Not sure what the problem is...

[TheOneGuy]

Okay, I'm not entirely sure what the problem is, but I think I've narrowed it down to the motherboard, the CPU, or the RAM.

About two years ago I got a new motherboard, CPU, and video card. The motherboard being an Asus A7V8X-X, the CPU being an AMD Athlon 2600+, and the video card being a Geforce 4 (I don't remember the exact specs, but that's not important anyway).

At first everything worked just fine and dandy, but then I started to have problems. Every now and then, my computer would just kind of...die. I really don't know what the problem was/is. The monitor would go black and then turn itself off. If I turned it back on, it would give me a message along the lines of "Video Mode Not Supported," or something like that. That's not the exact message, but this is also irrelevant. Anyway, I would always have to restart the computer to fix the problem. In time I got used to this, as it didn't happen very often. Then I decided it was time to get a new video card to play some of the newer games, and so I did. It's what I have now, an ATI 9600 XT 128MB. The problem persisted, but still not very frequently.

Then I got a new monitor (a 19" LCD Flat-Panel monitor), and the problem persisted once again. Still it goes black, then turns off, then when I turn it back on, it says "Video Mode Not Supported." (This time it is the exact message.)

So I'm starting to get fed up with the problem, as it seems to be happening more and more often as time goes by. Then I got a new hard drive (Western Digital, 160GB, if that's even relevant) and purchased a copy of Windows XP, hoping the problem might be solved by installing a newer version.

I did all this, installing Windows XP on my brand new hard drive, and suddenly the problem is out of control, it happens about two minutes after I turn the copmuter on each time. Same manner. But now I realize that it's onlny when I open programs. So I turn it on and leave it be, without opening any programs, and it doesn't kill itself for about three hours straight. Somehow I managed to install the newest Motherboard drivers/BIOS and the problem went away. Or so I thought.

The next day, it happened again. This was yesterday. It's not so consistent as it was before I updated the BIOS, but it's still much more consistent than before I got the new hard drive and installed Windows on it.

So, now I'll give you my exact specs...
OS: Windows XP SP2
Motherboard: ASUS A7V8X-X
Sound card: (the one that's attached to the motherboard)
Video Card: ATI 9600 XT 128 MB
CPU: AMD Athlon XP 2600+
RAM: 512MB

If you need something else, just ask, I'm not exactly the most tech-savvy person, even if I'm not completely ignorant, so I'm not entirely sure what is needed.

Also, sorry for the incredibly long post. I really don't have any idea how to explain it otherwise except for some vague "Oh, my computer likes to randomly die," statement" So, there you go. I sure hope someone can help, 'cause I'm about ready to throw this computer out the window. Thanks in advance for anyone willing to do so.

EDIT: Oh, yeah, I posted this topic in here because updating the motherboard BIO/drivers seemed to affect the problem Otherwise, I would have posted it somehwere else...not that I know where that'd be. I'm still not entirely sure if it is a motherboard problem, but it's the best thing I've got going for me so far, so...here you go.

[confus-ed]

Welcome to wd forums TheOneGuy, such an incredibly long first post deserves to be answered !

Well after all of that, I'd say overheating is the number one cause of 'my computer likes to randomly die' type comments - have you looked at that ?

& btw this isn't my forum else I'd just edit it myself.. but I'll give you hint that a phrase you used with the word 'legal' in it, is probably best left out !

[TheOneGuy]

The thought has occurred to me that it could be overheating. But my computer stays at a pretty consistent 85 degress Fahrenheit (inside the case, not for the individual parts, obviously). Also, this happens during the winter the temperature can sometimes get as low as 70 degrees inside the case.

(These measurements are according to the little display on the case that tells me so.)

[Campbell]

Two possibilities come to mind on this; the first being overheating and the second is the power supply. It's possible that a component is over heating without the cs. temp. monitor showing it. But I wonder how the inside temp can stay at 85, and how it could even get as cold as 70. Just doesn't seem possible.

