System locks up / BSOD issues HELP!

Ive had this issue for some time now and simply cannot pin point it:

What happens is at random intervals when playing a video, or game the system will bsod with the error

the system has halted contact your hardware vendor

now more than likely I dont get a bsod just a freeze during a game/video being played. For example, play a music video and the picture freezes, audio freezes too.


System

mobo: asus p5n32-e sli
gfx - ati 4870x2 2gb with accelero cooler
ram - 4gb ddr 2 corsair 800mhz
cpu - qx6850
hard drive one - kingston 128gb ssd
hdd2 500gb
hdd3 500gb
hdd4 wd raptor 32gb
hdd5 hitachi 150gb
optical - dvd rom (disconected)
psu - 1000w
cooler - scythe ninja 2


gpu temp - 52 idle 58 load
cpu temp idle 33



steps taken


tested all 8gb ram - one stick had errors, removed
tried ram in different slots
tried os on different hdd
tried different gpu
tried in safe mode
tried different psu (only for one hour test)
scanned for virii
tried gpu in 8x pci-e slot



notes:


has been acting like this for months with vista x64 and windows 7 x64
small piece of plastic broken off ide controller, as a result optical drive wouldnt fit properly - caused lots of shorts - so disconnected completely. Problem still occurs.

This problem occured before ide controller was broken.

This mobo has had issues before, ordered replacement and worked fine (this is the second one of this particular model).

removed battery, reflashed bios, same issue.


fresh install of windows, same issue
checked all hdds for errors - all fine

anyone recommend anything to try?

 

You've done some really good things already, good troublshooting!

I looks like it could be an overheating issue, how old is the PC?

Please clean out any dust you find.

And could you get some load CPU readings please?

 

I wouldnt have thought its an overheated issue, simply because this problem has happened from a cold boot with the pc off for over 12 hours.

There's around 3 120mm fans inside the pc, the cpu is around 45d at most times.

What program would you like me to use to get temps for you?
No dust inside pc, well at least nothing major!

 

For temps you could use a program such as speedfan

 

From what you have said it appears very unclear as to whether the problem is hardware or software related.

Heat issues may be at the heart of the issue but from the temps you have provided this would seem less likely. To eliminate heat issues, take the side of the case off. Get a fan (room fan, free standing) and point it at the mobo and try again.

Personally, if I were in your shoes I would try to boot with a Linux Live CD (Puppy my choice, 100mb download and loads entirely in ram).

Once you get to the Linux desktop, you will be able to navigate to your data files and play your videos or/and music files (Puppy has players for these).

See if the system freezes or not. If it does then the issue is likely hardware related.

If the above tests turn out fine with Linux, then the problem would point to a software problem, in which case a bare bones set up would be the way to trouble shoot.

Good Luck

 

I think this is obviously hardware related... I think you forgot that he tried it on another HD.

But it would be interesting to see if it freezes in Linux too.

Could be drivers tho.. right? are you video card drivers and sound drivers up to date? I don't see that in your troubleshooting.


Actually i just realized you said you replaced the GPU.. you updated chipsets too??

 

temps are 31d idle 60 load

drivers up to date, only thing not updated was the nvidia board (nforce) mobo driver, however I did try updating this in the past and mine was already the current version.

 

That is quite high, wait till you try linux, then i would put money on it being an overheating issue.

 

Cpu load temp is high. Higher than the 45c which you mentioned earlier and which I understood to be the cpu load temp.

In my book this pushes heat as a possible cause of the freezes higher up the suspects list.

Try the fan it's quick and easy (in the interim, if you have no fan just take the side of the case off). If you still have issues having tried the fan, then try Linux and see how you get on.

Good Luck

 

when this issue first came around I turned on all thw fan son full, so thats 3 120mm fans on third setting, plus the gpu fan plus the cpu cooler - still happened.

I somewhat doubt its a heat issue, as whenever the problem happens I immediately touch the back of the pc and its not hot at all.
Saying that, the gpu is red hot (atu 4870x2 depite it showing as cool) and the ram is also very hot.

