"New" PC crashes: No video signal,no Bluescreen, no Event Viewer entry

Alternate title: "Why am I not even getting Bluescreens? As in, never?"


Hi,

not 100% sure which sub-forum is the right one for this... Well, I'm trying to narrow down / exclude a hardware issue, so I figure I'll put this here, correct me if I'm wrong.

I've got a problem with a ... well, okay, the PC isn't actually new anymore - it's been in use for a bit over 2 years now - technically 3 - but the problem has been popping up every now and again since pretty much the beginning.

It's a custom build PC (Windows 10 Home 64Bit OEM) I ordered online, which was the first time I've done that. (For a list of components, see the end of the post.)

The issue happens fairly rarely and randomly, so it isn't currently a serious issue ... yet - but it's the type of problem I'd expect to crop up in a 7+ year old system, not a fairly new / ~2-years-of-use one, and it doesn't bode well for the future.

Warranty's running out soon, so I'd like to figure this out before then... preferably without having to send it in, as that would be... problematic, for a number of reasons.

^{(One of which is that I don't trust the postal service not to damage the PC ... any further. ...When delivering, they dropped it off right at my door step. As in: literally dropped the package, half a meter to the stone floor. Right at my door step. ...Styrofoam packaging and "inflated foam cushion" inside the PC and all, but still. Aargh! -_-# )}

This is a 1200 Euro PC, totally overpowered for my usual applications, bought with upgradability in mind, so I was expecting it to last a good long while.



Already in contact with One.de's support, but I'd like a second opinion in case anyone has any other insights to contribute.


Let's say it this way:

If this problem stays exactly the way it is now, and never gets actually worse - I can live with it.

If it's somehow a software or driver issue, it probably isn't gonna get any worse, and the PC won't need to be sent off for that. If some components in the custom PC aren't playing nice with each other, then maybe that won't get worse either? If it's indicative of a hardware fault however, then it'll probably get worse eventually.

If I can find out which component is broken, I may remove it and send them that by itself, minimizing risk in delivery. (How do I know the postal service won't do more damage than the PC currently has?! )






________________________________________________________________


PRIMARY ISSUE DESCRIPTION:

My PC will occasionally crash in the following manner:

• PC suddenly freezes up
• After a few seconds the screen goes dark, followed by a "no video signal" message from my monitor
• If any audio is playing, the current tone will get "stuck" for a while, then go quiet (more on this later).
• The PC itself is still powered on. Lights are on, fan's still running.
• "Read/Write" indicator light is no longer wildly flickering as usual, but either:
  a) blinking in a regular pattern that presumably means "idle" ("blip___________blip__________blip..."),
  or b) completely off.
  
  
• No Bluescreen. No Event Viewer entry. No "Windows wasn't shut down properly" message after reboot. ...Like Windows didn't even realize it crashed.
  (There was ONE exception to this, which left an Event Viewer entry and even a Memory Dump, see further below.)
  

Here's a video of the PC crashing:



(In the video, I tested if an external peripheral could be the cause, so everything except monitor and keyboard was removed at the time, including WLAN-stick.)

I checked after that crash, and literally NOTHING was added to the Event Viewer during the period the PC was on before crashing - or at least not in "Administrative View" nor "Windows Logs -> System", which I presume to be where I should be looking.





Often, there are minor warning signs prior to a crash, in the form of a brief audio stutter, i.e. music going 'brrt' for a split second. Doesn't always mean it's about to crash though, it's more of a "hanging at the edge of a cliff" warning. If I'm lucky, it won't crash.

Recreated audio example here:






________________________________________________________________



When does the problem occur:

• It happens fairly rarely and randomly. Couple times per month maybe. Sometimes no crashes for a few weeks, then more frequently for a few days, then no problems for a while. As stated before, not currently a real issue, but does not bode well for the future. If this is a hardware problem, I need to send it into repair before warranty runs out, which would be... problematic.
  