What is the output of the power supply? Is voltage constant to all component?

[confus-ed]

..But I wonder how the inside temp can stay at 85, and how it could even get as cold as 70. Just doesn't seem possible.

Welcome to wd forums campbell & thanks for joining in to help

I have to do some some math now to work out if what you are saying is right ! (those damned Europeans have got me thinking in celsius ) - so 85f is 29c & 70f is 21c near enough, I'd agree they were low for case temperatures, but quite possible I'd say if you don't have your PC in a room with central heating & such (also TheOneGuy does say 'in the winter' ) - however these seem to be case temperature readings & not cpu readings, so it isn't to say that isn't getting too hot - & I'd be interested to know what bios reports for cpu temp, &/or some windows utility to tell you when 'under load' (the motherboard manufacturer should have one)

Maybe xp itself can give us some clues ?

TheOneGuy: do you know how to use 'event viewer' (M$ stuff here - there ought to be messages there which might maybe help us diagnose matters )

& also xp creates a thing called a 'minidump' (if it gets chance) whenever it gets a 'fatal condition' & then tries to restart or shutsdown automatically, we want to see any messages, so we know what might be happening, so if you right click 'my computer', choose properties, click the 'advanced' tab, & then pick the button under 'startup & recovery'marked 'settings' - finally on that page there's a tick box marked 'automatically restart' - untick that & then if windows can it will report what 'bust it'

As for voltages, as suggested by Campbell, well yeah maybe - inconsistent voltages are the culprit of much strangeness, but I dunno if we are gonna be in a position to test this, as somehow I can't see TheOneGuy having a multimeter .. but you never know !

[Mayet]

Check your bios to see that the correct video settings are there.

Have you tried another stick of ram to eliminate that from the equation, ram errors bring up some weird messages at times and can be intermittent. Given that your issue happens when using more ram I would look at that

Power supply of course is another issue which is possible, you can eliminate that by testing it with a bigger better power supply

if you are having the same problem over two diff cards I would put the overheating behind some other possible causes but don't discount it

And if the correct settings are set in bios then it could be a dickie agp slot but would be rare.

anyhows thats just my opinion

[Ferrit]

I might be inclined to install Sisoft Sandra or Everest or motherboard monitor and check heat temps there to see if these really correspond to the ones you posted

[TheOneGuy]

..But I wonder how the inside temp can stay at 85, and how it could even get as cold as 70. Just doesn't seem possible.

I did say it was case temperatures. Most of my friends are pretty amazed at the coolness as well. It rarely gets as low as 70, it's only happened once or twice, I was just emphasizing that inside the case it's nto as hot as you'd think it'd be for overheating.

As I said, I haven't been monitoring the individual parts, just the case temp.

TheOneGuy: do you know how to use 'event viewer' (M$ stuff here - there ought to be messages there which might maybe help us diagnose matters )

No, I don't. Never heard of it before, but after posting this I'll be sure to look into it and edit/post with any new info.

Also, I've now unchecked the "Automatically Restart" box, so if something happens, I'll write down what it says.

Check your bios to see that the correct video settings are there.

I have done this before, but I'll do it again, just in case.

Have you tried another stick of ram to eliminate that from the equation, ram errors bring up some weird messages at times and can be intermittent. Given that your issue happens when using more ram I would look at that

This is to be the next new component I buy. However, right now I don't have the money and I don't have any extra RAM laying around. I think I have a friend who might, though, so I'll see if I can borrow it and see what happens.

Power supply of course is another issue which is possible, you can eliminate that by testing it with a bigger better power supply

When you say power supply, are you talking like the power supply that came with the case? Or like another outlet?

Excuse my ignorance. (:

EDIT: After looking into the Event Viewer thing, I'm confused. I looked at System errors, figuring that would be most appropriate for my problem, and I don't know which error to look at. Are there supposed to be hundreds of errors within 3 seconds that all say the same thing?