Im currently testing linux puppy, complete noob to it so have just run a video, got home after 8hours of work and found the linux video program crashed, couldnt play one of the videos - but the OS was still fully active. So am now playing a video it can recognise and see how long it'll last.
Will try the fan thing too if I can find one

 

I'm still thinking a heating issue, having 3x120mm fans does not mean that you are guaranteed to have a cool CPU. (I have 8, and still struggle to keep it cool)

Please download Prime 95, and SpeedFan. Prime 95 is a programme that pushes the CPU to the max, creating a load situation for a prolonged period of time, which rules in/out any OH issues.

Run Prime 95 and note the temp the CPU gets to.

 

do I run this within windows or linux, and for how long?

 

Up to you.

 

When stress testing the cpu, I usually run Prime95 (within windows) for 24hrs on the torture test (max heat). As Burrell said, run Speedfan to monitor cpu temps during this test.

To stress test the vga, run 3dmark (loop) and monitor the gpu temp with RivaTuner. RT can also monitor that the fan(s) on the video card are running at 100% when max heat is generated.

In the interim just take the side cover off. If the temps fall, it's generally an indication that case airflow is not efficient. Having more fans does not necessarily mean that cooling will be better simply because the airflow from the many fans may clash and/or create dead spots.

Good Luck

 

8AM-8PM running linux puppy, player did crash but OS still active , at 6pm ran 2hrs of music video, no issues.

8:10pm, exited puppy, booted vista x64, as soon as I get to the desktop, I right click in order to access a folder, I loose display immediately, have to reboot.
Reboot PC, get into vista x64, run prime95, about 20 secs in I thought I'd best run speedfan too to get temp readings so close down prime 95. The system hangs. Have to reboot.

8:15pm running speed fan, temps are 31d, start up prime 95, temps now 49d after 2mins.

9:03pm - still going strong, highest core at the moment is 67d, which ive gooled as alright really. I dont see that as super high. 80s yes!

10:39pm, 2hrs 22min test finished, no errors or warnings. Temps never got above 67d, average around 61 Id say, went down to 45's, 50s etc

I really doubt this could be the gpu, as Id have to be pretty unlucky to get 2 that fail.
Will try the GPU test tomorrow. Side panel was ON during this test btw, after test finished, PC was cool to the touch.

 

Hmmm i'll wait and see the GPU test before i say anything more.

 

I love how in windows one thing fails.. and it all goes down like dominoes. In linux you can tell something is wrong, but the whole system still works.. amazing... lol I had to say that.

 

Test 2

8am-8am ran video within vista x64 for 10hrs, no issues, still playing when I returned.
Just completed 3Dmark test, temperatures only rose by about 5d giving a load of around 55d total during the benchmark.
Only thing of note is how badly it performed, on my previous 9800 gx2 card, this same test ran fine with good fps, this one had about half the frame rate (about 12) and experienced alot of graphical glitches on the Jane demo, for example you would be in the middle of a texture at times. Demo ran sluggish.
Im surprised I havent had a crash yet, IM wondering if its because Ive moved the PC away from where it normally sits, ill get round to that later.

Advice?

 

Was your computer near a window in direct sunlight? not much is plausible by just moving the computer. That does not make sense either cuz you tested it at night..

 

The PC (now and then) is by a window, though curtain drawn

 

Lets sum up

You get what appear to be intermittent issues in windows and Linux, so it would appear to be some type of hardware issue/conflict.

When you stress tested the gpu with 3dmark, temps seemed reasonable. So I am assuming video card ok for now ie moved lower down in the suspects list.

When you stress tested the cpu with Prime95, the temps reached 67C but you only ran the test for some 2 hours which is not really any near enough.

At this stage, I personally only have 3 things to suggest (although I may think of others in a while)

1. The cpu is overheating and causing the prob.

2. I would try to remove the sound card (or replace with a spare if you have one) and see how the system responds ie any lock ups.

3. Bare bones, adding one component at a time and testing thoroughly after each addition. A pain but at least you will be in a better position to find the error.

Good Luck

PS. are the mobo's chipsets getting very hot..........

 

I cant see it being a CPU heat issue, as the PC is running now after 11hrs and still only at 31d which is damned good.

I did wonder about the soundcard so could give that a shot, I posted the chipset temps a few posts back.

Its certainly perplexing. No crashes as of yet though, 48hrs in.

 

Only thing that comes to my mind as a remote possibility is that along with moving the computer you now have it plugged into a different electrical outlet. Could there be a problem with the outlet it had been plugged into?

