Random BSOD, Freeze, or Black Screen Minutes After Startup

Hi All,

First post here, but have visited often over the years to learn about various subjects. Lately, I've been frequenting the many helpful BSOD pages thanks to the problem that is the subject of this thread. I'm hoping you wizards can help me pinpoint the offending driver or hardware before I resort to random uninstalls and eventually a clean Windows reinstall as more extreme procedures. I know what follows is a read, but it's hopefully well-written, so please bear with.

First, the System Specs

Specs are in my profile. Basically, this is a heavily upgraded OEM, the HP m9252p:
HP Pavilion Elite m9252p Desktop PC Product Specifications HP Pavilion Elite m9252p Desktop PC | HP® Support

The current configuration ran smoothly for over six months before the problem appeared. With the exception of the graphics card (more on that below), the most recent changes were the addition of the two HDDs at the beginning of this year, when migrating to SSD. Before that, the machine was completely problem free going back to its inception in 2008 (she's an old girl, I know).

The Mystery

Currently, the system is at a point where it always freezes, throws a random BSOD, or the screen goes black (monitor goes to sleep) within minutes after startup. It almost always happens within five minutes after the desktop loads, but occasionally up to ten minutes. This always requires a forced restart with a chance of one of these problems occurring again following reboot, for an unpredictable number of times. What began as typically a single occurrence has since become on average two crash/reboot cycles. The record was four.

Once the system is running for at least five minutes (the mystical "five minute rule" as I've come to know it), it works fine. I can leave it idle, or punish it with maxed out (ultimate) settings in Skyrim for hours on end, and it won't so much as hiccup. But after being shut down or in sleep mode for at least some hours, the startup problem always resurfaces in whichever flavors and quantities it decides to take. In all, it's a tolerable nuisance, but one that's surely the manifestation of one or more drivers or hardware gone wild.

Like the worst computing problems, this one arose seemingly out of the ether and has been in a downward spiral since. The variety and frequency of crashes seems to have gotten progressively worse over the more than two months since the problem first appeared. As far as I can tell, it all started with a sudden freeze on July 21 that had me thinking my wireless mouse and keyboard batteries needed replacing. Weeks went by with no problems, until the black screens started occurring. Still, these were intermittent and occurred days or weeks apart. Before long, the first BSODs started appearing. These were also intermittent, at first.

As time went on, it got to the point at which things are now. For basically the entire month of September, there have been one or more BSODs almost daily (uploaded for your debugging pleasure), but in any case, always one or more of the three types of crashes (freeze, BSOD, or black screen) shortly after startup.

Today's session went like this: (1) Freeze three minutes in, forced-reboot; (2) BSOD some minutes in, forced reboot; (3) different BSOD seven minutes in (broke the "five minute rule", I know!), forced reboot; (4) up and running six hours and counting. Each day the experience is different (last night was a black screen and BSOD), but the one constant is the crash(es) almost always occur within five, occasionally ten minutes post startup. Or at least that was the case until yesterday when the system threw a curve ball with a BSOD twenty minutes in. I pray this isn't a new symptom. So far, the two startups since then have remained true to the five (five to ten?) minute rule.

The Detective Work

Needless to say, I have tried everything I can think of in this harrowing driver/hardware whodunit. I started off investigating the basics. There were no manual changes to software, drivers, or hardware immediately prior to the first occurrence. If software or drivers are to blame, it's likely the result of one of those pesky auto- or frequent updaters. I'm looking at you Adobe Flash, Reader, Java, and MS Updates. I deliberately install the aforementioned updates manually, but if one of them, or one like them, caused the problem, it's a latent effect the likes of which are hard for me to trace.

I may have updated the Nvidia graphics and Realtek audio drivers once or twice each in recent history before the problem first arose in July, but it was a while before—three months prior for Realtek and at least a month prior for Nvidia. Besides, aren't we always told to make sure we have the latest drivers? I know driver updates can cause problems as often as they fix them, so I do make it a point to "test drive" them for a while before allowing them to stick around. If a driver update is to blame, again, the path to destruction was surely insidious.

Other measures I've looked into on the software/driver side of the coin include:

• System restore - Not an option because the oldest restore point is 7/30, after the 7/21 onset.

• Clean startup - Based on this tutorial: Troubleshoot Application Conflicts by Performing a Clean Startup. Problem still occurs, suggesting hardware.

