interesting hard/soft/firm ware issue(s)..

Hi all. Recently my computer conked out. The problem was traced to the motherboard (a P5B). A couple of days ago I finally got my system back with a new p5b, and observed a lovely range of system instabilities: screen freezes, screen going to black, and the ocassional BSOD. I eventually put an old IDE drive into the system, disconnected my primary SATA and everything else non-essential, and did a clean XP install. From there I managed to figure out that I had a bad memory stick (first clue: Prime95 crashed instantly. Second clue: memtest kept crashing.)

I'm now still running in my testing mode (clean windows install on the IDE drive) and the system is definitely much improved. However, I've left Prime95 running for prolonged periods (> 2 hrs), and have still observed a couple of system crashes. I do a lot of CAD and rendering on this machine (as well as 'ccasional gaming) so I would like to quash any final issues. Not sure yet if this is a hard/soft or fim issue, yet. (I'm also going to leave it running unattended tonight.)

System specs and my most recent minidump are below (the minidump points to ntoskrnl.exe. An article at computerhope.com suggested repairing ntoskrnl.exe with the windows disk. Has anyone had any experience with this?)

Any thoughts/suggestions/comments would be really appreciated.

-=-=-=-
System specs:

Windows XP Professional Service Pack 3 build 2600 (32-bit),
Core2 CPU 6400 @ 2.13GHz [1540.4 MHz],
1.0GB RAM,
Sapphire RADEON X1600 PRO, Sapphire RADEON X1600 PRO Secondary,
40GB HDD,

ASUSTeK P5B, Rev 1.xx, BIOS Version: 2101
BIOS Release Date: 12/30/2008

Intel® Core2 CPU 6400 @ 2.13GHz
Family 6, Model F, Stepping 6

Memory devices:
DIMM0:
1024MB, 800MHz,

Sapphire RADEON X1600 PRO
Chip Type: ATI Radeon Graphics Processor (0x71C2)
DAC Type: Internal DAC(400MHz)
Memory: 256MB
BIOS: 11x-2C84S5SA-002
Driver provider: ATI Technologies Inc.
Driver version: 8.573.0.0
Driver date: 1-13-2009

Realtek RTL8168/8111 PCI-E Gigabit Ethernet NIC

-=-=-=-

Most recent Minidump
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************

IRQL_NOT_LESS_OR_EQUAL (a)
An attempt was made to access a pageable (or completely invalid) address at an
interrupt request level (IRQL) that is too high. This is usually
caused by drivers using improper addresses.
If a kernel debugger is available get the stack backtrace.
Arguments:
Arg1: 808c4398, memory referenced
Arg2: 00000002, IRQL
Arg3: 00000000, bitfield :
bit 0 : value 0 = read operation, 1 = write operation
bit 3 : value 0 = not an execute operation, 1 = execute operation (only on chips which support this level of status)
Arg4: 806e4ad1, address which referenced memory

Debugging Details:
------------------


READ_ADDRESS: 808c4398

CURRENT_IRQL: 2

FAULTING_IP:
hal!HalpAllocateAdapterCallback+6b
806e4ad1 0fb68a98438c80 movzx ecx,byte ptr [edx-7F73BC68h]

CUSTOMER_CRASH_COUNT: 3

DEFAULT_BUCKET_ID: DRIVER_FAULT

BUGCHECK_STR: 0xA

PROCESS_NAME: System

LAST_CONTROL_TRANSFER: from 80696b94 to 806e4ad1

STACK_TEXT:
f79f7840 80696b94 865c4680 00000000 00000044 hal!HalpAllocateAdapterCallback+0x6b
f79f7dac 805ce858 80087000 00000000 00000000 nt!KdpSearchPageHitOffsets+0x994
f79f7ddc 8054534e 8069590c 80087000 00000000 nt!`string'+0x20
f79f7e64 00000000 00000000 00000000 00000000 nt!WmiTraceMessageVa+0x2ab


STACK_COMMAND: kb

FOLLOWUP_IP:
nt!KdpSearchPageHitOffsets+994
80696b94 ?? ???

SYMBOL_STACK_INDEX: 1

SYMBOL_NAME: nt!KdpSearchPageHitOffsets+994

FOLLOWUP_NAME: MachineOwner

MODULE_NAME: nt

IMAGE_NAME: ntoskrnl.exe

DEBUG_FLR_IMAGE_TIMESTAMP: 498c69f0

FAILURE_BUCKET_ID: 0xA_nt!KdpSearchPageHitOffsets+994

BUCKET_ID: 0xA_nt!KdpSearchPageHitOffsets+994

Followup: MachineOwner
---------


Cheers!

FJ.