Random Shutdowns

[PsychicTide]

I've recently built a new PC and decided... what the hell, I'll get an OS that causes baldness... Windows Vista Home Premium 64bit. So far it has locked up two times for no reason and restarted my PC completely (I want to catch the problem before it could possibly get worse). Sometimes this happens when I'm uninstalling a program (uninstalling PC probe.. a motherboard status program provided by ASUS)... sometimes when I boot up certain games like Empire: Total War (every time I open this game it happens). All of the newest updates and drivers are installed for every component.

Does this happen often with Vista? Maybe it's specifically x64 because of x32 software incompatibilities? I doubt that its a hardware issue but I'm not ruling it out.

• 550w BFG Tech Power Supply
• ASUS P5Q Pro Turbo (LGA 775 P45)
• Intel Core 2 Dou (2.8Ghz 3MB L2 Cache) Constant temperature of 30 (not in use) - 40 (heavy load) Degrees C.
• Aftermarket Zalman 120mm 2 ball CPU cooling fan and thermal paste
• 4x1GB DDR2 800 SDRAM (2 dual kits)
• 512MB GDDR3 ASUS EAH 2900XT (ATi Radeon) Constant temperature of 52 (not in use) - around 63 (heavy load) Degrees C
• Seagate Barracuda ATA IV (IDE - 120GB)
• Buslink DVDRW812D (IDE - secondary connection from same wire as hard drive)

Let me know what other info you may need or if you have any ideas on how to correct this. Any info is greatly appreciated.

[Quackware]

It might be a problem with the RAM.

Do you have the patch from Microsoft for 4gb ram systems?

You should probably check your BIOS and make sure all the timings etc are correct.

[crunchie]

You do not need a patch with 64bit systems for RAM. It is automatically recognised.
Could possibly be an overheating problem. Make sure that your HSF is in correct contact with the IHS and that you have the right amount of TIM in between.
Check if the shutdown feature (if your bios has it) is not set too low.

[PsychicTide]

Thanks for the replies guys.

You should probably check your BIOS and make sure all the timings etc are correct.

RAM timings seem to go with the RAM... as far as I know everything is in order.

Could possibly be an overheating problem. Make sure that your HSF is in correct contact with the IHS and that you have the right amount of TIM in between.

Hmm that's a thought I suppose... It would have to be a pretty big jump in heat fairly quickly... only thing I've seen get that hot that fast is a chip that doesn't have enough power supplied to it (although I have seen the no-heatsink videos online :) ). would improper thermal paste application cause that much of a jump in heat? I've benchmarked my CPU without it shutting down or going much over 40 Degrees C... I'll monitor it over the next couple days

[crunchie]

It could be software related. You would need to shut down programs 1 at a time and then wait to see if the freezes still occur.

[Yiuca]

You do not need a patch with 64bit systems for RAM. It is automatically recognised.
Could possibly be an overheating problem. Make sure that your HSF is in correct contact with the IHS and that you have the right amount of TIM in between.
Check if the shutdown feature (if your bios has it) is not set too low.

Actually there is a patch in Vista 64-bit to recognize 4GB of RAM because originally there were a bug that meant 64-bit didn't recognize all 4GB+ of RAM. Technically 32-bit systems can also recognize & address 4GB anyway, but the point I'm making here is that there were indeed a patch for that.

@OP: Watch voltages to see if you get any spikes, for example if the CPU is a big drop in voltage it can cause freezing & such drop in voltage can cause automatic shutdown, for example if you start a game, a component will require more electricity to be pushed to it, if the PSU can't supply that then the effects are the same as what you are experiencing. (This includes times where it's not able to supply enough even when you're not running a game but it usually happens slower or similar)

[crunchie]

Only said you didn't need one :).

[rs_sis]

Hi,

try to check the eventviewer !

i think that there might be a "overheating" problem with your videocard.
you can adjust the shutdown temp for your videocard
in BIOS/video card software..

[PsychicTide]

Once again, thanks for the quick reply guys.