• Safe mode - Problem still occurs, suggesting hardware.

• Startup repair – Inspired by: Startup Repair - Run 3 Separate Times. Problem still occurs, suggesting hardware.

• Antivirus/antimalware scans: Ran full tests with both MSE and MBAM as suggested throughout this forum. Nothing but a couple potential Java exploits (damn you Java!). Confession: I've been living dangerously for years with no antivirus software; just router and Windows firewalls and common sense. Based on the essentially clean scan results, I seem to have done okay.

Over the course of this lovely mystery, I've gone back and forth on thinking the cause is software/driver-related versus hardware-related. The progressive nature of the problem certainly suggests dying hardware (a real possibility given some of the core components of this machine have been in use since 2008: motherboard, CPU, power supply). But I find it paradoxical that the problem is limited to shortly after cold startup, and there are never problems under stress (e.g., the aforementioned nonstop hours of gaming). Nonetheless, I've done the essential hardware tests, all of which have come up clean:

• Windows Memory Diagnostic and MemTest86 RAM tests overnight, twice

• Chkdsk, HD Tune, and manufacturer utility checks of both HDDs

• HWiNFO shows temps within normal, even during stress

I've avoided running actual hardware stress tests (e.g., Prime95, IntelBurnTest) because the problem does not seem stress-related at all. Unless somehow the system is more stressed five minutes after cold startup than it is hours into maxed out gaming.

I should also mention that I actually upgraded the graphics card in August, partially hoping it would eliminate the problem. It didn't, and since the problem existed before and after installation, I'm confident ruling out the graphics card itself, though not necessarily the drivers (which are currently up to date).

I've done no overclocking of any kind (and can't with the OEM BIOS).

Finally, I've tried to make sense of the ample BSOD minidumps using BlueScreenView and WinDbg, but to put it simply, they're all over the place. Just about every major hardware driver has been implicated at one point or other, and the logs comprise a veritable who's who of bug checks. I've read that the abundance of Microsoft drivers being implicated suggests hardware failure, but don't know how to narrow it down. However, my expertise at combing minidumps is quite limited, and I hope someone with more experience sees something I don't.

Off hand, is there any specific hardware failure one might suspect given that the problem consistently occurs shortly after cold startup, but never after? I've never experienced power supply failure. Is it likely that? My other "gut feeling" suspect is the Intel Rapid Storage Technology controller driver I installed in March to address an issue where the system would hang coming out of sleep after installing the SSD (doing so worked).

Of note, the SATA controller mode also changed itself from AHCI to RAID when I installed the two HDDs in January, and I cannot switch back to AHCI in the BIOS without Windows failing to load. I've researched this and learned that a clean Windows installation is required to go back to AHCI. So for now, things remain in RAID, even though no array is set up. No idea if this has anything to do with the problem, but it's been like this since the beginning of this year, nearly seven months before the problem surfaced.

Thanks in advance for reading this novel. Any help is greatly appreciated.

You have described exactly my problem, dude!
If the crash (not necesarilly the BSOD) came in a middle of skype or VOIP conversation, can you still talk/hear for 30 sec - 1 minute, like me? With black screen on, in stand by. I can.
I hardly wait to find the solution for our problem...

Maybe is usefull to check this post too: BSOD minutes after desktop starts

Thanks for the replies!

temistocle, I do use VOIP almost daily, but never during a crash. I intentionally hold off doing anything with the system for at least five minutes after startup because I know a crash is coming. My problem could be attributed to so many different things, I won't be surprised if any of them proves to be the culprit. Best of luck with your problem.

Boozad, I in fact ran MemTest86, not MemTest86+. Will run the latter tonight. Are there any specific settings I should opt for, or is the default sufficient?

I have read in many places that Daemon Tools is known to cause BSOD, and it was indeed going to be one of the first to get the ax if I began randomly uninstalling things. However, it has resided on this system since 2012 to no ill effect that I know of. Of course, it may not be playing nicely with something that has changed since then, so I will uninstall it immediately.

I will remove PowerDVD as well. I know it throws errors after each startup session in the error log. Didn't see it in the dumps myself, but I take your word for it.

Unfortunately, the only way for me to test any troubleshooting measures will be to turn the machine off for a while, and then on again, as that's the only time the problem occurs. Not sure how long it needs to be, so did a test and can confirm the problem will occur after the machine is off for as little as 40 minutes. It may well happen in less time.

