Ok, I'm stumped. I have no idea what is wrong with my computer

[slightlyhuman]

Ok, I'm going to make this as short and painless as possible because I'm going crazy. A few weeks ago my computer started having random reboots and crashes. I thought my memory was bad. 9/10 it would happen in 3D applications(games) and would require a hard reset. It's done it once while in windows.


Specs:

DFI lan party nf3 250gb
a64 3200
2x512 3200 corsair value
new 1x1gb 3200 PNY
160GB WD HD
250GB SG HD
sp2
windows xp pro
x800xl

I went and bought a new stick of PNY 3200 DDr ram 1GB in size and put it in the last of the 3 memory slots and tried to run some games. played fine for an hour and boom the problem came back.

I've also tried just using the brand new 1GB stick and get the same outcome. I've tried only with the master HD connected and still, same thing.
I uninstalled the 6.7 ATI drivers and installed the latest omega and still, same problem.

Sometimes I can play games fine, it's never crashed on quake 4 but steam, FEAR and BF2, crashes.

I'm stumped and am about to lose my gord. I think it might be a windows problem/video card/ or mobo problem but I have no way to test because I don't have any spares.


Please help me and let me know if you need anymore info:

 

[Smitty308]

I would suspect the power supply.

Had a similiar problem with a computer recently. I have replaced the PSU and it seems to be ok so far.

Good Luck
Smitty

 

[slightlyhuman]

Originally posted by: Smitty308
I would suspect the power supply.

Had a similiar problem with a computer recently. I have replaced the PSU and it seems to be ok so far.

Good Luck
Smitty


hmm, it's a seasonic 380w. i wouldn't think of psu of this quality would do that. maybe i can borrow one from a friend to see.

 

[MrUniq]

You'd have better luck swapping out the video card and see if that's the problem. Don't forget to check event viewer also. I doubt it's the PSU too.... i think to many people blame PSU's when they are often the sturdiest pieces of hardware on a PC.

 

[SecondNova]

Sounds like your entire system is overheating, and that's what is causing those hard-reboots. It makes sense, since your system crashes while under heavy stress (BF2, FEAR, etc). Perhaps add a fan or two inside and get that cool air flowing?

 

[FP]

Installed SpeedFan and check your PSU voltages and temps while playing your games. Should give you good insight.

If that comes up dry then I would suspect your video card is crapping out.

You could try running Prime95 in torture mode to see if it is the CPU.

 

[yyrkoon]

speedfan or any voltage monitoring program is subject to error, and are often inaccurate.
380W does seem low for your system.

However, when was the last time you blew the dust off your systems CPU/GPU heatsinks ? Made any system changes recently ? such as OC'n the vid card etc ?

 

[oynaz]

I fixed a friends PC with exactly the same problem yesterday. Turned out the PSU was bad.

 

[Noubourne]

Sounds like you've got a lot more hairs to pull out. You've only checked a couple things so far. It could be anything.

I recommend changing only ONE thing at a time to test stuff so you KNOW for sure when you've found the culprit. If you start changing drivers and hardware etc etc all at once, you'll never know when you've found the exact problem.

Do one thing at a time. I'd start by checking temps, because if it happens after a certain period of time of gaming, that would be the first thing I would suspect. Check your temps at idle, and then check them after stressing it. Take the side off the case and see if that makes a big difference in temps. If it does, then you have poor cooling and you need to do something about it.

 

[slightlyhuman]

my pc stays clean, inside and out. i have good cooling. here's what happened last night. i tried playing quake 4(it's never crashed while playing this for some reason) and it crashed. i was almost done loading the map in MP and it went to the full screen console then crashed. well i held the power button for 5 seconds and then turned it back on. as soon as i got back into windows, crash. hard reset again and when i got into windows the second time, crash. so i finally got back into windows and went to quake 4 again. almost loaded the map and then went to full screen console then crashed.

what the fvck. i haven't overclocked anything. i don't have any extra parts in terms of psu's, mobo's or vid cards laying aroud to see if that's it


i don't see how it could be a heat issue. i've had this setup(minus the new stick of 1gb 3200) for over a year and NO problems whatsoever. nothing's changed so i have no idea why it would just act like this.

