Author Topic: PC reebots in mid game for no apparent reson. (Read 1547 times)

StarOfAfrica2 · « **Reply #15 on:** July 26, 2005, 04:26:28 PM »

Heat. Pure and simple. 60c (140F) = dead for your AMD CPU. If the CPU itself doesnt fry, you'll be popping motherboard components like Jiffy-pop if you spend too much time at those temps.

My post isnt entirely clear. AMD I think rates the CPUs up to 85c or 90c. But they also say in the papers I downloaded from their website that normal temperature ranges should never be above 60c. Damage to motherboard components can occur above that temp. If you are seeing 55c AFTER the reboot, you could have hit 60 and the motherboard reset the computer.

gnubee · « **Reply #16 on:** July 26, 2005, 06:22:38 PM »

Hey Stone-
I had the same problem a few weeks ago...

Only for me, it was a bad driver. It turned out that the drivers I was using for my Medusa 5.1 USB headset were not on good terms with AH2. I discovered this because of the error report and it just so happened that my computer re-start would happen when I was talking on my mic.

(To be clear) I would be fly'n, press T or f12 to talk to someone and my computer would turn off completly, and then restart. It didn't happen all the time, and appeared fairly randome so it took me a while to figure it out.

For me the fix was simple, I just un-installed the Medusa drivers and let XP recognize the divice instead. Not sure if this is what you're experiencing, but I thought I'd share just in case.

Good luck bud-

-scott

AmRaaM · « **Reply #17 on:** July 26, 2005, 09:26:17 PM »

its your power supply.

bloom25 · « **Reply #18 on:** July 27, 2005, 12:37:10 AM »

If this is Windows XP it will reboot rather than show a bluescreen by default. I'd go to control panel -> system -> startup and recovery and uncheck the option to reboot automatically after an error. This will now let your system display a bluescreen rather than just rebooting instantly. That might help you find out if this is a driver problem.

Stone · « **Reply #19 on:** July 27, 2005, 12:51:19 AM »

Memtest has run all night, and no errors are found.

I do the other tests.

I also take try to get the case heat down a bit. I think I have the case fans not on full speed.

I also set the Bluescreen thing on, as bloom25 says.

I hope its not the powersuply. Its brand new. My PC has now been runing the memtest for aprox. 12 hours without problems, so I think the power suply is ok, and the memory at this temperature.

I guess the case temp does not get to hot during the memtest.

Thanks all for reply. I will continue with the tests now.

Roscoroo · « **Reply #20 on:** July 27, 2005, 01:01:14 AM »

dont forget it could be a virous too .

as for an amd @54c thats were it starts getting pushed some of them work fine others start to crash there . your temp at idle should be down at 46c at idle even with a cheap fan set up .
one thing to look at is the side case fan (if applicable) is it blowing In ?? or outward fighting for air from the cpu ?

if the mem tests are ok .. then i would suspect a driver issue perhaps .

Stone · « **Reply #21 on:** July 27, 2005, 01:37:07 AM »

My PC did have the reboot option in the settings selected. Now I removed it, so I see the blue screen.

But I have the minidumps, that windows generates.

This also means that it IS windows who restarted my PC !

Good news I think:)

I remeber that I once used some tool to analyze the minidump files. They should have the same info as the bluescreen, and more.

I have to goole for the tools now

I run a CPU test for 10 minutes. It did not find an error, but did write at one point:
"spurious 8259A interupt IRQ7"

I dont know what spurious means, but I will google that also

I have two case fans. One in front blowing air in, and one in the back, blowing it out.

The case temp is now after some test running and now finlay back in windows at 40c, it is the alarm I set at 50c, but it never kicks in.

Stone · « **Reply #22 on:** July 27, 2005, 08:55:30 AM »

Well I have run the tests now. The CPU stress test get the heat up 61c and case temp to 41. That is not to bad I think?

The run "dumpchek" on the 4 dump files I have got generated, but they all gave different proboble causes

Last time I had a zeke in my sights at D200 when my pc went *puff*.

Oh well, just have to wait and see when the crash comes next time. Its not a daily problem.

Ohio43 · « **Reply #23 on:** July 27, 2005, 09:03:10 AM »

heh one time during a very hot summer I found myself opening the case and had a household fan blowing on it..overheat solved

llama · « **Reply #24 on:** July 27, 2005, 12:42:47 PM »

Quote

Originally posted by Stone

I run a CPU test for 10 minutes. It did not find an error, but did write at one point:
"spurious 8259A interupt IRQ7"

I have two case fans. One in front blowing air in, and one in the back, blowing it out.