It could be software related. You would need to shut down programs 1 at a time and then wait to see if the freezes still occur.

I'll go ahead and begin the long process of checking every running program and see if it still restarts when attempting to run the game. (may just be the game, but then why would I restart?)

...but the point I'm making here is that there were indeed a patch for that. (x64)

So would I need to manually install this update or would I automatically get this update if I've run windows update fully?

Watch voltages to see if you get any spikes...

So far I've run Futuremark (benchmark) and found the volts start at around 1.1xxx and at max go to about 1.2xxx I'm not very familiar with the average voltage on a CPU... what do you think?

try to check the eventviewer !

The only meaningful thing I've been able to come up is this.

WMI error ID=10... "Event filter with query "SELECT * FROM __InstanceModificationEvent WITHIN 60 WHERE TargetInstance ISA "Win32_Processor" AND TargetInstance.LoadPercentage > 99" could not be reactivated in namespace "//./root/CIMV2" because of error 0x80041003. Events cannot be delivered through this filter until the problem is corrected."

It looks as though I've had this error several times now (both while running normally and close to the restart events). However, I'm not familiar with where this is being pulled from.

you can adjust the shutdown temp for your videocard
in BIOS/video card software.

I'm not able to find this option in my BIOS or graphics control center.. is there a specific name for it or will it be pretty much spelled out for me? (would also like to check this for the other components if possible)

[Yiuca]

(may just be the game, but then why would I restart?)

Insufficient Voltage, Critical error within the game (unlikely unless widely reported) software conflict (possible albeit unlikely) overheating (possible but if you say you've been watching temps & they are ok then unlikely)

So would I need to manually install this update or would I automatically get this update if I've run windows update fully?

If you can see 4GB already then I wouldn't worry about it, it's automatically added in the SP packs & is applied by Windows Update anyway, however really if you see that you have 4GB then no need to worry since the patch is only to address a bug that stopped the OS from being able to recognize all 4GB.

So far I've run Futuremark (benchmark) and found the volts start at around 1.1xxx and at max go to about 1.2xxx I'm not very familiar with the average voltage on a CPU... what do you think?

Well generally every processor is different, there is a vdroop value used in overclocking communities it's used to express the amount of dropped voltage when raising the voltage to keep a core stable, that is if you up the clock speed it requires more voltage otherwise the CPU will be unstable however when you increase the voltage there is vdroop, this can be small or large & generally vdroop is the different between the actual voltage your CPU is getting compared to the setting you have in bios. Of course that doesn't mean you should wang the voltage up even more as that's just not very clever. :) That's about as basic as that explaination gets anyway the point is it's generally different & the usual voltage I see is around 1.3v, when underclocked (for example while system is idle) it's around 1.1xxx & max because of vdroop usually is 1.2xxx... It may not even just be CPU voltage however, it can be any other component not getting enough juice, RAM, GPU, Motherboard anything like that not getting enough will take the rest of the machine down with it when it shuts down.

The only meaningful thing I've been able to come up is this.

WMI error ID=10... "Event filter with query "SELECT * FROM __InstanceModificationEvent WITHIN 60 WHERE TargetInstance ISA "Win32_Processor" AND TargetInstance.LoadPercentage > 99" could not be reactivated in namespace "//./root/CIMV2" because of error 0x80041003. Events cannot be delivered through this filter until the problem is corrected."

It looks as though I've had this error several times now (both while running normally and close to the restart events). However, I'm not familiar with where this is being pulled from.

That is more of an annoyance than anything, it's because of the WMI filter being accessed without sufficient permissions. Paste the following script in notepad & save it with the extention vbs (visual basic script) after you've run this script, you also have to manually clear the event viewer of that error, after that it be ok.