Event Type: Information
Event Source: ati2mtag
Event Category: DVD_OV
Event ID: 62486
Date: 7/3/2005
Time: 3:35:20 AM
User: N/A
Computer: THEONECOMP
Description:
Invalid parameters
Data:
0000: 00 00 00 00 01 00 5a 00 ......Z.
0008: 3d 00 00 00 16 f4 00 40 =....ô.@
0010: 00 00 00 00 00 00 00 00 ........
0018: 00 00 00 00 00 00 00 00 ........
0020: 00 00 00 00 00 00 00 00 ........

All exactly like that withing three seconds. Does that help?

EDIT AGAIN:
Here's the current CPU temp and Motherboard temp.
CPU: 59 C/138 F
MB: 41 C/105 F

My current in-case temp, if you're interested is 91.0 F. (BTW, the 85 F thing hasn't held up to well the past couple weeks. It used to be like that, but not so much anymore.)

Is there any way to see the temp of everything immediately before it died? Like to store the data? I'm using the Asus Probe that came on the CD that came with my motherboard to do all the monitoring..

[Campbell]

Power supply that came with the case. It appears, though I may be wrong that the video card may be a source of problems. Since you've changed cards perhaps the AGP slot? The data line "invalid parameters" could indicate memory errors, (that would account for the loss of video drivers).

[TheOneGuy]

Yeah, I can't make heads or tails of what that stuff says, except I know ati implies video card, so I'll take your word for it.

Also, this may or may not be related, but when I unchecked that previously mentioned, "Automatically restart" box that confus-ed mentioned, everything seemed fine. Obviously. But then about two minutes ago, my computer suddenly started going incredibly slow. I was forced to restart in order to fix the problem, but I did manage to see that the CPU usage (according to the task manager) had spiked to 100% when it's usually at about 25-35%, with Winamp running pretty much perpetually and me surfing the internet. Related or not? It's the only thing I changed, except for installing Asus Probe. One more thing. Nothing new appeared in the Event Viewer.

[Mayet]

Probably unrelated to hardware issue but possible. i would be more looking at whats in your startup and what spyware you have onboard for that one. Once again though ram can be the cause of this. i would really discount and eliminate the ram before going any further...

[confus-ed]

He-he-he this thread is most appropriately titled .. .. as I don't think we do either, lots of conflicting info ..

So to eliminate one avenue of possible 'wrongness' run a memory testing tool :- Windows Memory Diagnostic or Memtest86

Can you manage to give us a screen shot of what's in event viewer ? (clear all the events, shut down straight away & then run again until it 'falls over', then show us the log ) - I'd say you've got some video driver issue myself at this point - most especially if you are getting 'hundreds' of entries.

Temperatures cpu wise seem a bit high too .. was that reading from asus probe, under 'idle' or 'load' ? - all this random turning off, could be your motherboard deciding your cpu is too hot & shutting it down - you reported 59c .. but most bioses have their auto-cutoff temps set at 60c (you ought to be able to look in bios & see a value for this)

[TheOneGuy]

Yeah, sorry about that. I'm trying to be as clear as possible, but it's not easy. Sigh.

OK, I've run Memtest86 and received no errors, so apparently memory isn't the problem.

I've cleared the event viewer, but I should say this...when my computer crashed earlier today (before I cleared it) no new events showed up, so I'm beginning to think maybe that's another problem I'm having entirely.

I wasn't doing anything at the time, so I'm assuming it was under 'load'. Um, it doesn't say either 'idle' or 'load', but I can see that it's usually at more like 55c. 59c is the highest it gets.

[Ferrit]

Memtest is inconclusive many times use Goldmem

[Mayet]

In my opinion memory testers do not work as well as taking the ram out and trying another strip. Thats my personal opinion anyway. i have seen many ram sticks pass mem tests that have defective bits and many fail the test that work with no issues.

That is only my opinion as a lowly bench tech anyhows....