Let's play, what the #$@# is wrong with my computer

[Vinceisg0d]

Okies. So, the specs are something like this:

x1900xtx
2gb ram
4200+ amd dual core
enermax 550W ? i think PSU

So, here's whath appens. WHen I was first installing the comp, every now and then it would 'glitch' out, like, hm. Screen would look 8-bitish, kinda like when a NES game didnt load right and it had that messed up 8bit pattern on the screen.

Ever since then, when I moved from place to place (i go home to parents a lot) it would take 10_+ startups to get it goign withotu messing up.

Now, it's been a whole day and I can't get itw orking. Here are some other things that happen:

-I hit the power button, and it starts' working' then shuts off.

-It loads and then randomly shuts off.

-it randomly glitches out in 8bit mode of doom anywhere at all sometimes. (Even in bios settings area place thing).

-I had one bluescreen labelled MACHINE_CHECK_ERROR

-The one time it didn't restart it said my vid card stopped responding it and it had to recover.

So, from what I se, it points at a video card issue.

But what can i do to fix it?

Also, excuse the typing. I'm pissed off and taking shots of tequila.

 

[NoStateofMind]

lmao at the title!

As to your other problems:

Do you have a spare video card to replace this one to see if that is the problem?

Have you used driver cleaner pro and reinstalled drivers?

Power connect connected to video card?

Is the video card correctly seated in the PCI-E slot? (try removing and reinserting)

These are just guesses to hopefully help you or others help you. Good luck

 

[SickBeast]

Sounds like either a heat issue or a power supply issue to me. Check the HSFs (CPU and videocard) to ensure that they are seated properly and that they are not clogged with dust.

If you have a crappy PSU you should replace it. :light:

*edit* If it's the Enermax you're probably fine. Keep in mind that the X1900 cards need a ton of juice. Make sure you have the power connector hooked into it!

 

[Vinceisg0d]

im too scared ot check any of that, because im running and havent crahsed for 30min now logner than ever today.

 

[SickBeast]

Originally posted by: Vinceisg0d
im too scared ot check any of that, because im running and havent crahsed for 30min now logner than ever today.

Must not be good tequila. That stuff usually makes people bold, no? :evil:

:beer:

Try running some Prime95 and see what it does. If it's stable, then your videocard is the likely culprit. Maybe it's loose in the PCI-E slot. :light:

 

[Jschmuck2]

15 bucks on "heat issue." Open up the side of the case and run it for a day - see what you get.

 

[Vinceisg0d]

it's been doing this with case open. That was my original idea.

I also tried a fan on it, no change.

Plus, it doesnt seem to happen in any peticular place that would usually be heat intesive. For example, it happens during the BIOS sometimes.

 

[Acanthus]

Graphical anomalies are almost always a video card failure.

Have you checked vid temps?

Are any of the power rails or thermals looking wild in the bios?

If its not heat or power, i would be quick to blame the graphics card.

 

[LOUISSSSS]

i dont see how it can be a heat error. how do u guys figure? if the gpu/cpu was overheating the hardware itself has a self monitoring device to automatically shut it down before any severe damage is done.

a possibility is that there is a problem with the CPU/MB/ram. mb something went bad or poor contact?

 

[Acanthus]

Originally posted by: LOUISSSSS
i dont see how it can be a heat error. how do u guys figure? if the gpu/cpu was overheating the hardware itself has a self monitoring device to automatically shut it down before any severe damage is done.

a possibility is that there is a problem with the CPU/MB/ram. mb something went bad or poor contact?

GPUs will show graphical problems long before they hit the shutdown threshold, both nvidia and ATI have their thresholds well over 100C.

 

[BernardP]

Here is another possibility, based on true experience a few years back with the in-laws' computer.

The machine had a behavior similar to what you describe. I couldn't find anything wrong, nor one of my co-workers who is a much better hardware tinkerer than I am.