  
• Crashes tend to happen either:
  
  
  • Most frequently: Shortly after startup, in the first 2 minutes, even if I'm not doing anything. Already had crashes during Login, too. (Which is impressive considering I don't have a password set up so it just shows my name for a few seconds and then goes to Desktop... and yet sometimes it still managed to crash here.) Chances increase if I try doing, err, anything during those 2 first minutes.
    
    
  • Less frequently: During relatively heavy load, I ... suppose? I remember it happening more frequently when playing video games, which I admittedly don't do often on PC, and once during video rendering.
    
    
  • Occasionally: During normal workload at some random point throughout the day.
    
    
    So it's not exactly dependent on workload, or any input on my end.
    
    
  • Oddity: Sudden workload increase after leaving the PC (relatively) idle for a longer period of time also caused a crash or two. May be a reproducible test case.

________________________________________________________________




The one single time an Event Viewer entry was made:

Every time the PC crashes, I always check the Read/Write indicator light for possible activity. There's either none at all, or a regular pattern presumably meaning "idle".

There was a single exception to the rule. Still no bluescreen when this happened, just "no video signal" again, but the Read/Write indicator light kept rapidly flashing for a few more seconds before stopping.

And whatdoyaknow, there's an Event Viewer entry this time:

- The previous system shutdown at 19:00:55 on 03/12/2022 was unexpected.

- Dump file creation failed due to error during dump creation. [Hmm. It did create a memory dump file though. Hope that isn't corrupted.]

- The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

- (A pointer device has no information about the monitor it is attached to.) [Unrelated]

- The computer has rebooted from a bugcheck. The bugcheck was: 0x0000007e (0xffffffffc0000005, 0xfffff801640fd5c3, 0xfffff5806f3a1768, 0xfffff5806f3a0fb0). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: c96f7f1c-38e6-44d2-bf4e-bbcce13164bf.

- (The driver detected an internal driver error on \Device\VBoxNetLwf.) [Unrelated]

XML from bugcheck:

Code:

<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
 <System>
  <Provider Name="Microsoft-Windows-WER-SystemErrorReporting" Guid="{ABCE23E7-DE45-4366-8631-84FA6C525952}" EventSourceName="BugCheck" /> 
  <EventID Qualifiers="16384">1001</EventID> 
  <Version>0</Version> 
  <Level>2</Level> 
  <Task>0</Task> 
  <Opcode>0</Opcode> 
  <Keywords>0x80000000000000</Keywords> 
  <TimeCreated SystemTime="2022-12-03T18:05:54.074394500Z" /> 
  <EventRecordID>333304</EventRecordID> 
  <Correlation /> 
  <Execution ProcessID="0" ThreadID="0" /> 
  <Channel>System</Channel> 
  <Computer>Toaster10</Computer> 
  <Security /> 
 </System>
<EventData>
  <Data Name="param1">0x0000007e (0xffffffffc0000005, 0xfffff801640fd5c3, 0xfffff5806f3a1768, 0xfffff5806f3a0fb0)</Data> 
  <Data Name="param2">C:\Windows\MEMORY.DMP</Data> 
  <Data Name="param3">c96f7f1c-38e6-44d2-bf4e-bbcce13164bf</Data> 
  </EventData>
</Event>

Memory Dump (both minidump and main dump): MEMORY DUMPS [03.12.22, 19;05 Uhr]

Big memory dump is 1.8 GB, compressed to 230 MB. Hope that's not just filled with NULL values again like some of the screenshots I took seconds before a PC crash the other day.





This was the single bugcheck in the event viewer's entire history log:



I hope that somehow helps.




________________________________________________________________




What I have already tested:

• Standby and Windows Fast Startup were already disabled.
• Testing the PC with all peripherals removed (except monitor and keyboard) -> ironically crashed while recording that, as seen in above video.
• RAM Test (Test Mix = Extended, Cache = ON, Pass count = 3) -> "The Windows Memory Diagnostic tested the computer's memory and detected no errors"
• Harddrive / M2.SSD check: CrystalDiskInfo and HDTune judge both to be in good health. Error Scan with HDTune found no damaged blocks.

