I'm an ADVANCED user, but this time I'm STUMPED!

I'll try to keep this brief, yet informative, but I'd like to say that if anyone can assist me with fixing this problem, I promise to make it up to you however I can!

(For hardware specs/dxdiag see the bottom)

That being said, my problem may or may not be hardware related... as a matter of fact, I may be leaning towards software, but it's iffy. Here's the progression of events: I install Windows XP (32bit, SP2) and my usual software (spyware, etc) and drivers and everything is fine. I install games and they run great. However, at a certain point, the games stop loading (none even start to the menu screen.) This is followed by my computer not loading Windows--what it does, however, is odd: it begins the Windows loading screen, then before it loads the desktop, the monitor loses signal and the screen goes blank. There is no drive activity at this point, so I know that nothing is happening in the background (i.e. Windows does not load).

Initially, one would suspect either a.) drivers or b.) hardware. But here's the oddity: I have formatted and reinstalled Windows at least 8 times (not an exaggeration--it may be more) now trying to isolate the problem. Each time, I am able to play the same games, run the same software, etc., and tax my computer to its processing limits UNTIL this problem rears its head again.

I have tried refitting my graphics card, though it shouldn't be the card since it can run everything fine until this bug happens each time I reinstall. I have tried numerous drivers, 2 different versions of Windows, and I do use anti-virus scanning so that isn't it either.

Also, this computer was working fine with this same setup for several months with no hitches until whatever is causing this anomaly began rearing its ugly head.

I consider myself an advanced user and I shy away from asking for help, but this has got me stumped. No man should have to reinstall his operating system this many times, and I'd like to prevent further cycles! Anyone with thoughts or questions can reply here or directly to me at aaronpaullow@gmail.com. Yea, if I'm giving away my e-mail addy, I have gotta be desperate!

Dxdiag at http://www.aaronpaullow.com/DxDiag.txt

Intel Core2 Duo 6600 @ 2.4 NOT overclocked
ASUS p5n-e sli
Sapphire Radeon x1950 Pro 512mb
Layla 20 sound card (pro audio card)
2 gb corsair performance memory
Video/CPU with arctic cooling heatsink/fans
4 raid sata HDs
600 watt antec truepower

 

Does this problem happen every time you run the PC? Does it only happen when running certain programs or will it happen after the PC sits idle for awhile? I'm thinking it could be a couple different things: flakey RAM, CPU, or hard drive; maybe a thermal problem with the CPU, hard drive, or video card. I would run some deep diags on the RAM and hard drive for starters. Try pulling one RAM stick for awhile, then pull the other one and see what happens. Sometimes just re-seating the RAM sticks can fix problems. I had a PC I was working on, and it wouldn't POST no matter what. I eventually pulled the RAM sticks (3 of 'em) and re-installed them one-by-one in different slots that where they originated. That fixed it! So give the diags a shot and try moving the RAM around and maybe make sure all your fans and heat sinks are clean and spinning freely, and double check all cable connections; both power and data connections.... GOOD LUCK!

hopperdave2000

 

Thanks for your reply. No, the boot problem (monitor blank) happens on average 4 out of every 5 boots. Safe Mode always works. Also, odd as it is, when I have NO graphics driver installed (totally uninstalled and after using driver cleaner) it does boot every time.

Whether it boots or not, and no matter what driver I do have installed, once the problem begins no games work.

The reason I can't imagine that this could be hardware is that, as I mentioned, when I reinstall windows and install my games/drivers/etc. everything works GREAT. But, after a while, the problem comes back, which leads me to believe this is somehow software related.

Another guy thought that this was power-related, but I've squashed that theory by trying a different power supply and even removing all but the most crucial components.

I will try re-seating everything in the computer and see what happens. I just can't imagine that if I have a hardware problem, it would work perfectly every time I reinstall windows (!!!) Does this make sense? I mean, even if it were a hard drive problem, it shouldn't work fine again when I reinstall windows, right?

 

I have bad news for you rocketmorton. Alas, you're video card is failing. Since it only shows up when "drivers" are loaded, it points to the accelerated graphics portion of your video card. Optimistically, it might be heat related. In other words, if you can cool your video card a bit more, you might find that the "problem" vanishes. Best of luck to you, rocketmorton.
.

 

If this is the case, why would the video card work fine when I reinstall Windows? I've reinstalled many times and every time, all of my games work again, perfectly... that is, for a couple days until the problem comes back. WHat do ya think? Does this still sound like a hardware issue?

 

I still think it's your video card overheating. Run this diagnostic program. It's called Everest and there is a section which shows the temperature of various system components. http://www.majorgeeks.com/download4181.html
It might give us a better idea of what's happening.
.