So I took the machine to the Seanix service center (it's a Canadian brand), and the tech found it in a few minutes: a defective front panel on/off switch. There was a false contact in there. A new on/off switch cured the problem for good.

Worth a look?

 

[Vinceisg0d]

Also, what could cause this to happen consitently irregularly.

I mean, when i first set it up yesterday it was happening every 1sec-2min. Then it started happening ever min to 5min, then 15 to 30, then rarely every now and then. Same thing today...

 

[SergeC]

I have to agree with the person who suggested you reseat your card. I'd take it further, and check all the connections. Reseat the cards, etc.

Sounds like you've moved it fairly frequently, I'd bet something's just a bit loose.

 

[Vinceisg0d]

As a side note, this started happening the day I got it. We originally thought it was something in the BIOS as it was happening before we could get XP on it. We changed some settings and it randomly fixed.

I took out the video card and put it back in. Crashed immediately on startup and crashing regularly since I did it.

The same effect of moving the computer to another location.

It cant be a heat issue on the video card, as it happens during startup at times. Doesnt seem likely.

I think I should just take it to a repair shop. How much ya think they'd charge me to find out the problem?

And, just got another VPU not responding error followed 10 seconds lat er by a crash.

I'm running Prime95 now, a torture test.. not sure what to do, but that seemed like a good way to force-crash it.

Also, any other tests like this would be welcome, or any programs I can use to report statistics/etc of my temps/any other things you need.

I wish I could screenshot and show you guys what happens when it glitches out. Ill try to find a screenshot. If anyone thinks they know what it looks like and happens to have a screenshot lemme know

EDIT: http://www.nvnews.net/vbulleti...ad.php?t=82074&page=13

That's the closest pic I could find as to what it looks like. Also, the computer totally seizes up when this happens.

 

[jjzelinski]

Two things, one should be able to answer immediately.

First, how does the card bhave in another system and how does another card behave in your system.

Second, check the 6 pin socket solder points on the card for 12VDC. One row should have 12vdc, the other zilch.

EDIT:

My concern is that the GPU isn't being powered properly because of a flakey rail/coupling which would cause the PCI-e bus to compensate (bad.)

If you can make it into the BIOS also consider setting your PCI-e bus frequency to around 110Mhz.

 

[nitromullet]

I'm leaning towards either video card or power supply, but it's pretty difficult to even begin to tech this with the lack of information about your components...

"enermax 550W ? i think PSU", isn't exactly detailed info. It might help us if you could check the PSU and tell us exactly what model it is. If it turns out it's a Raidmax 250W instead of an Enermax 550W, this thread would be over. Also, exact motherboard and RAM brand and model are useful as well.

Sorry for your troubles with your rig, but the more info you can give us the better.

 

[Deinonych]

Do you have a spare video card you can pop into your system? When I suspect a component failure, I always swap in a component that I know is working.

 

[AdamK47]

Originally posted by: Vinceisg0d
WHen I was first installing the comp, every now and then it would 'glitch' out, like, hm.

Like hm you say? In a hm situation like this I feel it's best to take the entire computer apart and put it back together piece by piece. Please report back with your reply of great success.

 

[Vinceisg0d]

A8N-E Mobo
OCZ 520W PSU. The reason why I wasn't unsure was because I originally wanted the enermax, and couldn't remember.

prime ran for 9hrs also, no errors.

 

[Cr0nJ0b]

Ah...A8N-E....my nemesis...

No, Vinceisg0d, you are not some mad cow...I have this board and and regardless of what anyone says...It's Crap!

I have replaced my BIOS with a new physical chip at least 6 times. I have replaced the Fans 3 times, which you should check (the chipset fan is crap and will break at least once). I have swapped the board completely twice...and in one of the swaps It took 3 times for them to send me a working board! I'm not kidding...When it works it works, but God help you if you are having issues.

Here are a couple things I have learned.