What I'm still gonna test:

• Remove and then put back in the RAM sticks and the graphics card in case something is lose.
• (Will edit this section if I hear back from One.de support for what else I should try.)

Reproducible Test Case:

As I said, the problem happens randomly and fairly rarely, and most of the time within the first 2 minutes after booting up.

It's taken a lot of experimenting, and a lot of other, more resource intensive "stress tests" quite ironically did not cause the system to crash---

...but I think I may (tentatively) have a test case to, err, deliberately crash the PC. You know, so I can see if anything I do fixes that.

Test case setup: Put Chrome (500 tabs), Waterfox (9000+ tabs), and a poorly optimized 3D game ("Event[0]") on autostart. (Plus the Epic Games store app which is already on autostart.)

^{(Yes, 9000 tabs sounds like a lot, and I guess at some point I'm gonna have to do some spring cleaning again, but this is normal load for me. Waterfox is better optimized than regular Firefox, and it makes little difference if you open 200, 2000, or 10.000 tabs. And my PC has 32 GB of RAM, this is nothing. In any case, I don't normally have this on autostart so it's hardly a culprit. ;-)}

Observed behavior: When turning the PC on in the morning, trying to open all of these on autostart predictably crashes the PC shortly after start up.

However - and I don't have enough test data for this yet - there seems to be a difference, depending on how long the PC was turned off prior to this stress test.

• Restart PC = No crash with same setup (1 or 2x tested) (eh, one of those two was a bit unclear)
• PC left off for an hour = No crash with same setup (1x tested)
• PC left off overnight = Crash (3x tested)

Again, not very much data (some unaccounted factor may have skewed results), but could this be a motherboard issue? A broken capacitor, maybe?? I sure hope not.

I mean - why am I not even getting any bluescreens? As in, never?

I vaguely remember getting a bluescreen on this PC once or twice at some point, possibly for unrelated reasons, but that was so long ago I can't recall what year it was.

Is the motherboard intercepting errors before Windows gets a chance to handle them?? Are bluescreens deactivated somehow???


Or is this behavior something that commonly happens with the "MSI B450-A Pro Max (Chipset: B450 / ATX)" motherboard?



________________________________________________________________







Rare oddities / potentially related issues:
(Stuff that may or may not provide a hint, or may simply be a fluke)

• This one's a dozy; I haven't observed this in 1.5 years, and MUCH more rarely then the primary issue even then, but it's very possible it did happen again and I simply missed it - a fake crash.
  
  "No video signal" as described above, happening during medium to heavy load, and Windows seemed to have crashed while I was away from my PC. Standby is disabled, by the way. Wiggle the mouse, mash the keyboard, nothing. Turn monitor off/on -> no video signal.
  
  ...but then Windows came back from its "no video signal" state after ~20 minutes, then taking another ~15 or so minutes to stop acting like a snail and become properly responsive again.
  
  Managed to look at the task manager during this once or twice. CPU/GPU/Memory was fine. Only thing of note was: Performance -> disk -> D-drive showed it as being in ~99% use. (C is an M2 SSD with the OS installed, D is a regular 4 TB hard drive. D is disk 0, C is disk 1.)
  
  Not sure what was taxing the D-drive, though.
  
  And yes, I did "wait it out" during a regular "crash on boot up" for half an hour once or twice to check if this was the same issue. But alas, Windows didn't return that time. Who knows if I just wasn't patient enough, or didn't test it often enough.
  
  
• LESS RELEVANT: This next bit happened only once, not sure if it's relevant.
  
  Remember when I said it sometimes crashed when playing video games?
  
  On one singular peculiar occasion, it crashed within less than a minute after closing a particularly resource intensive video game ...at least according to how loud my fan was spinning.
  
  And it crashed only after the heavy load was gone. Weird. Only had once instance of this.
  
  So take it or leave it as a clue or a fluke.