 

[kungfoo]

Originally posted by: Noubourne
Sounds like you've got a lot more hairs to pull out. You've only checked a couple things so far. It could be anything.

I recommend changing only ONE thing at a time to test stuff so you KNOW for sure when you've found the culprit. If you start changing drivers and hardware etc etc all at once, you'll never know when you've found the exact problem.

Do one thing at a time. I'd start by checking temps, because if it happens after a certain period of time of gaming, that would be the first thing I would suspect. Check your temps at idle, and then check them after stressing it. Take the side off the case and see if that makes a big difference in temps. If it does, then you have poor cooling and you need to do something about it.

I agree with this poster. I had a problem VERY similar to yours. Do the ATI utilities come with something that tells you your GPU temp? (im an nvidia user, and they have one) Regaurdless what you'll want to do is find someway to check your GPU temp. Start your game of choice, play for a little while and then check your GPU temp. I checked every part of my PC before realising I had a crap card and RMA'd it. Replacement card has worked perfect since. Also pay attention to Case and CPU temps too.

Generally if your GPU is overheating it wont cause a total system lock up, but it can. Generally with GPU's overheating you see alot of artifacting and screen-blanking.

So, those are some things to keep in mind and try.

 

[slightlyhuman]

k, i've been running atitool's artifact scanner and it freezes everytime i run it between 30 seconds and 1 minute.

maybe it's my video card?

i was thinking of running prime95 but would my video card interfere or does prime only focus on the cpu?

 

[kungfoo]

Originally posted by: DP
k, i've been running atitool's artifact scanner and it freezes everytime i run it between 30 seconds and 1 minute.

maybe it's my video card?

i was thinking of running prime95 but would my video card interfere or does prime only focus on the cpu?

Leaning toward video. You dont have any way to get your hands on another card to test?

 

[slightlyhuman]

Originally posted by: kungfoo

Originally posted by: DP
k, i've been running atitool's artifact scanner and it freezes everytime i run it between 30 seconds and 1 minute.

maybe it's my video card?

i was thinking of running prime95 but would my video card interfere or does prime only focus on the cpu?

Leaning toward video. You dont have any way to get your hands on another card to test?

just tried torture test using prime95, it reset after about 5 minutes.

i could go to best buy and get a video card to see and just take it back afterwards but now that p95 crashed it could be anything.

gdi

 

[kungfoo]

Originally posted by: DP

Originally posted by: kungfoo

Originally posted by: DP
k, i've been running atitool's artifact scanner and it freezes everytime i run it between 30 seconds and 1 minute.

maybe it's my video card?

i was thinking of running prime95 but would my video card interfere or does prime only focus on the cpu?

Leaning toward video. You dont have any way to get your hands on another card to test?

just tried torture test using prime95, it reset after about 5 minutes.

i could go to best buy and get a video card to see and just take it back afterwards but now that p95 crashed it could be anything.

gdi

Yeah honestly I would do that. Find the cheapest card they have there and just hang on to the receipt. Pop the card in and run a few tests. You should be able to tell really quick if it was your card, and you can just return the card you bought. Atleast you'll be sure its not video/is video at that point.

 

[Parasitic]

Maybe underclock your video card to a lower frequency?
Sometimes it may simply be th fan crapping out on your video card - in that case unless you hate Best Buy very much don't go out and buy a video card for testing purposes.

Downclock your video card below stock speeds, then run artifact tester to see if it still freezes. If it doesn't, then leave it at the same frequency and take off the side panel to see if the GPU fan is functioning.

