Abit BX133-RAID problem's

[BearCat]

I am posting this for a friend of mine, any input's will be appriciated.

[quote:92bfee198e]
After upgrading from WinME to WinXP (final build 2600), I have got
allot of trouble with the stability of my computer.
My system consist of the following components:

-Abit BX133-RAID Motherboard
-Pentium3 FCPGA 700 MHz processor
-2 ea. IBM-DTLA 307030 30 Gb Harddrive, RAID 0, 64K stripe, NTFS
partition
-256 Mb PC133 SDRAM
-ASUS V6800 GeForce DDR graphic card

(I have a number of other components, but it's not essential, since
I've tried unplugging all w/o any improvements).

Here is a description of my problem:
When I stressing my harddrives, i.e defragmenting or using SiSofts
Sandra 2001te Pro's "drive benchmark", I get the "blue screen of
death" saying "Hardware Malfunction". This happens at the default
speed of the processor (700 MHz). At higher front side bus setting,
the problem get worse; at 133 FSB (933 MHz), the "hardware
malfunction" happens at normal use of my computer.

I had no problem running WinMe. Stable as rock at 933 MHz. Only
difference is, in WinXP I'm using NTFS instead of FAT32, and 64K
striping instead of 16K.

This is what I've tried to solve the problem:
-Unplugging every PCI card, CD/DVD ROM and external devices.
-Disabled RAID, installed WinXP on each harddrive.
-Setting a lower PIO mode
-Flashed the BIOS with different versions, including one "modified" ZW
mainboard BIOS with the HPT370 1.2.0612 BIOS.
-Uppgraded the original HPT370 1.0.5 windows driver to v1.2.0612.
-Tried different BIOS settings
-Installed IBM's Drive Fitness Test (version 2.10) and stress tested
each harddrive in DOS (no errors reported from the S.M.A.R.T. system)

None of the above made any difference, still getting "hardware
malfunction".

I also installed HD Tach v2.61 and the result from the test indicates
that something is not right. The graph is very erratic, totally
different from what i got in WinME. I graph was almost straight in
WinME, and the "Random Access Time" around 8,5ms. The "Sequential
Speed" was also much higher, but that could be the difference between
16 and 64 K stripe. I have attached a snapshot of the test to this
mail, so you can see for yourself.

Since everything was ok in WinME, could there be a incompatibility
issue with motherbord/harddisk controller/RAID controller and WinXP,
or could there be a faulty motherboard that only shows with WinXP?

[/quote:92bfee198e]

[b:92bfee198e]Anyone[/b:92bfee198e]

[Sephiroth]

i think it is probably going to be a problem with xp and the ide controllers, try different windows 2000 drivers for the ide and raid controllers and see what happens

stock bios, no overclocking, no pci devices, no raid....bet it works...then build up and see where it pukes


and most definately a CLEAN install

[ October 03, 2001: Last edited by zonko ]</p>

tha seems odd.
has your ambient house temp increased?
you cpu could be overheating
check for a fan failure
and make sure the heatsink is setaed proeraly
then try to underclock the FSB and see if it is stable then
if it is none of the above it might just be so sort of hardware/software conflict

[Sephiroth]

it shouldn't be hardware defect such as bad motherboard or the cpu overheating, if it was that then it would show up under me too, the only thing i can think of other than trying different raid drivers is try to use fat 32 on the partition instead of ntfs and see what happens, and i'm not sure if this is going to be possible or not as i've never had a lot of experience using raid or scsi but can you use 16k striping under 2000/xp? try that too