Windows 7 Using CPUID Hardware Monitor and what to look for?

[Kaidonni]

I've been posting over in the BSOD forums, and I've downloaded CPUID Hardware monitor in a bid to find out what exactly is wrong. Only thing is, I am not certain what I should be looking out for, so I've attached the monitoring data file to this post if anyone is able to help me make heads or tail of it.

My hardware setup:
ASUSTeK Computer INC. M4A78LT-M-LE
AMD Athlon II X4 640
Nvidia GeForce GT 240
Memory 4.00GB
Microsoft Windows 7 Ultimate 64-bit 7601 Multiprocessor Free Service Pack 1

I notice that the idle temperature for the graphics card is holding at 53 at the moment (it has a 105 degrees maximum under load). On Sunday I plan to run a series of tests to ascertain what happens prior to BSOD events as regards hardware performance, since the last three weeks none have happened - they all started in Windows, but seem to be more likely in games (add to that, they magically started after weeks of smooth running of said games). Replaced harddrive because that failed all the manufacturer tests, and the issue had been fixed...then came right back (obviously it wasn't the harddrive causing it ). I have kept off games for a bit because I rely on my brother to assist in fixing problems, and he's been away (if anything goes wrong, who do I have to blame?).

 

[Digerati]

Although you did not actually say, I am assuming you are posting here because your system keeps crashing with BSODs.

What is the error code/message of the BSOD? Do you always get a BSOD, or does it sometimes just hang, or reboot.

I see nothing wrong in your attached file, but it really is not that informative. For example, it does not show what the PSU's +5VDC voltage is sitting at. I might suggest Link Removed from the makers of CCleaner. It includes a great option to "publish" your system specs (less any personal information) and provides a convenient link to post to forums for easy review.

Download, install and run Speccy
When the scan completes, click on File > Publish Snapshot,
Click Yes,
Copy provided link
Click Close.
Return to forum post, right-click in the post text field and select Paste to copy clipboard contents (link) into your post.​

BSODs are often caused by a driver conflict or other hardware issue, perhaps caused by a flakey PSU, or high temps, or a failing motherboard or other component. The fact it is not happening now that you have backed off gaming suggests it could be heat (since gaming is so demanding), but it might be a problem with your graphics solution too (since games tend to tax graphics solutions too).

Obviously, because so many things can cause these symptoms, troubleshooting can be very difficult and time consuming and often works only by eliminating all potential causes before finding the real cause.

I might suggest you pull all but one stick of RAM and see if you have problems, swapping in your other sticks as a process of elimination.

 

[Kaidonni]

I'll be trying memory tests tomorrow evening, first using Windows Memory Diagnostic, now that my brother is soon to be back. No one will be needing the computer tomorrow evening, so it's the best time. I know of Memtest 86+, but I was hoping for a quick go at a lighter solution first before asking him to do anything.

My BSOD errors have been PFN_LIST_CORRUPT most commonly, with DRIVER_IRQL_NOT_LESS_OR_EQUAL thrown in, and a DRIVER_CORRUPTED_EXPOOL (the last one struck when Empire Total War froze the system, and I tried rebooting into safe mode - it went BSOD on me when loading the driver list). The funny thing is, the very first BSOD errors I had were not in games - one all the way back in December (yet my problems first started at the start of February), then one on the 23rd February (days later, the BSOD events triggered by the chief suspect, Skyrim, returned). Skyrim is what caused most of the errors (i.e. it was running when they happened, not that it's actually responsible - played too much without any issues at all, and the BSOD events magically started from out of nowhere). The reinstall on the new harddrive only fixed it temporarily. Those BSOD events in Windows were shortly after boot-up, and both PFN_LIST_CORRUPT events (there was also an IRQL_NOT_LESS_OR_EQUAL when no one was in the house, no games running, hours after boot-up).

Actually, checking Bluescreen View, this time around SYSTEM_THREAD_EXCEPTION_NOT_HANDLED has also popped up among the errors.

I don't think over-heating because before the reinstall I tried a heavily modified Oblivion without issue. Not that it wouldn't have gone also - that's the similarity between Skyrim and Empire; that they worked for a good long while, then all of a sudden they caused serious issues.

 

[Digerati]

Well, I found this: Bug Check 0x4E: PFN_LIST_CORRUPT (Windows Debuggers)

 

[Kaidonni]

Snapshot from Speccy:

http://speccy.piriform.com/results/HhPSArH7FZJumKOmpVR7Lfu

 

[Digerati]

All looks good in there. I did notice you have Windows Update disabled. That's fine as long as you manually keep Windows updated. It looks like you are a couple weeks behind.

Still, all looking good does not mean everything is.