 

[Modular]

I'm betting on PSU even now. Especially after the hard-reboot whcih started to load windows and then hard-booted by itself. Pretty indicative of a PSU problem. Time to bust out the volt meter.

 

[bwnv]

I would run memtest86 first, if you get any errors bump up the voltage a bit. A vid card error hardly ever causes this type of problem, though it's a possibility. Also, if you have a multi-meter you could check your voltages on the power supply. Sometimes lower voltages on one of the rails can reak havok on a system.

 

[exdeath]

Originally posted by: bwnv
I would run memtest86 first, if you get any errors bump up the voltage a bit. A vid card error hardly ever causes this type of problem, though it's a possibility. Also, if you have a multi-meter you could check your voltages on the power supply. Sometimes lower voltages on one of the rails can reak havok on a system.

Just be careful, voltage testing alone isn't always useful without some kind of current load testing as well. I had a machine that wouldn't power up at all (dead PSU suspected immediately). The standby light was on and all voltages were correct (tested all pins), but under load (hitting the power button) the machine wouldn't even try to power on. The problem was remedied with a spare power supply, confirming that it was a bad power supply, even though the voltages were all fine.

 

[exdeath]

Originally posted by: Modular
I'm betting on PSU even now. Especially after the hard-reboot whcih started to load windows and then hard-booted by itself. Pretty indicative of a PSU problem. Time to bust out the volt meter.

Hmm when you have hard restarts like that, especially when they become more frequent after the machine is first powered on, usually 99% it?s caused by CPU overheating (fan failure or improper heat sink seating usually the cause).

Makes sense, takes a few minutes to warm up, after which it starts spontaneously rebooting, often rebooting even in the middle of a reboot. Come back 5 minutes later and you manage to get to the desktop for a short while before it starts again.

 

[Denithor]

In my experience I have usually seen this type of problem either from the power supply or the video card. I have seen it recently with a brand new x1900gt that appeared to work ok until you got into a 3D game, at which point it would crap out, either freezing solid, dropping to windows, or even rebooting the system completely (nothing more fun that being in the middle of a firefight and suddenly you're seeing your BIOS screen). Unfortunately there isn't a definitive method to test either of these without simply switching for a different part. Your best bet is probably to buy or borrow a power supply and/or a video card at BB/CC/CompUSA to test with and return when done.

I would probably start with the video card as the problems seem to occur more in games than other applications.

Good luck!

 

[GuitarDaddy]

I'm betting on PS as the problem

 

[knght990]

Does your x800 have its own place for a power connector? (i thinkt hey all do now don't they)? If so, is it hooked up? (sorry i have to ask the stupid questions) You didn't mention that you checked the windows error logs. If you did were there errors? The error logs are located by right clicking on my ocmputer, then manage, then open the event viewer list and click through each of the sub menus and look for red x's and yellow exclmations.
I have had similar problems with a bad install of windows and a bad install of the video drivers (which you usually find errors). Also with an over heat of the nforce chipset (no errors but the chipset's ehatsink is usually very hot to the touch).
Maybe some of this will help.
-TL

 

[Bozo Galora]

I go with temps (maybe MB), then PSU, then Nforce drivers (especially RAID SATA IDE), then vidcard,

 

[slightlyhuman]

Originally posted by: exdeath

Originally posted by: Modular
I'm betting on PSU even now. Especially after the hard-reboot whcih started to load windows and then hard-booted by itself. Pretty indicative of a PSU problem. Time to bust out the volt meter.

Hmm when you have hard restarts like that, especially when they become more frequent after the machine is first powered on, usually 99% it?s caused by CPU overheating (fan failure or improper heat sink seating usually the cause).

Makes sense, takes a few minutes to warm up, after which it starts spontaneously rebooting, often rebooting even in the middle of a reboot. Come back 5 minutes later and you manage to get to the desktop for a short while before it starts again.

it will go all day in windows with winamp, limewire, firefox and trillian open and run without a hitch.

i'll get a psu and vid card next week and try.