What's more, following that 40 minute downtime, I saw an entirely new phenomenon: Five minutes after initial startup, the system just auto-restarted to boot. No BSOD, freeze, or black screen. On the second startup, within seconds, the screen froze like this:



The squiggly white line in the middle is actually on the screen, not a reflection. My spidey sense says hardware failure. I'd suspect the graphics card, but as stated in the OP, the card is new and the problems predated it.

I am now working from the third startup, and all's well for now. I'll report back after uninstalling DT and PowerDVD.

Thanks again for the help!

Well, it wasn't Daemon Tools, sadly.

Uninstalled it as instructed and shut the machine down for a while. Turned it on a couple hours later and there were no crashes. Turned it off. Checked again a couple hours later and no crashes. I was both hopeful yet skeptical that the problem was solved.

Sure enough, today, I was greeted with all three flavors of crashes, as if the culprit is saying "Not so fast!"

First boot: Yet another new BSOD (this time pointing at the Hauppage TV Tuner in BlueScreenView, and the Xbox controller receiver driver in WinDbg, both of which I doubt are the true cause).
Second boot: Freeze.
Third boot: Black screen. But wait, there's more: Along with the black screen, there was, for the first time ever, a BIOS beep code. Two short beeps about a second apart, every three seconds. It kept going until I powered off.
Fourth boot: I am now working from the fourth startup, 50 minutes in, with no problems.

According to AMI Bios Beepcodes, two short beeps indicates either:

Memory parity error - A memory parity error has occurred in the first 64K of RAM. The RAM IC is probably bad

or

POST Failure - One of the hardware testa have failed

However, I'm not sure if those beep codes only refer to beeps during startup, or if they also apply to beeps during a crash.

I didn't uninstall PowerDVD yet and didn't run MemTest86+ overnight as I wanted to see how today's startup fared after the promising results from uninstalling DT. I'm taking it all one step at a time to try and isolate the cause. After today's events, I'm 99% certain it's hardware failure, but the question is still what.

I will definitely run MemTest86+ for at least 8 passes tonight.

Today's dump file is attached in case anyone cares to look.

Oh, man... Blueworld, keep up the good fight! :P

I forgot to add earlier: besides stressing drivers with "verifier" win7 command, I followed an advice on another similar thread here and installed a program which check and update the drivers (I used freeware Driver Booster). I discovered that all my drivers are damn old. I mean these drivers:

Attachment 335537
Attachment 335538

So I let the DB to update them.

I have reach that thread because I experienced the same problem (discovered in Windows Events) with RoloDman: Nvidia GeForce Live Kernel Event.

I want to mention that, in Windows Events, I discovered a lot of this Errors:
Event 12, HAL
The platform firmware has corrupted memory across the previous system power transition. Please check for updated firmware for your system.

I hope I dont have to update my BIOS...

After my last post a week ago, I uninstalled PowerDVD and began a battery of MemTest86+ testing spanning nearly a week.

The testing was inconclusive:

On the seventh pass of the initial test, with all DIMMS installed, a ton of errors showed up (5,120 to be exact). I promptly exited and began one-by-one testing of each DIMM and slot to try and isolate any problems, keeping a log as I went. The third test--DIMM 3 in Slot 1--coughed up 80 errors during the third pass.

Eight tests later (11 tests total), with at least 8 passes each, tests 1 and 3 were the only times I saw errors. Naturally, much testing was focused on DIMM 3 and Slot 1 in combination with other slots and DIMMs, respectively, but the errors of test 3 could not be reproduced.

Nor could the numerous errors produced when testing all DIMMs simultaneously be reproduced, not even after another 8 passes. I may try to re-run this all-DIMMs test for at least 10 passes, but it's a very, very long test approaching 20 hours, so rarely convenient. All DIMMs are currently back in their original positions.

Despite the inconclusive testing, I am now inexplicably on day two of not encountering crashes!

I know it's too soon to celebrate, but if I never see the problem again, it will have vanished just as mysteriously as it appeared. The only things that happened since last week, were the uninstallation of PowerDVD (could this really have been it?!) and the extensive MemTest86+ testing (could the simple act of removing and reseating DIMMs over and over have cleared out some voltage anomaly or something?).

Anyway, just wanted to post an update and let anyone following this know I'm still at it. If anyone has any thoughts on the strange test results and the even stranger seemingly-resolved status, I'm all ears.

I remain cautiously optimistic.