Rich said:
If you got a SATA WD Hard Drive do you load the drivers at beginning of
install??
The computer might not be a modern one. The following would apply if
the motherboard doesn't properly support drives of >137GB size (so
called 48 bit LBA mode of operation for IDE interface).
The motherboard is a P4B266-LA (Danube). At least the link at the bottom of this
page suggests that. Nearer the top of the page, the chipset is Intel 845.
There would likely be two IDE connectors.
http://h10025.www1.hp.com/ewfrf/wc/...7048&cc=us&dlc=en&lc=en&jumpid=reg_R1002_USEN
It is possible the P4B266-LA OEM board from Asus, bears some similarity to
P4B266 retail board from Asus. In this table, Asus lists support for
48 bit LBA on some IDE interface. On boards that have both a separate
RAID controller, and also some chipset IDE ports, you can't tell whether
this table applies to both onboard hardware devices or not. The P4B266
seems to fall in the middle of the 48 bit LBA transition period, so the
idea would be, to use the latest BIOS available as one solution.
http://support.asus.com/technicaldocuments/technicaldocuments_content.aspx?SLanguage=en-us&NO=501
support 48bit HDD (137+ GB HDD)
Model Status Since BIOS release
P4B266 Yes 1007
I would try one of two things. These are *test* cases, and not
final installations. They're to prove the problem is 48 bit LBA
related.
1) Partition the disk into two pieces. Make the first
partition less than 137GB. Leave the rest of the disk
unallocated. That way, Windows won't attempt to go over
the 137GB mark.
+-------------+----------------------+
| < 137GB | rest of 500GB |
| WinXP C: | unallocated |
+-------------+----------------------+
2) Use the "Clip" jumper on the drive. A hard drive can have two
jumpers installed at the same time. One jumper controls Master/Slave/CS,
while a separate jumper position controls clipping. If the Clip
jumper is installed, what the drive reports as CHS geometry will
change, in such a way that the OS will interpret the disk size
as 33GB or so. This number is less than 137GB, so the Windows
install cannot go past the 33GB mark. Later, if the clipping
jumper is removed, the full drive size will be visible, but
the existing partitions on the drive will not be magically resized.
The Clip jumper is not always documented on the drive label. Usually
the drive label, concentrate on Master/Slave/CS information. The
manufacturer web site, may give details for where the Clip jumper
goes. (Compare Fig.4 "normal" to Fig.6 "clipped" for more info,
in this example document.)
http://www.westerndigital.com/en/library/eide/2579-001037.pdf
So those are some experiments to try.
If either of the above seems to help, these would be
some other more permanent solutions.
1) Install a DDO. The downloadable Data Lifeguard tool can install
a DDO. That apparently solves the 48 bit LBA problem (I've never
tried it). Most of the time, disk utilities will understand a DDO
is in place. I prefer to keep well away from this myself, but
a lot of people like the simplicity of the solution (lets them
get on with life...). When I'm doing stuff, I generally try
to do things as "vanilla" as possible, with the idea being there
will be less resistance later when there is a problem with the
computer.
http://www.westerndigital.com/en/library/eide/2779-001001.pdf
http://en.wikipedia.org/wiki/Dynamic_Drive_Overlay
2) Install a PCI card with an IDE controller. This will potentially
require pressing F6 and installing a driver for the card, early
in the Windows installation. The card should support ATA/ATAPI 6
standard or later, as I think that is the one that handles 48 bit LBA.
At one time, a Promise Ultra133 TX2 would have been recommended,
but those have ceased production as far as I know. Chances are,
something with a VIA VT6421 might work. More research will be
needed, to be sure it all works. I don't know if the SIL0680
works for this or not - I'm still seeing cards with those on it,
and the chip was slightly tweaked, part way through its production
life. For more info on 48 bit LBA in general, see the Seagate document.
http://www.seagate.com/support/kb/disc/tp/137gb.pdf
The reason that can work, is the PCI card has a disk controller
chip on it, but also has a second chip containing BIOS code. That
BIOS code is loaded by the motherboard during power on self test.
The code supports Extended INT 0x13, which provides the ability to boot.
And hopefully, also supports big drives.
3) Update the motherboard BIOS to the latest revision available, then
try the install again. Whether this is viable, will depend on whether
HP bothered with BIOS updates. There are OEM motherboards that never
see a BIOS update, so there is no guarantee HP will be able to help.
Some companies feel their product is "perfect" and cannot possibly
have bugs. Judging by the release date of the BIOS shown here, I
have a suspicion this BIOS is not modern enough. I wouldn't take a
chance on this one. If you're already running this version, then
there is nothing more to do for this solution (maybe the initial BIOS
screen will show release info).
http://h10025.www1.hp.com/ewfrf/wc/softwareList?os=228&lc=en&dlc=en&cc=us&lang=en&product=71016
HTH,
Paul