Just so you know it's safe, It's dished out by MS: http://support.microsoft.com/kb/950375

strComputer = "."
Set objWMIService = GetObject("winmgmts:" _
& "{impersonationLevel=impersonate}!\\" _
& strComputer & "\root\subscription")


Set obj1 = objWMIService.Get("__EventFilter.Name='BVTFilter'")

set obj2set = obj1.Associators_("__FilterToConsumerBinding")

set obj3set = obj1.References_("__FilterToConsumerBinding")



For each obj2 in obj2set
                WScript.echo "Deleting the object"
                WScript.echo obj2.GetObjectText_
                obj2.Delete_
next

For each obj3 in obj3set
                WScript.echo "Deleting the object"
                WScript.echo obj3.GetObjectText_
                obj3.Delete_
next

WScript.echo "Deleting the object"
WScript.echo obj1.GetObjectText_
obj1.Delete_

Question, when your comp shuts down, does it go through the shut down procedure or does it instantly power down for seemingly no reason? Like one moment it's running then bam it's powered down... Or do all the programs close & Windows shuts down & then turns off the computer as though you'd clicked to shut down?

[PsychicTide]

It says 4094MB everywhere I look for RAM so I assume its alright on the patch. I'm not familiar with Voltages enough to really mess with them... but I never go under 1.1v or over 1.3v.

Critical error within the game (unlikely unless widely reported)

It looks like there are a lot of crashing problems with this particular game on Vista 64, but they seem to just crash to the desktop (not restart the PC)

software conflict

Currently doing some research on this

As for the overheating... I've placed my PC so as to directly take in the cool air from my central air and brought my average CPU temp down to 20 Degrees C while not in use (my AC seems to be running all the time... maybe this is why my electric bill is so high :))

Thanks for the event log message removal program :).

I'm still up for any suggestion posted or to expand on current suggestions.... even if its just a topic to google let me know. I'll continue to watch my volts and temp.

[PsychicTide]

Forgot to mention this... I shutdown just fine it seems... closes programs, brings up shutdown screen, 8 second shutdown total.

[Yiuca]

Empire Total War I hear is totally bugged out, reason why I haven't bought it & played it myself is because of that! :D Does the problem happen with any other games? Go into specifics, it can sometimes really help knowing all the ins & outs, since sometimes it can be the smallest most irrelevant thing you could ever hope to think of.

Sounds like the voltage is alright, have you got any programs that can monitor & actually record voltages as they happen? I remember when I used to have an Asus board, I think Asus Probe or something else they had could be set to record voltages constantly, it were how I found out my PSU had normal voltage for the most part but every so often the voltage dropped quite abit in just a spike for 1 second or so after a few of those the comp would just power down... So that's the sort of thing I'd probably be looking out for but obviously you won't want to sit there constantly looking at it, so it helps to have something that records it constantly for you to look at later. (Gives you an idea of the last moments before shut down too)

Just to try it, (so it's been tried) it might be worthwhile disconnecting the PSU from say the motherboard, GPU & HDDs, then reconnecting them up, along with that re-seating the RAM... It strange how many problems have come from such stuff, so it's better to try it just so you know for sure, sometimes they come a little loose, I'd assume because of the small vibration of fans/HDD sometimes but I don't know for sure why! :D I just know that it is a small lead to try, computers so complex there is alot to try but it doesn't hurt to try the unlikely, when I've fixed other peoples comps, it's these unlikely things that have saved me many hours of headache! (Hence why I suggest them I guess)

[PsychicTide]

Does the problem happen with any other games?

So far this is the only game I've found to cause the problem, but as I said before it did happen one other time while uninstalling the PC Probe 2 provided by ASUS.

have you got any programs that can monitor & actually record voltages as they happen?

I've got AI Suite provided by ASUS that monitors and records temps, volts, and speed of fans... Hopefully if anything goes funky it will catch it.

it might be worthwhile disconnecting the PSU from say the motherboard, GPU & HDDs, then reconnecting them up, along with that re-seating the RAM

As soon as I get done posting this I'll go ahead and make sure every thing I can unplug or pull out (short of the CPU and HSF because thermal paste is a pain) is properly reseated... That's a great thought actually.

I'm going to go ahead and mark this thread as solved because it seems to be an intermittent problem. Thanks alot for all the input from you guys... It definitely helped out for troubleshooting.