• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Crashing in games :(

Craig321

Craig321

Craig321

Soldato
Joined
2 May 2004
Posts
19,950
At first I thought the crashing was because of my 8800GTX overheating, so I set up RivaTuner and adjusted it so the fan spun up to 100% when the graphic card got to a certain temperature. The temperatures are roughly the following when I've been in a game for a while:

GPU: 65
GPU Diode: 70
GPU Ambient: 50
GPU Memory: 65

As far as I know these temperatures are fine for a card under load?

I was in Oblivion for a couple of hours earlier and my PC decided it'd like to restart itself :(

I've got "Restart automatically" on blue screen unticked, so it definitely wasn't a blue screen. It was a plain reset.

I am able to run Orthos overnight with absolutely no problems so it's not power supply/memory/something else, however after a few hours in a game my PC will crash.

Is there anything else I can try to make sure it is the graphic card crashing? I don't want to RMA it if it's not faulty.

[Edit]
Gigabyte P35-DS4
Intel Core 2 Duo Conroe E6600 @ stock
BFG 8800GTX @ stock
Geil 6400 4-4-4-12 2.1v @ stock

CPU rarely sees much above 40°C
Motherboard temperature (which I'm guessing is the northbridge?) stays at about 50°C
Drives are perfectly cool - ~30°C

Other than the random crashes/resets when playing games the PC is perfectly stable. It never crashes under high memory/CPU load.

Thanks,
Craig.
 
Last edited:
Craig321 said:
At first I thought the crashing was because of my 8800GTX overheating, so I set up RivaTuner and adjusted it so the fan spun up to 100% when the graphic card got to a certain temperature. The temperatures are roughly the following when I've been in a game for a while:

GPU: 65
GPU Diode: 70
GPU Ambient: 50
GPU Memory: 65

As far as I know these temperatures are fine for a card under load?

I was in Oblivion for a couple of hours earlier and my PC decided it'd like to restart itself :(

I've got "Restart automatically" on blue screen unticked, so it definitely wasn't a blue screen. It was a plain reset.

I am able to run Orthos overnight with absolutely no problems so it's not power supply/memory/something else, however after a few hours in a game my PC will crash.

Is there anything else I can try to make sure it is the graphic card crashing? I don't want to RMA it if it's not faulty.

Thanks,
Craig.
Click to expand...

Similar probs as with my card, it appeared to be my XP installation borked, in vista it's 100% stable, at much higher temps too I might add ( 86 C idle)...
Although mine mostly just BSOD'D blaming the nvidia drivers ( even after drivercleaner pro & fresh drivers a few hundred times)
Try it on a clean windows install ( be it another xp if you like)
 
snowdog said:
Similar probs as with my card, it appeared to be my XP installation borked, in vista it's 100% stable, at much higher temps too I might add ( 86 C idle)...
Although mine mostly just BSOD'D blaming the nvidia drivers ( even after drivercleaner pro & fresh drivers a few hundred times)
Try it on a clean windows install ( be it another xp if you like)
Click to expand...

Running Vista x64 here.

I've re-installed recently to try fix the problem :(

Gonna run Furmark and maybe some other games for a few hours again to make sure it is the graphic card.

This card has crashed in:
Titan Quest
Trackmania Nations
Trackmania Nations Forever
Oblivion
Dawn of Magic (I think, can't quite remember)
Battlefield 2

I think I've ruled out the possibility of the game plain not liking the hardware/software. I can't see my drivers going wrong twice in a row, even after a complete Vista reinstall.

I'm going to run Furmark stability test for a few hours, if it crashes then I'll do a driver sweep and run Furmark again. If it still crashes I'll RMA :(

Then I get to use a P4 3GHZ XP machine with a 7600GT for a while... yay! :/
 
Last edited:
Is there anything in C:\Windows\Minidump? That's where dump files are stored, so if there are any new dump files we could inspect them to find out what caused the machine to restart.

The only thing I can think of that could forcefully make the machine restart without causing a BSOD would be lack of power, and the machine cuts out. Running Orthos overnight isn't a viable test since it's only testing the CPU and power to that on full load is no where near the power consumed by a graphics card playing Oblivion, let alone the added CPU usage from Oblivion.
 
Nope, nothing in the Minidump folder.

I've got the Corsair HX 620W here, surely if it was faulty it'd be crashing more often and not only in games?

I still suspect the graphic card though as most of the time rather than resets the PC will freeze, the screen goes into standby, then after a while it'll usually BSOD with nvlddmkm.sys
 
Last edited:
Craig321 said:
Nope, nothing in the Minidump folder.

I've got the Corsair HX 620W here, surely if it was faulty it'd be crashing more often and not only in games?