 

When I said I moved it I meant like out form its original position, so its around a foot from where it is, same electrical socket. Its been moved out to gain access to the side panel.

 

Maybe a card was slightly dislodged in the move and is now making poor contact, causing the computer to intermittently believe that a card has been removed, locking it up.

Try removing and reseating all cards.

 

Did the same test today, i.e played a video at 8am, all day while I'm out, the only difference to this test was that I moved the pc back to where it normally stands. I returned home today - PC is OFF. My partner said it froze around 2pm, but could have been sooner.

So, its been working for 48hrs fine, no crashes, move it back, crashes. Now obvious thing is somethings loose, and I think it could be the GPU, which isnt currently screwed down - now dont go DOH at the same time, as this issue has happened when screwed down and not.

I tend to think the gpu card is so heavy its slightly pulling out of the PCi-e slot.

Perhaps the "jossle" of moving it back is dislodging it a little, it doesnt look as if anythings amiss though...

 

DOH

I am sure that is the fault, and screwing it in tightly should resolve it, especially if you make sure the clip on the PCI-E slot is engaged properly.

 

I did originally screw this, and the older 9800 card in, same problem happened though, the small blue catch lever is broken off the PCI-e slot though.

Ill try screwing the card in tonight and test again in the morning.

 

GPU tightended, checked ram, connectors etc, all seem to be ok. Turned on PC, initially didnt get a display, rebooted again and it came up.

Testing again.

 

PC was put back to original position last night, all components checked to see if they're secure, GPU was screwed down tight.
Left PC on overnight.


This morning at 8:20am - crashed, dont know how long.

What the HELL is this issue!!

 

Next test - move the pc halfway from its standard location to where it worked.
Clutching at straws or what.

 

Try leaving the PC on it's side and leaving it on. This will use gravity to hold the cards in place.

 

Test results after moving PC halfway - crashed after 6.5hrs

 

Have now moved PC onto side, in its original location, testing again

 

Test results - PC moved back to position where it previous worked for 48hrs - CRASHED!!!!

What the hell guys?

Am now taking this pc upstairs trying elsewhere, same test, any other suggestions from the info posted?

 

So did it work on it's side?

 

no crashed

 

Further update;

PC moved upstairs, changed MONITOR, changed KEYBOARD, changed MOUSE.
Played for around 3.5hrs, and then..froze again.

Now testing with above config, minus all HDD's except the boot HDD
From now on Im going to remove a certain item of hardware until only the barebones is left.

I was thinking this afternoon though about how it only seems to affect videos and games, leave the pc on doing nothing and it'll remain unfroze.

All signs point to the GPU, yet it has been testing in 2 different PCI-e slots, and with 2 different cards.

Should I try a third card?

 

Might be a motherboard component getting heat stressed and loosing electrical connection.

 

how can you test every conponent though?

Easiest would be get a different mobo to try

 

This would be the only fix if it were the case, there are no testing methods.

 

So what now? A problem that cannot be fixed by the Geeks? Surely not? :cool

 

The fix is to get a new motherboard. Sorry.

 

Update on testing, last night removed all HDDs except boot HDD - this morning after around 9hrs.. crashed.

Am on my penultimate test today.

Have removed one stick of ram, changed other ram to different slot, removed optical drive, removed soundcard, removed firewire and USB cables.

If this doesnt work, last test is barebones with new PSU and possibly GPU

 

and after the barebones test, yup - crashed.

AM now on barebones, whole pc outside of case, with different psu, even cut my hand on the frigginly awkward cpy cooler scythe ninja 2

 

If you removed the CPU cooler you will need to replace the thermal paste between it and the CPU otherwise you will reduce the efficiency of heat transfer between the two.

 

havent removed the cooler, this is the last test Im going to do, an RMA with asus is mind numbingly awkward, so I'll most likely just buy a new PC, or something

 

new psu, barebones, still crashes.

Time for a new mobo

 

http://www.play.com/PC/PCs/4-/11565895/Gigabyte-GA-G41M-ES2H-Socket-775-Intel-G41-Micro-ATX-Motherboard/Product.html?ptsl=1&ob=Price&fb=0


would this mobo be ok for my 4870x2 card and qx6850 cpu.
Its micro ztx so take it it will be smaller than the asus p5n32-e
or should I just get another cheap full size board?