Note that MemTest 86+ (and Windows 7 built in RAM tester) is a good program, but none of those RAM testers are conclusive. If they find errors, the RAM is likely bad, but sadly, RAM often tests fine, but fails under real-world use - or (as is often the case) simply fails to play well when installed with other sticks.

 

[Kaidonni]

Well, I've decided to wait until tomorrow evening for any RAM testing. This evening, I'm capitalisng on the fact I can finally mess around with games again, and I'll be testing one that most certainly shouldn't have a problem working, one I've been modding for a while. If the errors are memory related, then being forced to load, exit, and reload the game to test changes may trigger an event (and I've never had any issue having to do this, so I'll know something hardware related is wrong if it causes the system to go unstable).

 

[Digerati]

One think you can try too when gaming (pushing the computer) is opening the side panel and blast a desk fan in there.

 

[Kaidonni]

Unfortunately, haven't gotten around to testing temperature yet. My brother had downloaded a CPU temperature monitor, but it only monitors the CPU and nothing else. Only got HWMonitor the other day, and the graphics card is holding at 48 idle (is that right?).

While I could do Windows Memory Diagnostic tonight, the problem is if it doesn't find anything - and when reading up on another forum on how to use it, someone was having trouble exiting (it always ran the diagnostic when they rebooted, it only stopped after they removed the RAM sticks and put them back). I've never actually touched the RAM sticks, no idea how to remove them or put them back (I need to be shown). Testing a few games should allow me a chance to see what is happening with temperatures on the graphics card, and since a fair few BSOD events seem to stem from instability of a certain element, a stint at modding ought to see what happens.

 

[Digerati]

GPUs (graphics processing units) tend to run warmer than CPUs so 48°C is great.

Removing and replacing RAM is actually pretty easy. Your motherboard manual will show you how. The important thing to note is the modules are keyed so they only go in one way - though they can be forced the wrong way. The other key thing to remember is static in your body can destroy ESD sensitive devices without you (the human) even being aware a static discharge occurred. So you MUST ensure you unplug the computer from the wall and you touch bare metal of the case interior to discharge any static in your body (and put you and the computer "at the same potential) BEFORE touching the RAM.

 

[Kaidonni]

I actually went with the inbuilt memory test, and it didn't find anything wrong. Not that this means nothing is wrong. Next up would be memtest, and possibly trying out Furmark or something else to stress the graphics card. I'd test the games more, but I'm terrified! I know unless I use something Like Furmark I have no choice, and that it may have been a driver issue (updated Nvidia drivers three weeks ago, right about the time no one was around so I couldn't test to see if it helped), and trying out Skyrim might help answer whether or not it has gone, even temporarily, but still...scared!

 

[Digerati]

I actually went with the inbuilt memory test, and it didn't find anything wrong. Not that this means nothing is wrong. Next up would be memtest,

No, it does not mean nothing wrong, but no errors is still encouraging.

No need to be scared but certainly cautious. Make sure you have a recent copy of any data you don't want to lose.

 

[Kaidonni]

All is not well...

From my other thread in the BSOD forums, Link Removed:

New BSOD event. Yesterday I was able to do some testing in games, had to restart the computer to do so (nothing serious, just one newly [re]installed game required a restart as directed by the installer). No issues with that, or with Empire either, then decided to go for Windows Memory Diagnostic. It didn't find anything wrong. Played about an hour of Star Trek Online, followed by forty-five minutes of web surfing. The shut off period was ~10 hours. This morning it must've been on for thirty-forty minutes, and while surfing the web it went (ironically, I was visiting a site named 'Can I Run It' at that exact moment - apparently I cannot!). Attached is a new set of logs, it's the latest - MEMORY_MANAGEMENT. I went through the rebooting process, and it may have happened a second time, as it was back at the screen for selecting to boot up normally or allow Windows to try repairing the issue a short while later - I allowed Windows to try it's magic.

A concerned family member thinks it's the games that did it, since there were no issues for three whole weeks... Maybe it is a game, but not the way they are thinking. I believe that something is causing system instability, and I'm thinking the RAM has something wrong with it, or one of the slots. The MEMORY_MANAGEMENT issue also happened on the 25th December when no one was on the computer, and it was basically left idle (not sure if a game was running, but it would have been alt-tabbed at the time). The only other thing that caused that was the Creation Kit hanging on being closed (31st January), and the dialogue box popping up to ask about waiting for the program to respond, to find a solution, or close the program.

Unfortunately, a family member is convinced games are the cause since nothing happened for three weeks. Since as they know little of computers at all, and one person who didn't even know how to load Skype criticised me, I'm just about done with this issue and going to build my own computer. I can't even be bothered with figuring out anything further because they don't want to know. I've wasted my time troubleshooting for nothing.