________________________________________________________________

PC and Operating system information:

OS:
Windows 10 Home 64Bit OEM
Version 1909, OS Build 18363.1139

Processor:
AMD Ryzen 7 2700X 8x 3.70 GHz 105W

RAM:
32 GB DDR4 Crucial 2666 MHz (2x 16 GB)

Grafics card:
8 GB AMD Radeon RX 590 Sapphire Pulse

Mainboard:
MSI B450-A Pro Max (Chipset: B450 / ATX)

M.2 SSD (with OS installed):
500 GB M.2 2280 PCIe x4 Samsung 970 EVO Plus NVMe SSM

Hard drive:
4 TB SATA III WD Blue WD40EZRZ

Power adapter:
600 Watt be quiet! PURE POWER 11 80+ Gold

Processor cooling:
be quiet! Pure Rock

DVD drive:
1x DVD-RW / DVD-ROM ASUS DRW-24D5MT black

PC case:
Fractal Design Define R6 Black (ATX)

________________________________________________________________

Long and windy explanation......absolutely uniqe!

Repair Install Windows 10 with an In-place Upgrade

Short answer: I would have tried this......I also was struggling a couple of years with a PC that worked 90-95%.
It was fixed during a small hour.


And a german ISO you also could find here. Thrustfully.
Deskmodder.de - News zum Thema Microsoft, Windows 11 / 10, sowie anderen Neuigkeiten aus dem Netz, viele Tipps und individuelle Hilfe bei Problemen in unserem Forum ● Deskmodder - seit 2003

While it is understandable you would like to try sort your PC yourself to avoid the hassle involved with sending it in, you definitely do not want to do anything that may void/impair your warranty - which is why it may be best to inform the provider of the PC of the problems you are facing with it (as thoroughly as you have done here), before making any changes. Once you have opened up a dialogue with the provider you can then let them know of what steps (suggestions provided here) you wish to take and get the go-ahead from the provider before making them so as to avoid voiding the warranty. If you are unable to get it sorted, then you will still have the option to send it in under warranty. Some providers may be able to supply the box to send your computer back in, else, some couriers will wrap the PC for you and take responsibility for damages - the provider should be able to inform you on such.

Hope you can get it fixed,

Best regards
Fox Computers

Thanks for your quick replies!

I'm a bit in a rush atm, so sorry if I miss any detail from the posts already made, I'm gonna go over them again tomorrow.

NTN said:

Long and windy explanation......absolutely uniqe!

I confess, I feel like a bystander in a Sherlock Holmes novel - I have no idea what's a clue and what's not, or what any of it means, I'm only guessing. So I figured I'd provide all the information I can in case any of it proves relevant to "solve the case"!

Fox Computers said:

which is why it may be best to inform the provider of the PC of the problems you are facing with it (as thoroughly as you have done here), before making any changes

Which I already did!

Most of the tests (see "What I have already tested") and "remove X and put it back in" from "What I'm still gonna test" I mentioned were direct suggestions from the One.de support, except for the stress tests and autostart tests I came up with myself, which I needed as a "control" to see if any of the things I did fixed the issue or not, since crashes are random and sporadic. Wouldn't want to think "Oh that fixed it", let the warranty expire, and suddenly "oops it crashed again".

I sent them a mail containing (almost) all of this information yesterday ... minus some more recent results from the autostart tests I did earlier today implying different crash-likelihoods depending on if the PC was turned off for a longer period of time or not, which makes me worried about a possible capacitor issue. But admittedly I only have very little test data there, so who knows, hopefully it isn't that.

NTN said:

Repair Install Windows 10 with an In-place Upgrade

Short answer: I would have tried this.

So, you think it's a software issue rather than a hardware problem?

'Cause the One.de support's answer to my last mail containing roughly the information described above can be summed up as "reinstall windows to fix your issue".

Well, I cannot really go and just bulldoze my operating system - for a number of reasons. If it is a software issue, well then, "problem solved", I don't really need to fix it, it means the problem's not gonna get any worse than it currently is, and it's only mildly annoying at the moment. My concern was that it might be the onset of a serious hardware problem cutting the life time of my PC short, and the warranty period is running out to have that fixed.

itsme1 said:

Make a small partition, install windows on it with the latest updates. If you have no more problems, install the drivers for the computer.

