Frequent crashes - Help requested

Treebiter · 25 Jan 2009 at 17:23

Ever since building my rig 6 months ago I've been having issues with it. It crashes regularly in different situations and just 'hangs', screen frozen, sound going crazy. This occurs primarily when on the net as certain websites cause this hanging without fail. It seems to be connected to flash/activeX components on these sites.

Some games also cause it to hang. Supreme Commander and Call of Duty WAW both hang regulary and others like World in Conflict and L4D do it very occaisionally. Strangely, equally intensive games like Fallout 3 and Farcry 2 never cause this hanging and play very well on max settings.

I haven't OC'd this rig, I'm running the latest drivers for everything and I'm running Win XP 32 home. The comp's running cool and is otherwise very stable. Anyone have any idea what's going on?

Thanks

Specs:

Intel Q9300 quad
Asus P5E3 mobo
OCZ 2GB DDR3 PC3-10666C7 1333MHz Platinum (2x1GB)
ATI 3870HD X2 1024mb
Creative Sound Blaster X-Fi Xtreme Gamer 7.1 Sound Card
Samsung SpinPoint F1 750GB SATA-II 32MB Cache
Corsair HX 620W ATX2.2 Modular SLI Compliant PSU

A2Z · 25 Jan 2009 at 17:28

Have you tried taking 1 of the RAM sticks out and seeing if it happens? And swapping them.

J.B · 25 Jan 2009 at 17:35

A2Z said:
Have you tried taking 1 of the RAM sticks out and seeing if it happens? And swapping them.

Or run a memtest.

normally Id say a driver problem but if you deff have the latest drivers like you say then Im not sure.

What kind of temps are you getting on the GPU?

Treebiter · 25 Jan 2009 at 17:44

I've taken the sticks out and they both work fine solo. I've also run mem test and all ram checks over 100%. Temp-wise, Pretty frosty considering. 45-50 idle on the GPU, 70ish max under full load. The CPU idles at 45 and seldom tops 55-60.

lunarwolf · 25 Jan 2009 at 17:50

temps seem fine, although CPU a bit hotter than i would have thought but nothing to create crashes

i have read many posts with asus boards and ocz ram issues, not sure this necessarily applies but it may be worth investigating

A2Z · 25 Jan 2009 at 18:07

Treebiter said:
I've taken the sticks out and they both work fine solo.

No...does the same problem happen with just 1 of the sticks in...

jambus · 25 Jan 2009 at 18:40

Sounds like what was happening to me, I had the wrong screws in the motherboard, overvolting everything. Have you checked you've used the right ones? Very unlikely but just a thought. If not and if it's possible, try switching out certain components, such as trying a different graphics card, sound card etc. I know it's a bummer but it should narrow the problem down. I would've thought it's your graphics card if things are just "hanging". When did you buy the 3870x2 and is it still under warranty? Only other thing I can think it might be is the Sound card messing you around. Have you tried running Orthos or Prime95? How long did you run memtest for?

Treebiter · 25 Jan 2009 at 18:52

Hmmm... Appears that I can't get it to crash with the second stick removed. Unless it's just sods law, which I doubt considering the frequency of the crashes, I reckon we've found the culprit. I've heard OCZ isn't best paired with ASUS mobo's so does this have anything to do with it or we thinking faulty RAM? Like I said, memtest came out clear for what it's worth.

Certain the sound card isn't related as this was added after the problem reared it's head. The GPU seems stable and I've tested it extensively under stress. Mobo is also screwed in correctly as far as I can tell and the crashes appeared too uniform for that to be the problem.

Treebiter · 25 Jan 2009 at 19:05

Ok, scratch that last message. It did it again with one stick... Next!

simbady · 25 Jan 2009 at 19:08

Well if you could also try running Prime .. http://www.techpowerup.com/downloads/385/ .. you can use that to test your RAM and CPU if you want to in Windows usally used to see if a overclock is stable but will also check if you RAM is broken. If not it could be your Hard Drive?

Treebiter · 25 Jan 2009 at 19:53

Simbady - Ok, downloaded that program. How long should I let it run for?

simbady · 25 Jan 2009 at 19:55

I would say about a hour if it doesnt become unstable then its not your RAM only stress RAM for this you can do other things at the same time but your PC will be a little but slower.

Treebiter · 25 Jan 2009 at 21:04

Hmmm. Test crashed but ran for about an hour/40 mins before doing so.

jambus · 25 Jan 2009 at 21:56

either your cpu or ram then, you say memtest was fine, it's the cpu

Treebiter · 25 Jan 2009 at 22:22

CPU check came back fine (stable for an hour). Mem test crashed whilst testing memory in Go-Orthos on both sticks. Weird as mem test ran fine.

On another note, do you think it could be a power problem?

Treebiter · 25 Jan 2009 at 22:46

Tried testing ram again in Go-Orthos and this time instead of locking-up it gave this error message:

Type: Large, in-place FFTs - stress some RAM Min: 128 Max: 1024 InPlace: Yes Mem: 8 Time: 15
CPU: 2504MHz FSB: 333MHz [333MHz x 7.5 est.]
25/01/2009 22:41
Launching 2 threads...
Using CPU #0
Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Torture Test ran 1 minutes 13 seconds - 1 errors, 0 warnings.
Execution halted.

Using CPU #1
Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
Torture Test ran 1 minutes 13 seconds - 0 errors, 0 warnings.
Execution halted.

This was a 'Large, in-place FFTs - stress some RAM ' test, so is this a RAM error or a CPU one?