Intermittent BSODs on Vista

[flyingmeatball]

Hello,

My computer is experiencing the BSOD at random intervals, and I'm not quite sure what to do to diagnose it. It typically happens when I'm playing video games for a while (rarely when it first turns on), although I've had the guy on for a while and it will randomly have happened without me doing anything on it. Once it bsod's it will keep crashing on boot until you completely unplug the machine, let it sit for 30 seconds, then power it back up. First, an error message appears saying overclock failed, please change settings. If i shut it down and restart then it will work fine again...until it crashes 30 mins-1 hr later.

I just changed the power supply with one I had laying around (580 watts - modular cords, good supply) and keep having the same issue. I downloaded a temp program speedfan which is saying my cores are between 51-56, and this is right after playing, and high 40's when just surfing the web. Plus the case is lying competely opened, so heat shouldn't be the culprit? When my system arrived I had an issue with one of the sticks of ram, and corsair replaced it, so unless something has broken since then it shouldn't be the RAM either.

My system is:

Intel 2 quad duo CPU at 2.5ghz (4 cores)
2 * 2048 Corsair DDR2 1066 ram sticks (although for some reason in dxdiag is says 3326 mb ram?)
ATI Radeon HD 4800 series, with 2913 mb ram
500 gig @ 5400 RPM Seagate drive
ASUS P5Q-E LGA 775 Intel P45 ATX Intel Motherboard
Windows Vista home edition.

Any suggestions for troubleshooting this would be much appreciated, I have no idea how to tell what is failing, especially how to diagnose what the BSOD's mean, and how to tell if its the mobo?

 

[Archean]

ATI Radeon HD 4800 series, with 2913 mb ram

Which card is that?

Can you please upload minidumps for analysis with your next post? You can find them in C:\windows\minidump. If there are multiple dumps zip them in a single file and attach with your post. Regards

 

[flyingmeatball]

Here is the graphics card's full specs

ASUS EAH4870 DK TOP/HTDI/512MD5 Radeon HD 4870 512MB 256-bit GDDR5 PCI Express 2.0 x16 HDCP Ready CrossFireX Support

I've also attached a log of the last few BSOD's I've gotten.

Thanks for your help!

 

[Archean]

All the minidumps points at different errors with different OS processes, however, two of the dumps cited memory corruption as the issue.

I suspect that their may be one of these two issues:

1. RAM may be plainly bad, to check that out download memtest from TechSpot's download section and run it for minimum of 8 passes, preferably run it on each individual stick separately.

2. RAM may not be configured according to its specs on box i.e. timings/speed/voltage. Please verify that your BIOS is correctly set according to RAM's original specs. Regards

 

[flyingmeatball]

I ran memtest, and found that one of my sticks of RAM was bad. This is suprising because A) they're Corsair sticks and B) I already had a problem with my RAM, and RMA'd one of the sticks successfully about a year ago.

I'm currently running my system with 1 stick (2 gigs). I had adjusted the settings to what they were supposed to be: 5-5-5-15 with 2.1v. I thought this had cleared up the problem, except it just crashed on my again. I didn't even get a bsod, just straight to reboot. (I was in a game when it happened) Sure enough, I got the overclocking failed message on restart. Since I know my 1 stick of RAM in the machine is good, how do I take care of this "overclocking" problem? Where can I disable overclocking in the BIOS? I'm worried my mobo is burning out ram sticks...could I possibly "Underclock"? I'll sacrifice a little performance if it could ensure me stability.

thanks again for your help!

 

[Archean]

Can you please give me exact model number of your Corsair RAM ?

By the way it says 3326 RAM because you are using 32bit (also called x86) OS, and with that VGA card's frame buffer (i.e. Memory) and other memory allocations only leave 3326MB to be seen by the OS (out of total 4GB it can actually see at best), I've tried to keep it as simple as I could I hope it make sense.

Go to Control Panel, then to Administrative Tools, and open Event Viewer, try to find out what may have been recorded at the time of this unexpected reboot.

Lastly, what is the make / model of your PSU.

 

[flyingmeatball]

The memory is CORSAIR DOMINATOR 4GB (2 x 2GB) 240-Pin DDR2 SDRAM 1066 (PC2 8500) Model 8500C5D

The processor is Intel Core 2 Quad Q8300 Yorkfield 2.5GHz 4MB L2 Cache LGA 775 95W Quad-Core BX80580Q8300

I'm running Vista 64...shouldn't that mean I can use the full 4 gigs?

I'm getting this message on Startup "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."

but I'm not getting any errors on the crashes.

 

[Archean]

Can you please download CPUZ, then install and run it, once you have done that, go to About tab, and save report (.txt), post that file with your next post. By the way I somehow missed that you were running x64 OS which the debugger clearly tells

It is weird to get this error, I have seen it only once and that was on a Vista machine with SP1. Anyway, please try this and keep an eye on your event logs to see whether this resolves the issue.

Please open a command prompt with administrator account, and press enter after typing each line:

net stop winmgmt
DEL C:\Windows\System32\wbem\Repository\*.*

You need to reboot once you have done that, and let the system sit idle for 5-10 minutes, You will get several WMI warnings in the event log, but that is OK. After say 10 minutes or so, reboot your system one more time, hopefully this should resolve the issue. Regards

 

[flyingmeatball]

Here's the CPU-Z post...Do I do the system32 portion of that line even if running vista 64 bit?

 

[flyingmeatball]

Still getting the BSOD at random intervals...

 

[Archean]

Ok here is the link for your RAM's specifications (its a pdf file).

According to this, your RAM is certified with 5-5-5-15-2T at 2.1V settings, which you have said is the case as well. So that leave us with only one other option, i.e. to try to operate it at lower 800Mhz frequency, with 5-5-5-18 with 2.1V, and see how that works out.

 

[flyingmeatball]

That's how the RAM was running on Automatic. I had to manually adjust the ram to 5-5-5-15. Neither scenario has alleviated the crashes. Oddly, CPU-Z says my ram is still running at 5-5-5-18, even though in the bios settings I have it set to 5-5-5-15...

 

[Archean]

Frankly Asus board's can be 'royal pain .........' (you know what i mean) to configure sometimes .

Did you followed my suggestions given in post # 8 about WMI issue?

Also, please do post your newer minidumps.

 

[flyingmeatball]

Ok so thanks for all your help! I know a lot of ppl don't post what the actual solution was, but for me I got some new non-Corsair memory, (G SKill). Apparently corsair and that particular motherboard don't go well together. Also, I adjusted a BIOS setting about how the memory was used. The option was stability or performance. I put the setting (can't remember the name exactly...) all the way to stability. I'm crash free for 2 days now...pray for the best!

 

[Archean]

That is great news, I think it is always good idea to go through QVL of whichever Asus board you are planning to purchase. Goodluck and let us know if you need any further advice. Regards