Computer blue and black screens of death (1 Viewer)

[tmyers]

To those who are more versed in hardware then myself, any idea what could cause constant BSoD's? The error varies when it crashes, but the most common one I see if Memory Management. I also see Kernel errors, Corrupted Page errors, IQL (dont know if thats right, its some 3 letter acronym) and security errors. The screen also graphically tears and artifacts when it happens, which I have not personally seen before.

As the problem initially only reported the memory management error before branching out to others, I had suspected bad RAM. I have swapped the sticks out as well as tested them in another computer and they caused no issue. I then suspected maybe the motherboard had a bad channel or something until the screen tearing during the crash started. I removed my GPU and ran off the on board graphics but that also did not stop the crashing.

Over the last week it progressed from blue screens to black screens and as of Sunday the computer will no longer boot. Think the motherboard went bad?

 

[The_Doc_Man]

If your computer will not boot now, it is sort of a moot point.

If that odd error was "IPL" rather than "IQL" then that is a reference to the interrupt priority level, which is a CPU hardware thing. Memory management could be a memory stick problem but could also be a memory controller problem, which is on the motherboard separate from the CPU. Kernel errors and corrupted page errors can be motherboard or in the CPU chip. With the security error, that is a "who knows" case that really doesn't point in any one direction - means it is an effect, not a primary cause of something. The observed screen issues also are a "who knows" case but the test you tried for GPU vs. board graphics was good. It suggests that the screen artifacts are a side-effect of some other problem.

As a future item, whenever you have a BSD, look for the error code, which will be a hexadecimal 8-character string. To get a BSD, you will usually have a code like either 8xxxxxxx or Cxxxxxxx. Of the remaining 7 hex digits, the first three will be 0 for a Windows problem or some other number for a utility program problem. Not going to swear to it, but the file system might have its own number, 001. From your description, I'm betting that a lot of the errors were C000xxxx, as having some other number in the place of the 000 would indicate a privileged program that would be tough as Hell to identify, because ordinary programs don't trigger a BSD.

Now I'll offer a blind guess, but the screen artifacts, multiple error types, and slowly degrading reliability suggest EITHER a failing power supply OR a bad fan (in which case your 'puter died a slow heat death.) If it had gone from working to instantly dead, I would not have made such a guess, and my guess could still be quite wrong. But that is what it smells like. Your guess as to a bad motherboard is also possibly correct but an effect, not a primary cause. Motherboards don't fail in that particular pattern without being helped to fail. If you can take it to a repair shop, good luck - but you will probably be replacing that motherboard.

IF you could reboot that machine and get to the Event Viewer, your system event logs will show you each crash AND will have the primary error code. They will also name the failing software module in some cases. Sadly, not all cases; e.g. a kernel interrupt error won't show the failing module because that interrupt code is intrinsic to the O/S. If your event logs would be available, you could also derive a history of how often the failures were occurring and whether they were escalating in frequency.

 

[CJ_London]

I had a similar problem a few years ago with not booting. Turned out I needed to replace the battery on the motherboard- which kind of fits with docs comment about power supply

 

[The_Doc_Man]

The slow degradation as the computer in question devolved from artifacts to Blue SoD to Black SoD isn't likely to be a bad battery, because that usually reaches a point where it just stops working. I.e. not the same profile, and those other failures would not be related to the battery. But yes, a bad battery will do you a bunch of dirt, too.

 

[arnelgp]

Incompatible drivers is the most common cause of BSoD.
It happens to me many times. No matter if I first create a Restore Point and
install the driver, it crashed and won't allow me to Restore (Esp on Win11).

 

[tmyers]

I had considered driver problems and such. Not long after having tested the RAM I had completely reformatted my drives and did a fresh windows install and update and the problem persisted.

I did not however think the PSU could cause the problem. I will pull it out and run a voltage test on it to see if that is what had eventually killed it. All the parts in it however are likely scrap if that was the case but hopefully I can identify the problem in the future before it results in another dead computer.

CJ, I have dealt with the CMOS battery on the motherboard giving out before as well and to Docs point, that is typically a the computer works and now it doesn't scenario. This slowly degraded over the period of several months then got rapidly worse in a few days time until it eventually died which sounds like an under voltage problem now that I think about it. Bet when I test the PSU it will reflect that.

 

[The_Doc_Man]

IF it is the power supply and an UNDERvoltage problem, you are less likely to have damaged the machine than if it was an OVERvoltage issue.

 

[tmyers]

That would be correct as I would assume if it was under voltage, I would have gotten temperature shutdowns from the hardware overheating due to it.

Either way, the computer will likely be "office spaced" when its all said and done.

 

[The_Doc_Man]

If there is any valid data on the hard drive, you can at least buy a new machine and try to add the previous hard drive as a secondary drive - as long as it isn't a super-compact laptop.

 

[AccessBlaster]

IQL (dont know if thats right, its some 3 letter acronym) and security errors.

Maybe IRQ:

In a computer, an interrupt request (or IRQ) is a hardware signal sent to the processor that temporarily stops a running program and allows a special program, an interrupt handler, to run instead.

I would at least reload the OS since there is no downside, especially if it's going to the ash heap

 

[The_Doc_Man]

That's another good possibility, AB.

 

[tmyers]

Maybe IRQ:


I would at least reload the OS since there is no downside, especially if it's going to the ash heap

When the problems progressed beyond just memory management, I had completely wiped my primary hard drive and reloaded a fresh install of windows and the problem didn't improve at all.

I did manage to test the PSU last night and Doc was right, the voltage was was higher than it should be and fluctuated a decent bit. I am honestly surprised I never once considered the PSU to be the problem given its the single oldest part in the computer at 7 years old.