This idea I like. That's a great way of testing if it is a software problem or not!

How small of a partition are we talking here?

I'm gonna have to look into how I can do that, obviously without deleting anything currently on my hard drive.

Then I guess I can set boot order in the BIOS, or something?

A clean install on another partition won't help unless you're only using this one for 4-5 months.

Huh? Wait, you mean I should use the new partition for 4 - 5 months to see if it crashes?

Well, I mean, if it is a hardware defect the crashes would be reproducible pretty much immediately, wouldn't they? Like I said above in "Reproducible Test Case", with some experimenting I found a test-case to cause the PC to deliberately crash, just don't have much test data on this yet. It isn't even anything too taxing on the hardware either, just "do this on autostart" instead of "wait 5 minutes to do it".

Again, if I can definitely rule out a hardware defect, then this problem is "solved", since it'd mean it isn't gonna get any worse than it currently is (and right now it's acceptable), and no hardware needs to be replaced / fixed by a repair shop within the limited warranty period. :)

You reduce a partition on the main hdd/ssd or secondary hdd/ssd from the disk manager, from 30 to 50 go it should be enough. At startup you will have the choice to start between the 2 windows installations, no need to define the order.
I saw it after my message that it was reproducible so it's fine, no need to use it for 4 to 5 months.

However - and I don't have enough test data for this yet - there seems to be a difference, depending on how long the PC was turned off prior to this stress test.

Restart PC = No crash with same setup (1 or 2x tested) (eh, one of those two was a bit unclear)
PC left off for an hour = No crash with same setup (1x tested)
PC left off overnight = Crash (3x tested)

-------------------------------------------------------------------

A little whisper in my left ear tells me "check up the Power States"....or PSU.
And maybe also the Control Panel\All Control Panel Items\Security and Maintenance\Reliability Monitor if there should be anything....

-----------------------

No clue what kind of Power State you got. (powercfg /a in CMD(adm))

Standby (S3) status. If you see the "This standby state is disabled when S0 low power idle is supported" message, it means you can disable Modern Standby to enable Legacy Standby (S3).

This article shows how to disable Modern Standby...
How to Disable Modern Standby in Windows 10 & 11

No logic why Modern Standby should cause this, but it is still new. Probably you don't got it either.
-----------------------

OCCT, a free CPU/GPU/PSU stress testing tool. OCCT stress tests your CPU, GPU, and motherboard, and in the process, can discover any faults with your power supply unit.
How to Fix a Windows 10 Kernel Power Error in 5 Easy Steps

And also the Power Plan with advanced power settings, but there I think you have had a look some hundreds times...

HedgeToaster said:

I mean - why am I not even getting any bluescreens? As in, never?

I vaguely remember getting a bluescreen on this PC once or twice at some point, possibly for unrelated reasons, but that was so long ago I can't recall what year it was.

Is the motherboard intercepting errors before Windows gets a chance to handle them?? Are bluescreens deactivated somehow???

The difference is that blue screen could effectively stay indefinitely without any harm to the computer, but windows juist restarts the system once it has done the dump. i also think earlier computers way back in the past may of stayed on a blue screen and the user would of had to reset it themself but i am not certain. The main point is blue screen can happen without harm.

Black screen to reboot means fatal error in terms of something has happened and the computer thinks it needs to cycle to save itself from damage hence why there is no dumps and it just turns off and reboots. Although it may not be in any direct harm regardless.

I don't really know in depth with diagnosing crashes like these only basics but i guess virtual box is at hand here and maybe to do with your hard drives or virtual hard drives or even virtual hardware.

The images you relayed are saying kernel power, volmangr and vbox so there is probably an issue here with vitual box and the volume manager driver maybe even hardware virtualization.
its hard to say personally without knowing enough but just my guess and an easy way to test would be remove virtual box and/or disable hardware virtualization in the bios and then test over the span of a week.

You could also just run the bsod tool from the respective forum section and then relay your logs there. There is one user that seems quite knowledgeable about this stuff.