Constant Crashing, Semi New Build, Win 7, Intel

Associate
Joined
30 Oct 2004
Posts
139
Location
Newport
Heylo.

Would appreciate some help if at all possible as im at my wits end.

Built a new PC, Christmas pressie for a family member. Due to her old one giving lowframe rate on game she played ( LOTRO ) and also on facebook Bejewelled slowed down

Her old system.

AMD FX57,
2gig OCZ Plat
Asus Sli-Premium
9800GT
Win Xp Sp3

Shouldnt of given the problems it did.

Now Built and running on.

Intel Wolfdale E6300
Asrock N7AD-SLI
4gig (2x2) Crucial Pc2-6400
9800GT
Win 7 x64

Figured replacing half the PC should give some form of improvement,
Ill also point out that Win 7 was a latest attempt to fix this, Problem was still on new hardware on XP 32x

No matter what shes playing be it Lotro or WoW, or Flashgames, it seems to crash pretty heavily. The only clue i have is in event viewer it states.

Event id 6: Kernel-Processor-Power

Some processor performance power management features have been disabled due to a known firmware problem. Check with the computer manufacturer for updated firmware.

I have updated the drivers all the way, Motherboard and Vid card etc, Updated bios, etc.
I did read some reference above error relating to power saving features in bios and windows so have turned as much of that off as possible. but still crashing.

Scanned the harddrives with Victoria and found 1 or 2 bad sectors on one.
Ran P95 on torture for an hour or so without problems.

Found a program to look at the minidumps, reports as >

Bug Check Code : 0x00000124
Caused By: ntoskrnl.exe


anyone got any other ideas or anything that can help, let me know about anymore info thats needed. Thanks
 
PSU is : Thermaltake ToughPower 750W Modular Power Supply

ill run memtest overnight see what that brings up :)
 
I would normally guess PSU here as well seeing that it crashes under load.

Disable "automatic recovery" on bluescreens in order to see if it is windows crashing or your pc simply rebooting.
 
Surely the PSU wouldn't cause problems with flash games tho?
Ill zip up the minidumps if they'll be any use?

ty
 
Surely the PSU wouldn't cause problems with flash games tho?
Ill zip up the minidumps if they'll be any use?

ty

While flash games usually are not taxing the GPU they can still use loads of cpu. Check your browsers CPU util in task manager will surprise you... Some of the more well known FB games are horrible.

Memtest like suggested is always a good idea.

It's often enough to simply check the actual bluescreen to see the error raised. If you see for example references to NV4disp.sys it is likely GPU related.
 
Memtest been running for over 8 hours now, gunna leave it on till morning, 12am atm.

Used Bluescreenviewer to check the bluescreen info references ntoskrnl.exe as the problem driver.
 
Mem test ran for just over 18 hours with no problems. Still no wiser to whats going on :(

The only thing i did check was the compatability charts for the mobo, Ram and CPU aint listed on it directly, But that shouldnt be a problem really?

/edit 2

This also in the event log:

Event ID 18 WHEA logger

A fatal hardware error has occurred.

Reported by component: Processor Core
Error Source: Machine Check Exception
Error Type: Bus/Interconnect Error
Processor ID: 0

The details view of this entry contains further information.
 
Last edited:
Possible faulty gfx card? Might explain the low framerates, also the fact under stress tests its crashing within 5 mins, And then the event log shows the above error.

But to me that wouldnt explain it crashing on flash games?
 
Google for "prime95" from mersenne research, download it and run a stress test for an hour.

If your mem test ran fine thats good. prime95 will task your cpu very hard. If that runs fine it goes onto the gpu stress testing.

However ntoskernel is the very core of windows. So that sadly does not reveal to much.
 
Prime was one of the 1st things i did run, was on for around an hour or so, I know i should probably leave it longer, but the systems been crashing that fast, i figured that the problem would arise sooner.

but no :(

Gunna swap out the graphics cards when i get back. See if different ones make a difference
 
No, But with the lowframe rate persisting though a semi rebuild. I thought it would be more hardware based.

"The NVIDIA Display Driver Service service has reported an invalid current state 32."

is something else thats cropping up in the eventlog, Googled it a bit, and installed some new beta drivers to see if that helps as per some reading ive done.
 
Lastest update for reference sake. :)

Thrown in one of my 8800GTXs,

Also plugged the molex into the motherboard, the one thats for sli / xfire. To see if that helps at all. Mobo said to plug it in if using more than one card, but who knows

Stress testing using same program and its been running for over 30mins now, whereas the other crashes in under 5.

will be testing previous card soonish to see if not plugging that molex in was the problem >.<
 
Power distribution might be weak point in that Nvidia mobo if plugging that extra power connector helps...



You could try this disabling of one Nvidia's service:
http://forums.nvidia.com/index.php?showtopic=105792&view=findpost&p=601638

And what Nvidia driver you're using?
(191.07 has bugs which can damage/kill Nvidia hardware)

prime95 will task your cpu very hard.
Only small FFT mode stresses CPU hard... and even that can't match Linpack.
But Prime's blended mode is good for overall stability testing when run for long enough. (at least overnight)
 
Driver is the 195.81 beta driver. Read some people have had luck with it stopping the state 32 error.
 
Last edited:
To tie up, My 8800 Ran till i turned it off on her games, Not worlds greatest framerate but beggers > choosers.

Reinstalled her 9800Gt, started stress testing it, Crashed in 6mins. Gunna see about sending it back if it fails in my machine tomorrow.

Thanks for the help as always :)
 
Back
Top Bottom