffff80576d04d73-fffff80576d04d78 6 bytes - dxgmms1!DXGGLOBAL::GetGlobal+f (+0x10bde)
[ ff 15 3f 98 01 00:e8 28 e5 05 00 90 ]
fffff80576d04da6-fffff80576d04da7 2 bytes - dxgmms1!VIDMM_PROCESS_ADAPTER_INFO::ReleaseProcessAdapterInfoLock+a (+0x33)
[ 48 ff:4c 8b ]
fffff80576d04dad-fffff80576d04db0 4 bytes - dxgmms1!VIDMM_PROCESS_ADAPTER_INFO::ReleaseProcessAdapterInfoLock+11 (+0x07)
[ 0f 1f 44 00:e8 ae 61 5d ]
fffff80576d04db2-fffff80576d04db3 2 bytes - dxgmms1!VIDMM_PROCESS_ADAPTER_INFO::ReleaseProcessAdapterInfoLock+16 (+0x05)
[ 48 ff:4c 8b ]
fffff80576d04db9-fffff80576d04dbc 4 bytes - dxgmms1!VIDMM_PROCESS_ADAPTER_INFO::ReleaseProcessAdapterInfoLock+1d (+0x07)
[ 0f 1f 44 00:e8 32 6e 5d ]
fffff80576d054de-fffff80576d054e3 6 bytes - dxgmms1!ADAPTER_DISPLAY::DdiSetVidPnSourceAddress+12 (+0x725)
[ ff 15 d4 90 01 00:e8 bd dd 05 00 90 ]
fffff80576d055be-fffff80576d055bf 2 bytes - dxgmms1!VidSchiReportHwHang+7a (+0xe0)
[ 48 ff:4c 8b ]
fffff80576d07416-fffff80576d07417 2 bytes - dxgmms1!DXGPUSHLOCK::AcquireShared+56 (+0x1e58)
[ 48 ff:4c 8b ]
fffff80576d0741d-fffff80576d07420 4 bytes - dxgmms1!DXGPUSHLOCK::AcquireShared+5d (+0x07)
(this is just a short part of the dump text) you notice a pattern.
dxgmms1 = Direct X. 99% of the time I see this in a dump the cause is GPU drivers.

This happened after you swapped cards right? as the AMD drivers are "new" in that they weren't there before

Did you run DDU and remove the Nvidia drivers before installing AMD? I can see parts of Nvidia drivers on here but not the Display driver, which was confusing.

All the stack text spells a picture of PC not talking to GPU right.

Nov 04 2015atikmdag.sysATI Radeon Kernel Mode driver
Nov 04 2015atikmpag.sysATI video card driver
typical AMD gpu drivers

Are there any newer?
Mar 14 2019nvvad64v.sysNvidia Virtual Audio driver http://www.nvidia.com/

Nov 29 2019NvModuleTracker.sysNVIDIA Module Tracker driver
Jan 10 2020nvvhci.sysNvidia Virtual USB Host Controller driver http://www.nvidia.com/
isolated Nvidia drivers. No actual display driver.

I am just a guy trying to work these things out as he goes, but dxgmms1!VidSchiReportHwHang+7a could mean Video Scheduler Report Hardware Hang
 
Last edited:
this is the windows kernel. If this was corrupt windows wouldn't be running right now.

but if you look at the big picture...

(this is just a short part of the dump text) you notice a pattern.
dxgmms1 = Direct X. 99% of the time I see this in a dump the cause is GPU drivers.

This happened after you swapped cards right? as the AMD drivers are "new" in that they weren't there before

Did you run DDU and remove the Nvidia drivers before installing AMD? I can see parts of Nvidia drivers on here but not the Display driver, which was confusing.

All the stack text spells a picture of PC not talking to GPU right.

Nov 04 2015atikmdag.sysATI Radeon Kernel Mode driver
Nov 04 2015atikmpag.sysATI video card driver
typical AMD gpu drivers

Are there any newer?
Mar 14 2019nvvad64v.sysNvidia Virtual Audio driver http://www.nvidia.com/

Nov 29 2019NvModuleTracker.sysNVIDIA Module Tracker driver
Jan 10 2020nvvhci.sysNvidia Virtual USB Host Controller driver http://www.nvidia.com/
isolated Nvidia drivers. No actual display driver.

I am just a guy trying to work these things out as he goes, but dxgmms1!VidSchiReportHwHang+7a could mean Video Scheduler Report Hardware Hang
This happened after you swapped cards right? as the AMD drivers are "new" in that they weren't there before

Did you run DDU and remove the Nvidia drivers before installing AMD? I can see parts of Nvidia drivers on here but not the Display driver, which was confusing.


I did not run DDU or remove the nvidia drivers before installing my old video card, and yes it happened after i swapped the cards. Kinda why i figured it was about the the older card being in there.


I was using geforce experience to update my drivers for my main 2060 super card.

Would you suggest running a DDU and then trying to install all the latest nvidia drivers (the ones you named, idk if there's more though) and seeing if I get a crash again?

What should be my next step, I guess is what i'm trying to figure out. I was running process of elimination and I had my old video card in, my computer still did the restart thing and crashed like normal. Not the BSOD but the normal random instant restart. So I have my 2060 back in atm.
 

Colif

Win 11 Master
Moderator
When you put the AMD card into PC you should have removed the Nvidia drivers first.
I don't want you to update those 3 Nvidia drivers I listed, I only listed those as they seemed to be left overs from you removing the drivers (I assumed at time you had removed them).

That error does look like it was caused by the GPU drivers, it was much clearer in that way compared to the previous 2. I assume that card worked fine in last PC, it might be the age of the drivers.

Swapping card out didn't prove it was GPU, as you just got different errors.

Run ddu, remove any AMD or Nvidia drivers and start with the latest drivers form Nvidia.
 
When you put the AMD card into PC you should have removed the Nvidia drivers first.
I don't want you to update those 3 Nvidia drivers I listed, I only listed those as they seemed to be left overs from you removing the drivers (I assumed at time you had removed them).

That error does look like it was caused by the GPU drivers, it was much clearer in that way compared to the previous 2. I assume that card worked fine in last PC, it might be the age of the drivers.

Swapping card out didn't prove it was GPU, as you just got different errors.

Run ddu, remove any AMD or Nvidia drivers and start with the latest drivers form Nvidia.
So i did this 2 days ago, and I was feeling pretty good about it as i didn't have a restart for a full day and a bit. Just got another unfortunately so I guess that's not the fix =\. Seems i'm back to power supply / motherboard / processor again I think?

Thoughts on what my next step would be?