The case temp is now after some test running and now finlay back in windows at 40c, it is the alarm I set at 50c, but it never kicks in.

....

Well I have run the tests now. The CPU stress test get the heat up 61c and case temp to 41. That is not to bad I think?

OK, we're getting somewhere.

First of all, I am going to go on record to say that anyone saying it is the power supply is just plain wrong.

Second of all, I think this shows that this isn't a heat problem. That CPU heating program I directed you to gets my CPU hotter than Aces High, or indeed, any other game I've tried, and if it runs OK for a long time and your CPU never got hotter than 61c, I think you're in good shape.

Now then, IRQ 7 usually relates to your parallel port, or if you don't have one, your sound card or network card. Usually. Not always, but usually. Here's how to find out:

Run the Windows Device Manager (right-click My Computer, choose Properties, click Hardware tab, then click Device Manager button). Pull Down View menu and choose Resources by Type. Click the "+" next to Interrupt request (IRQ). See what hardware is linked to IRQ7. Let us know.

-Llama

Roscoroo · « **Reply #25 on:** July 27, 2005, 07:35:14 PM »

I wouldnt totally rule out the power supply .... but normally it would crash during a stress test .

61c on an amd is hot .. mine hits 50-54c when i stress test it .

i would check into that irq 7 .. you could have too much stuff sharing on that one and its causing a conflict . such as a added sound card ect .

one question . have you tested with a different high stress type game ??? and does it only crash while running ah2 ?

if thats the case then i would suspect a driver issue or an irq , sound card issue perhaps . could even be a bios problem..

bloom25 · « **Reply #26 on:** July 27, 2005, 09:38:52 PM »

What information does the bluescreen give you?

I highly doubt temperature is the root cause of this one. It's probably a driver issue.

Stone · « **Reply #27 on:** July 28, 2005, 12:50:34 AM »

The IRQ 7 thing is nothing to wory about. I found tons of the "spurious 8259A interupt IRQ7" error, and its not an error, just some anoying Linux thing?

I did check the IRQ7 and its free.

I have not got a blue screen since the start of this post, but the minidumps windows writes at a crash all have different reasons.

I have 4 dumps:
1) Probably caused by : ntoskrnl.exe ( nt!CcWorkerThread+269 )
2) Probably caused by : Unknown_Image
3) Probably caused by : kmixer.sys ( kmixer!MxSetMixLvlTable+148 )
4) Probably caused by : HIDCLASS.SYS ( HIDCLASS!HidpDestroyCollection+cd )

I have no other games installed, so I dont know if it just AH that would crash the PC. And it is not a daily thing either. Has hapend about once a week, but always in game.

Shamus · « **Reply #28 on:** July 28, 2005, 10:44:33 AM »

I had the same problem with an asus A8V. updated the via mb drivers and it went away

shamus

llama · « **Reply #29 on:** July 28, 2005, 01:03:25 PM »

Quote

Originally posted by Stone

I have 4 dumps:
1) Probably caused by : ntoskrnl.exe ( nt!CcWorkerThread+269 )
2) Probably caused by : Unknown_Image
3) Probably caused by : kmixer.sys ( kmixer!MxSetMixLvlTable+148 )
4) Probably caused by : HIDCLASS.SYS ( HIDCLASS!HidpDestroyCollection+cd )

Well, let's see...

1. NTOSKRNL.EXE is a process used to boot windows properly. You should NOT be seeing it in the task manager (process list) unless you've been infected by w32.bolzano and variants.

2. Unknown is just that. Unknown. Sorry.

3. kmixer.sys is the audio mixer/volume manager for Windows. I think this is telling.

4. HIDanything refers to a Human Interface Device - mouse, keyboard, joystick, etc. - plugged into a USB port.

I still think we're looking at a multimedia device driver issue. If I were you, I would try updating the following, maybe just one thing a week to see if anything is improving things:

1. Motherboard drivers. (especially if you have onboard sound or network in use).

2. Sound card drivers.

3. Video card drivers.

4. DirectX.

5. Joystick or mouse or keyboard drivers.

Obviously, we're all here for assistance if you need it.

-Llama