1) If you boot the system with USB cables in certain slots you can get errors. For a while my system wouldn't boot with certain peripherals in certain slots...lots of messing and I'm fine now.
2) Don't F&*$ with the BIOS unless you have to.
3) hook up a PC speaker to the boards speaker pins to hear to the boot beeps. You can quickly diagnose memory and graphis issues that way.
4) Heat is not your friend. My rig is not in an ANTEC P180 with 6 fans and a sunbeam controller...with Everest It stays quiet and cool
5) lots of memory issues with the Mobo especially working with Dual Channel and trying to get them recognized...had to play slot polka.
6) There was an issue where the BIOS wouldn't allow me to move around my the boot order of my RAID group...but I got a work around.

So with all of that said...here are some more thoughts.

If you are seeing that funky 8-bit display during boot, before the OS loads then I'm going to say 85% it's your graphics card. Probably power, hardware or heat in that order. Sounds like you aren't getting enough power to the card, but sometimes it get's just enough, other times not. I bought about 200W more than I needed just to make sure that this wouldn't be an issue. It's really hard to diagnose. If you can get a bigger PSU to test...even, just one that you have laying around that you can plug your peripherals and junk into to reduce some of the load....then see if that helps. I have had hardware issues with my Graphics card too, so I wouldn't rule out a bad card...again, very hard to diagnose...I usually just RMA the board and see if that helps. Heat I'm doubtful about unless it's your chipset that's overheating and taking down your PCI-E lanes. That will do lots of bad stuff...

But, please do try booting with no peripherals...or minimal cards and peripherals in the system to see if you can make some sense of this.

good luck and God Speed.

 

[nitromullet]

prime ran for 9hrs also, no errors.

What happens if you try something graphics intensive?

 

[Vinceisg0d]

Move to my town and figure this out for me. I dont feel comfortable doing eveyrthing you've suggested.

If WoW is graphic intensive, then the same thing that laways happens. Random, inconsistent crashes. Sitting here idle with nothing running will crash the same amount as with WoW running and a bunch of other things.

 

[jjzelinski]

If you're not comfortable doing what others have suggested (basic PC troubleshooting) then you should probably abandon your effort here and take it into a shop. Hopefully they won't charge you $70/hour, but the probably will Imagine if they tell you that you need a new $200 card on top of the $200 you owe them for labor

Good luck

 

[nitromullet]

Originally posted by: Vinceisg0d
Move to my town and figure this out for me. I dont feel comfortable doing eveyrthing you've suggested.

If WoW is graphic intensive, then the same thing that laways happens. Random, inconsistent crashes. Sitting here idle with nothing running will crash the same amount as with WoW running and a bunch of other things.

My vote is that the PSU is not supplying the video card with enough juice. I'm not sure which OCZ 520W you have, but they appear to have offered a number of configurations for 520W PSU's...

ModStream OCZ52012U 520W: +3.3V(28A), +5V(52A), +12V(28A)
PowerStream OCZ520ADJ 520W: +3.3V(28A), +5V(40A), +12V(33A)
PowerStream (SLI-Ready) OCZ520ADJSLI 520W: +3.3V(28A), +5V(40A), +12V(33A)

If you have the ModStream, it's possible that the +12V rail isn't sufficient for the video card or that one of the modular cables is not working as it should. I actually had an issue similar to your's with a GeForce 6800GT that I resolved by getting rid of an extension cable. High end video cards can be finicky with their power. I would suggest maybe try re-arranging your wiring so your video card gets power from different cables. Also, make sure that you are not sharing the power cable that goes to the video card with anything else.

All of OCZ's 520W PSU's have been discontinued btw... Not sure if that is significant or not.

 

[Vinceisg0d]

Is there any sure-fire way to figure out if its a PSU or not? I fucking hate how everything with computers is guessing. And if you don't have another piece to replace and test individuals, you're screwed.

Honestly if I had a better PSU to test it out I'd just be using it not this one anyway...

And the PSU is the last one,

PowerStream (SLI-Ready) OCZ520ADJSLI 520W: +3.3V(28A), +5V(40A), +12V(33A)