Reading between the lines (and trying to interpret how the mix of windows
and chipsets and drivers, it looks like a design problem(?) that causes a
slowdown in disk I/O. There may be other factors involved, the interrupt
controller may be operating in (freaked out mode) and causing additional
slow downs in overall performance. There may be something going on that is
eating a bunch of IRQ triggers.
Guess, guess, guess.
Ymmv.


http://www6.tomshardware.com/mainboard/01q1/010117/kt133-17.html

This correlation is evident particularly in the case of MSI. The
manufacturer offers two versions of the MSI K7T Pro2. The classic K7T Pro2
has the ATA/66-Southbridge (VT82C686A), while the K7T Pro2-A is based on
the ATA/100-Southbridge (VT82C686B). Apart from this difference, the
boards are constructed identically. The result shown above reveals that
the VT82C686B appears to slow down the Northbridge. A different graphics
driver version from NVIDIA does not change this trend, either.


http://www6.tomshardware.com/mainboard/01q1/010117/kt133-18.html

The new VIA-Southbridge with ATA/100 functionality leaves a stale taste.
Especially when used with a single hard drive - as most computers are
operated nowadays -, boards with this chip run slower than their ATA/66
pendants in the field of 3D games and OpenGL applications. The ATA/100
chip only leads to a slight improvement in system performance with classic
2D office programmes if you own a modern hard drive. It seems that VIA
should revise the VT82C686B-Southbridge, otherwise the ATA/100 option as
sales argument may go up in smoke.



http://www.msi.com.tw/support/bios/more.php
Lots of bios updates
http://www.msi.com.tw/support/bios/bios.php?model=MS-6330+Lite+V3.0