I still suspect the graphic card though as most of the time rather than resets I get the nvlddmkm.sys BSOD.
Click to expand...
Yeah I'd be looking at the graphics card then if you are getting BSOD's as well (sorry I assume that all your restarts were automatic and you never got a BSOD). If it's BSOD'd though, you should have something in C:\Windows\Minidump, unless you've formatted as mentioned and you got the BSOD's before that then there would be nothing in there.

I'd definitely isolate the issue to either the power supply being unable to cope with heavy load for extended periods, or a graphics issue (card or driver), or even both. Drivers wouldn't be able to just make the computer restart however without causing a blue screen or some sort. Power failing, and graphics card failing would cause the machine to restart though without any form of a BSOD.

Being on Vista you normally get a box opening up telling you why the computer was restarted when you next boot up, have you had any of those?
 
Yep, I've had them and I've analysed the Minidumps as well. Not been able to get much useful information out of them though apart from the fact that it thinks nvlddmkm.sys caused the crash.

The Vista box telling my why the computer BSOD'd just says it was something to do with the NVIDIA Corporation drivers.

If the computer crashes with Furmark again I will run Driver Sweeper and reinstall the drivers, but I seriously doubt it's the drivers considering I've reinstalled Vista.
 
Don't suppose you have another machine you can test the graphics card in? Or maybe a friends machine and let him full load the graphics card on his machine for X hours and seeing if it restarts?

That would at least remove the graphics card and leave you with drivers or PSU.
 
I just ran a Furmark for 2 hours. Surely if there was a problem with the power supply not being able to handle the load it would have crashed after half an hour or so? I ran a Furmark stability test with 4x MSAA.

I guess I should be running Furmark for longer?
 
Craig321 said:
It'd crash under Orthos/when running other intensive programs/processes if it was RAM. The PC only ever crashes in games.
Click to expand...

It could be a mutlitude of things. i set up my pc for music studio (cubase e.t.c.) use with the 3gb switch and a midi interface e.t.c. If i tried playing games using that configuration everything would crash,bsod,freeze, all sorts of problems. Switch back to normal and everything is perfect.

What i am getting at is that it could well be a conflicting bit of hardware or software? Try just running games using the barebones of the pc.
 
I cannot see where you have mentioned which driver versions you are having the issue with??

163.75 WHQL's are probably the most stable nvidia 8 series drivers I have used - late 2007. The more current versions seem to cause me little problems now and again with my 8800 ultra, including the odd BSOD. However, no problems what so ever with 163.75 for me.

Could give them a try? I would recommend uninstall your nvidia drivers, reinstalling the drivers that came on cd with your card, then sticking the 163.75s over them.
 
Last edited:
Craig321 said:
I am able to run Orthos overnight with absolutely no problems so it's not power supply/memory/something else, however after a few hours in a game my PC will crash.
Click to expand...

Power drain while gaming could well be higher than when running Orthos, as the 8800GTX can draw quite a lot of power under load.
 
eggnoodle said:
I cannot see where you have mentioned which driver versions you are having the issue with??

163.75 WHQL's are probably the most stable nvidia 8 series drivers I have used - late 2007. The more current versions seem to cause me little problems now and again with my 8800 ultra, including the odd BSOD. However, no problems what so ever with 163.75 for me.

Could give them a try? I would recommend uninstall your nvidia drivers, reinstalling the drivers that came on cd with your card, then sticking the 163.75s over them.
Click to expand...

I've reinstalled Vista trying to fix this problem, I've tried both 175.16 and the other older ones (December 07 IIRC). I really doubt it's a driver problem, especially as my PC full on reset itself in Oblivion :(

Just ran a 3 hour Orthos while I was watching TV & everything in that area is 100% solid.

I guess I should really run a longer Furmark.

HangTime said:
Power drain while gaming could well be higher than when running Orthos, as the 8800GTX can draw quite a lot of power under load.
Click to expand...

I haven't ruled PSU out yet, when I said what you quoted above I wasn't thinking about how much power the 8800GTX can take.

I think I'll run an overnight Furmark with 4x MSAA. (My poor poor GTX :()

But meh I only feel sorry for it when I stress it because it's fan spins up :p
Drama queen.
 
Last edited:
Started Furmark stability test at 3am last night when I went to bed, according to Vista the PC was unexpectedly shut down at 03:40:09 (so about 40 minutes into the test).

There's no minidump and no report of a blue screen, so the PC must have reset itself again.

I've been running the same Furmark test on my brothers PC for the past 4 hours with my graphic card in it. It's been solid.

How likely is it that it's my PSU giving up considering it's resetting rather than BSODing?

Thanks,
Craig.
 
Very likely, but have you looked into anything else, eg. the cpu not overheating ? Or the north bridge, the mainboard auto reboots if some stuff gets too hot ? If it's not either of those then I'd say it's very likely it's your psu, try your bro's PSU the next night in your own pc and see how that works out.
 
You must log in or register to reply here.
Back
Top Bottom