Author Topic: Odd, Odd, Odd (Read 357 times)

Charon · « **on:** September 24, 2003, 04:52:11 PM »

Been having gremlins in my computer lately. Would post it in the hardware section, but I though this qualifies as a general interest mystery for chipheads. Plus, the problem is apparently "Solved" at least for today

Asus A7V8X-X Via KT400/2500 barton
512 crucial
Ati 9500 pro
Maxtor 30 g HD (few years old)
Win 98SE
SB Live 5.1
Latest drivers all

Started getting some real wierd lockups after a recent rebuild after having no problems for several weeks. Started when I loaded CIV3 Play the World - quick lockups in game and it generally seemed like it kicked off an overall stability decay. Could still play Ghost Recon and AH, but started to get lockups there too, and system seemed to slow down with IE after a few minutes (click on a browser button and it would happen a second or two later).

At first, thought it was a HD problem (which it was actually). Crosslinked files, delayed actions on screen with HD light on solid, eventually no boot. But, I could fdisk/reformat, and reloaded windows. Graphis worked with stock windows 2D driver. When I loaded Ati drivers it wouldn't finish booting. Through much trial and error turning off fast writes in BIOS (which had been on duing the previous working period) allowed the boot, but then a total signal loss to monitor when going from the desktop to any 3D application (or at least 3D mark which I used for testing purposes).

- So, swapped ram - no joy. Reswapped ram - no joy.
- Reset ram bios values to lowest - no joy.
- Tried a variety of older and newer drivers for MB and video card - no joy
- Fdisk/reformat/reload Windows again to use some earlier MB drivers - no joy. Tried more drivers (now going on a second time) - no joy.
- Tried both direct x9a and 9b in various combinations with various MB and video card drivers on the happenstance that I had somehow loaded a buggy newer version of one after the initial HD crash.
- Loaded Nvidia drivers and tried GF3 - worked!

So, today my buddy brought by his radeon 9700 to see if it worked and I just had a bad card (same drivers as 9500). It worked fine, but then, suddenly my 9500 worked fine also using drivers that failed the previous day. Still working. Turned fast writes on - still works.

So, computer docs, what's your guess? I've been doing this since 1988 (only bought 1 "new" computer a 286/12) and I'm pretty good at problem solving but this has me stumped. I thought it might be a corrupted bios (an issue with some Asus boards) but didn't get around to flashing it before it started working. Or a MB hardware fault leading to the video and HD problems. Bad power supply that was having a down period? Can't figure out the link between the deterioration, solid hardware-like failure, and sudden recovery.

Charon

Sandman · « **Reply #1 on:** September 24, 2003, 05:12:00 PM »

I'd check the power supply.

I'm using the same mainboard with:

AMD XP2400+
512MB 333MHZ PC2700 DDR
SBLive
GF4 4400 vidcard

AKS\/\/ulfe · « **Reply #2 on:** September 24, 2003, 05:12:53 PM »

You had no problems with the PC other than in Windows?

If so, did you use an image of a fresh install to rebuild your computer? If you did, its possible there was something corrupt in there that led to all of your problems.

If you aren't having problems outside of windows, or in Windows safe mode, then its more than likely corrupted drivers... or windows is trying to share IRQs.
-SW

AKIron · « **Reply #3 on:** September 24, 2003, 05:25:13 PM »

I think SWulfe is on target. Maybe putting in the 9700 caused an interrupt reassignment solving an interrupt conflict. The 9500 then used the recently reassigned interrupt.

OIO · « **Reply #4 on:** September 24, 2003, 11:37:16 PM »

powercycle the network hub

*hides in shame*

Charon · « **Reply #5 on:** September 25, 2003, 09:59:08 AM »

Good point about the IRQs. Didn't show any conflicts but then I didn't write them all down either and compare manually either.

I think it might be two coincidental problems. The Graphics card being a conflict issue and some form of hardware issue related to the HD.

The system's been pretty stable since yesterday, but I still get some odd stuff related to the hard drive. I was getting a lot of micro writes to the HD from some unknown application after using IE (best I could tell it was IE). I'd just notice the hourglass pointer for a second's duration about every 5-10 seconds. Haven't noticed it since I did the full update to IE 6 (but, oddly, this was happening after the same IE6 update before the last main failure ). Civ3 is now playable in spite of an initial "Font" error. The thing is, all of this had been loaded and trouble free a month ago before the new chip, ram and MB. The new system had been pretty stable up until I loaded CIV3 a week ago and tried to play.

The event that kicked off the graphics problem was a corrupted harddrive that eventually refused to boot after a bunch of typical HD issues (quickly following the CIV3 crash) like excessive access noise and really slow access times. There were a few failed boots in recent months even before the new MB (OS not found type) that would be fine on the next reboot. The fdisk/format restored it to a functional but still quirky state (as noted in the previous paragraph). The only question I have is whether it's the hard drive itself, or something in the MB interface that is the real source of the problem. Time to just see what happens and maybe run the lengthy surface test. Since this is exclusively a Game machine, I can afford to wait and observe.

Charon

boxboy28 · « **Reply #6 on:** September 25, 2003, 10:19:51 AM »

I have the A7V8X (no -x) board and one thing i had to do just so i could play Strike Fighters is set the AGP to 4x in the BIOS
and i also have the ATI 9700 pro.

At first i had severl problems due to my TV tuner card sharing IRQ's and after moving it around and my sound card around in different PCU slots i got every thing to work just fine!