how 2 get this junk off my machine ?

  • Thread starter Thread starter marshalx
  • Start date Start date
M

marshalx

vista64 is running SO SLOW that the drive is making
creaking sounds, and it seems to have grabbed control
of the boot, so i cant issue Format c:
is that a new MS anti-abandonment technology ?

what do MVP do, throw out the whole machine,
or just cut the wires to the drive ? thanks for help.
 
marshalx said:
vista64 is running SO SLOW that the drive is making creaking sounds, and
it seems to have grabbed control of the boot, so i cant issue Format c:
is that a new MS anti-abandonment technology ?

what do MVP do, throw out the whole machine,
or just cut the wires to the drive ? thanks for help.

Boot from a WinXP disc .. go through the motions of an install .. when you
get to the disk paritioning part, remove the partition(s) Vista is on and-or
boots from etc. etc... that done, use F3 to abort the install and reboot.
 
WGA said:
Boot from a WinXP disc .. go through the motions of an install .. when
you get to the disk paritioning part, remove the partition(s) Vista is
on and-or boots from etc. etc... that done, use F3 to abort the install
and reboot.

That didn't work for me. I did exactly as you
describe and XP would hang on the 1st reboot. I
finally had to boot to a DOS prompt with a WinME
disc and execute "FDISK /MBR" before XP would do
a complete installation.
 
marshalx said:
vista64 is running SO SLOW that the drive is making creaking sounds,
and it seems to have grabbed control of the boot, so i cant issue
Format c: is that a new MS anti-abandonment technology ?

what do MVP do, throw out the whole machine,
or just cut the wires to the drive ? thanks for help.

You installed a beta operating system and expected what...?

Either reinstall Windows XP (or whatever you were using before) or your
ghosted disk image backup.
 
Nothing really to add - curious though - I was under the impression that the
Win ME startup disk was FAT oriented and did not recognize NTFS - thereby
not indicating a NTFS "C" drive - Vista requires NTFS- interesting you were
able to run FDISK/MBR.
 
Neither a WinXP SP2 CD nor a WinServer 2003 CD is as good as a WinXP SP1 CD
for fixing an MBR.
 
Does the DOS's FDISK require the presence of a FAT formatted partition to
run with the /mbr switch ?? I'm not sure that it does.
 
You cannot format a system drive while thtat system is running, but
formatting the Vista partition when XP is running should work. However, if
depending on where Vista wrote the Boot Configuration Store you may not be
able to boot into XP until you do some repairs.
 
No, not when your at the DOS prompt. I never
booted into WinME, I just used the install CD to
get to the DOS prompt. FDISK recognizes NTFS
partitions.

You can get to the same place with a WinME boot
disk.
 
Colin Barnhorst said:
You cannot format a system drive while thtat system is running, but
formatting the Vista partition when XP is running should work. However, if
depending on where Vista wrote the Boot Configuration Store you may not be
able to boot into XP until you do some repairs.


***dont worry folks those nice people at Partition Magic told me to drill a
6mm hole in the drive and squirt some oil in, and that did the trick.
 
Sure.

I used to like to tweak the formatting of my drive partitions. I'd go to a
few lengths using the tools supplied by the Windows XP CD and Win98 Boot
disk to get my first primary partition (C:) formatted with 32KB cluster
sizes (allocation unit size) [the Windows XP and Server 2003 install routine
only formats in 4KB cluster sizes for the most part].

This meant creating a second primary partition, making it active, installing
Windows there .. then using Windows XP's Disk Management to format the first
primary partition NTFS with the 32K allocation unit .. then making that
first partition active .. then installing another copy of Windows there.

When Windows XP SP2 came out and I started doing this with the WinXP SP2
disc, I would get an error on reboot during the install routine when
installing to the first primary partition with the 32KB clusters [error:
NTLDR missing]. Neither the FIXBOOT nor FIXMBR of the Windows XP SP2 disc
and the Windows Server 2003 disc would not do the trick fixing it. But the
older Windows XP SP1 disc's FIXMBR did. Then I could successfully reboot and
finish installing WinXP SP2 or Server 2003 onto the first partition with the
32KB clusters.

Someone explained why to me a year or so ago [I'm not sure they were
correct] - now I forget what he said. All's I know is that the WinXP SP1
disc works at fixing the MBR even when the first primary partiton is
formatted NTFS with unconventional cluster sizes.
 
Interesting. Thanks for taking the time to reply.

Now this begs the question, why do you prefer a larger cluster size?
Conventional wisdom dictates that you're not wasting as much space with a
smaller cluster size, but I can't think of any circumstances where a larger
cluster size would still be preferable...

Or have I missed something you've already explained?
 
***dont worry folks those nice people at Partition Magic told me to drill
a
6mm hole in the drive and squirt some oil in, and that did the
trick.

You're sure you had the right number and didn't accidentally get Dogbert
Tech Support? :-p
 
Homer J. Simpson said:
Interesting. Thanks for taking the time to reply.

Now this begs the question, why do you prefer a larger cluster size?
Conventional wisdom dictates that you're not wasting as much space with a
smaller cluster size, but I can't think of any circumstances where a
larger cluster size would still be preferable...

Or have I missed something you've already explained?

Performance. Larger cluster size, in some circumstances, can increase
performance.
 
Interesting. Thanks for taking the time to reply.
Performance. Larger cluster size, in some circumstances, can increase
performance.

In a consistent, measurable fashion? I'm genuinely curious. I'm dying to
see the hard evidence to back this up.

If I was trying to get all the performance I could get out of a machine, I'd
more readily believe that RAID would provide more significant performance
gains than just a larger cluster size. I'm also more inclined to believe
such gains would be more significant on a plain data drive (eg, hosting a
large database file) rather than the OS drive itself--thus entirely avoiding
the hassles you've described trying to change cluster size on the OS drive.
 
Hm hm .. In my opinion - to be honest, arising from dabbling - my Maxtors
using Intel RAID0 + 32KB stripes + 32KB clusters load data quicker. I
notice when lauching a newgroup with, say, 150,000 messages etc. etc. I
didn't notice much in terms of OS performance.

Generally speaking: A while ago I went about reading on what is the best
combination of RAID0 stripe size and cluster size. I found it's nothing one
can pin down as it varies for setup to set up e.g. Maxtors behave
differently from, say, Western Digitals. So if one had Maxtors they'd set
the strip at 32 whereas if something else they might get better performance
at 128. One would have to uniquely test harddrive + controller + cluster
size to figure the best performance for their situation.- and they widely
vary. Sorry I can't be more specific - it's been a while.

Maxtors, BTW, generally do seem to work speediest when the stripe size is
set to 32KB - go figure - like something internal was made for that size.
They seem pretty speedy in a RAID0 32KB stripe and 32KB cluster.

The thing to do is search the web and read numerous experiences and reports.
If you are really handy you might perform some real-world tests e.g. HD Tach

In more recent times, this past year or so, I stopped caring about it. It
basically doesn't matter at all for the small business / gamer / hobbiest /
home user - not at all - although the hobbiest tweaker might enjoy it just
because.

However, for someone running a busy web farm dishing out data like there's
no tomorrow, then such sublties magnified can make real differences in
customer happiness, in equipment and in money. I did notice that guys who
shelled out large files often reported better performance from 32KB
clusters - or bigger - fewer seeks or something .. ;)
 
Back
Top