 

This is what it shows:


Field Value
Sensor Properties
Sensor Type ITE IT8712F (ISA 290h)

Temperatures
Motherboard 43 °C (109 °F)
CPU 22 °C (72 °F)
Aux 38 °C (100 °F)

Cooling Fans
CPU 2909 RPM
Chassis 2801 RPM

Voltage Values
CPU Core 1.31 V
+2.5 V 3.22 V
+5 V 4.84 V
+12 V 12.03 V
+5 V Standby 4.87 V
VBAT Battery 3.10 V
Debug Info F E8 F1 FF
Debug Info T 38 43 22
Debug Info V 52 C9 00 B4 BC 02 05 (01)

 

That's odd. I don't see a GPU temperature. I was expecting something like this:

Temperatures:
Motherboard 28
CPU 33
GPU 42
GPU Memory 39
GPU Memory 42
Maxtor 6Y080L0 34

Cooling Fans:
CPU 2021
Power Supply 1383
GPU 3763

Voltage Values:
CPU Core 1.52
+3.3 V 3.28
+5 V 5.00
+12 V 12.04
+5 V Standby 5.09
GPU Core 1.41
GPU Vcc 3.18
GPU AGP VDDQ 2.51
GPU FB VDDQ 2.94
Debug Info F FF A7 7A
Debug Info T 28 33 208
Debug Info V 61 C6 CD BA BE 04 00 (03)

 

I went into the Computer/Sensor thing and I copied exactly what's there. I have used Ati Tray Tools, of course, to examine the GPU core temperature and it never exceeds 52 degrees C. It has a pretty nice heat sink on it...

However, it doesn't tell me the temperature of the graphics card's memory...

 

Not being an expert on Everest, I am wondering to myself, if the Motherboard Chipset Drivers need to be updated. I regularly update my own Chipset Drivers, so perhaps I get a more detailed diagnostic. The only other thing that pops out, is the 4.84 Volts on your 5 volt rail. I don't expect to see a voltage that dipped with a 600 watt power supply. I'd re-inspect my plug ends. Pull them out one at a time, look at the pins and look for burnt/discolored or pushed back pins. Then wiggle them back into place when you're re-seating them. I realize that your video card is a major power user, but a modern 600 watt ps should do better than that.

 

I'm beginning to think you are right about the power, but not the unit itself!

I don't know why I never thought about this, but I neglected that I live in Upper Manhattan in a pre-War apartment (they're all pre-war... or pre-SOME-war). The electrical systems aren't exactly perfect, and my building shares its energy very closely. Power usage goes up post-winter and I'm beginning to think that my power supply isn't pulling its weight when plugged into my power conditioner (I'm an audio guy, so I have to have everything plugged through it to prevent noise). When I plugged the computer directly into an outlet, it starts up fine 100% of the time (so far at least). Games still don't work, but that could be a result of driver corruption at the Windows level (directx integration has some serious setbacks when things go wrong).

No, I'm NOT claiming victory yet, but this is a promising start!!!

I don't know how to raise that voltage setting though... is this something I can adjust manually?

 

I have bad news and sorta good news...

I did a longterm test on the memory, and a test on the hard drives. All were fine with no errors.

I bought a Thermaltake Toughpower 750watt to test out the power supply theory, and I'm still having the problem. I will probably try a fresh install just in case, but it seems it is not a power issue.

The GPU is cooled by an Arctic Cooling hs/fan that keeps it around 60 degrees Celsius during games (yes, I've tested) so it's not a heat issue.

Any other thoughts? I can't keep any of you guessing too much on the cause of this problem for long (at least, before you reluctantly give up), but believe me when I say that I am trying every option suggested--even buying a $200 power supply! (which, if it doesn't help, I may just return)

Here's hoping there's something obvious that we missed!

 

Can anyone tell me how to test my video card's memory?

 

My next piece of advice ( lame as it may sound ) is to replace your video card. Don't buy another video brute from ATI, just get a DX9 run of the mill type card from nVidia. Currently, your card is from ATI and I'm not sure how well that brand works on the SLI type mother board. SLI boards use nVidia chipsets and there may be issues at work that are difficult to detect. It would be nice to resolve this.

 

Success!!!!!!

I mean.... Failure!!!!

Whatever you wanna call it, it was in fact a problematic video card all along. It failed that "Video Card Stability Test" a couple times, and that got me thinking! So, I did some more tests and lo and behold, it's defective. Odd, though, it seems as if it is something minute that is defective, like a single onboard ram dimm or something of the sort.

Why do I sound so excited that my video card is dead, you may ask? 2 reasons:
#1: you guys helped me solve the most unfathomably vexing computer problem I've had in 10 years

and..

#2: newegg.com let me rma the thing, so the replacement won't cost me anything save shipping.

So, thanks to everyone for the help! I doubt I would have come to this conclusion on my own!

 

That's what I call success. Even better than that, a lot of people just learned something, because you posted back. Thankyou for that, rocketmorton.

 

A UPS will usually fix brownout problems, by basically taking what it is given, internally turning into DC voltage, and then remaking the AC from scratch. There is a catch though, not all UPSes do this, they sit on standbye waiting for power to fail. They also do make variable transformers to boost the voltage. Both options are expensive, and probably not worth the effort.

Personally I lean towards memory issues. Switching power supplies don't care that much about input power voltages, they just suck more current and compensate. Most of what has been said sounds reasonable, so I can't improve on it a bit.

Opps, just read the last two posts, now I have egg on my face. Sorry about that.