 

[Digerati]

It seems like you are being awful hard on your family for not being computer experts. It is hardly fair to criticize someone over Skype if they are not familiar with it. Not everyone uses Skype - in fact most people don't.

I tend to agree with your family member who suggests it is a game.

 

[Kaidonni]

Games that never caused issues for months and months beforehand, and didn't cause any last night.

 

[Digerati]

That does not mean there was no an update by the game maker that did not change something, or an update with Windows, or an update with one of the hardware drivers affected by a game. Anyone could have changed something. Often, a fix for one thing ends up breaking another.

I'm just saying.

 

[Kaidonni]

Certainly no updates by the game developers for at least two of those games, and they were fine previously. I would have also expected an error to occur last night as opposed to this morning while on the internet. The MEMORY_MANAGEMENT is an error that has not ever occured while playing a game, they have always been PFN_LIST_CORRUPT, DRIVER_IRQL_NOT_LESS_OR_EQUAL, or SYSTEM_THREAD_EXCEPTION_NOT_HANDLED (and a majority usually as a cascade of BSOD events after one in-game).

Yes, I was harsh, on both you and my family. I'm obsessive compulsive, so I keep going over the same things in my mind, and did that with my family. It's as if going over it a billionth time somehow will make it make sense. Unfortunately, not enough has been done to properly diagnose these errors - no memtest or driver verifier, plus the Device Manager doesn't even know what the motherboard is (it classes it as an unknown device). If you read my main thread, this is a problem that magically went away after the full reinstall + new harddrive, only to return from out of nowhere once again.

Something is clearly wrong somewhere, and given the errors are exactly the same as before the reinstall, I presume it's the same cause. Prior to the reinstall, I did test Oblivion, Stalker, and Star Trek Online without issue, only Skyrim started causing trouble. No next-day BSOD events or anything. It's darn inconsistent. Until the other tests have been run, and the motherboard sorted out... Like I said, OCD, so I go over it and over it, and that isn't helpful when the last 3 weeks I've not been able to troubleshoot further.

EDIT: I lie, only one MEMORY_MANAGEMENT, 25th December. The 31st January was a PFN_LIST_CORRUPT. So this is indeed a rare error. Different code for that parameter one (it being 8885 as opposed to 5003).

 

[Digerati]

Yes, I was harsh, on ... you

I don't take things personal in forums, unless they get personal - if you know what I mean.


As far as memory management error - hard to say if that is bad code banging into good RAM, or good code banging into good RAM. Do note that if bad code is in the OS, most Microsoft updates fix things, not break them. But when they do mess up, millions of users scream about it. That does not mean other software coded to use the bad code will word once WU pushes out the update.

Or the memory manager on the motherboard could be going - that's new motherboard time.

 

[Kaidonni]

I've been given the all clear on memtest. The errors have always seemed...odd. They started on the 19th December by all rights - a PFN_LIST_CORRUPT, thought nothing of it. Then the 25th December, and I also thought nothing of it. Nothing until the 31st January, then one on the 1st February (both in Windows at the time, no games running). The first one in any game was 2nd February (a game I'd played for a month prior to that), and it was consistent (less than 5 minutes in, it'd happen, and cause at least one more BSOD in Windows after reboot). Figured it might be the harddrive due to a bad block warning, went through the diagnostics that seemed to confirm this, replaced and fully reinstalled.

For a whole week and a half, no issues - then a PFN_LIST_CORRUPT struck on the 23rd February shortly after boot-up, in Windows. Tested the game after that, no issues until Feb 27th when it became consistent (although no additional BSOD after reboot). Feb 28th confirmed this, plus two additional BSOD events after reboot. Since Windows was updated on the 25th, I think it may have returned due to driver and WU incompatibility issues (didn't actually play the game on the 25th or 26th of Feb), but I haven't tested that game with the newer drivers, and I haven't considered scaling them back either yet to what worked prior to February. The MEMORY_MANAGEMENT isn't anything that has popped up from the game-related BSOD events, those have pointed towards drivers usually (have not run driver verifier yet, do not leave the computer running over night so no idea when to run that), so they may well be unrelated. It may even be the other BSODs were only driver related, and before the reinstall brought on by corrupt harddrive, this time brought on by a WU and having Nvidia drivers that don't agree with it.

The errors brought on by the games have funny timing - I'd expect them to be absolute from the beginning.

 

[Digerati]

I wish I had more to offer. Certainly many games push our systems to the limits - especially games that exploit extreme graphics features.

The errors brought on by the games have funny timing - I'd expect them to be absolute from the beginning.

Absolute from the beginning. I wish. If problems were consistent, they could be reproduced easily. As a troubleshooter, if you can reproduce a problem at will, it typically is easy to fix.

Since it is likely no two gaming sessions proceed through the game in the exact same manner, consistency is not likely.