Tony

Distinguished
Aug 5, 2001
1,944
0
19,780
Archived from groups: microsoft.public.windowsxp.help_and_support (More info?)

Xp(home)_SP2

When trying to minimize Windows, I compressed
ServicePackFiles, burned the original to CD then deleted it
from the hard disc. sfc /scannow was then run,which brought
up the expected demand for the XP Pro disc.
Rebooted and added 'zip' to the SourcePath key. No joy.
Changed the key, unzipped and repeated, whereupon I was
asked for the Home CD. Inserted my slipstreamed version,
which was rejected, so had to use the original.

So far so good but, when I looked at the unzipped
ServicePackFiles it had doubled in size! It contained i386
and a second folder, ServicePackFiles, (empty), but still
occupying 367MB. Deleted this and carried on.
Then compressed the folder using NTFS compression, but the
only change I can see is that the filename is blue, while
the contents occupy the same amount of disc space as before.

Questions:-
1) Does the rejection of the slipstreamed disc mean that
it is corrupt, or is it only useable for reinstallation?
How can I test - without reinstallig <g>?
2) Why was the unzipped version twice the size of the
original?
3) Why is there no reduction in size after NTFS compression?

More questions:
4)Are there any folders in Windows which can, safely, be
deleted or permanently compressed, yet,if compressed,
accessed when needed, seamlessly? I know that the
installation backups,excluding the first two folders in Windows,
can be deleted if there are no problems with updates, but
Help\Tours, Resources\Themes,SoftwareDistribution, InternetLogs
and DownloadedInstallations look likely candidates for
compression, particularly the last because I have the
original setup programmes. NTFS compression seemed the way
to go but, if there is no space saving, pointless.

5)Internetlogs contains an unreadable 20 MB text file
called tvDebug.txt. QuickView shows that it contains thousands
of mainly blank pages with the odd heiroglyphic on just a few
of them. Is this recording every connection I've made since I
bought the PC and, if so, could it be renamed BAK, a new file
created, then the bak file deleted?

6)While examining the ServicePackSourcePath, in registry,
I noticed a reference to ServicePackCache, but was unable
to locate it in ServicePackFiles. Is this of any significance?

Unrelated to any of the foregoing:-
7) Why - if it's true - is it more efficient to have the
pagefile on a non active partition, away from the OS?

TIA for all advice and information.


Tony.
 

Tony

Distinguished
Aug 5, 2001
1,944
0
19,780
Archived from groups: microsoft.public.windowsxp.help_and_support (More info?)

Thanks for all the info, Galen, and 'here here' [or is it 'hear
hear'?]to the Sherlock quote; it's why I'm performing this exercise.

Not short of space - disc less than half full - just curious to know
how much can be stripped out without causing total collapse, although,
of course. I've backed up and created a restore point first.

Solved the slipstreamed disc problem. Just needed a couple of registry
tweaks -
took hours to find them though. Used Sherlock's dictum 'When you have
eliminated the impossible, whatever remains must be the truth.'

Probably addressed these queries to the wrong group, as I'm only
interested in the *why* rather than the *how*. Conventional wisdom
says put the swapfile on a separate partition, but I've not seen
anyone explain why. My own surmise is that
when on the OS partition it 'floats', changing size and position, so
windows has to 'search' [don't you hate anthropomorphic words applied
to a PC?] for it whereas, in a fixed location it 'knows' and can head
straight there. But that's pure speculation on my part.

I understand why decompression often produces a larger file than the
original but have never experienced anything like a doubling.

A batch file, incorporating windows 'sageset' and sagerun, together
with items of my choosing, keeps my disc clean and includes clearing
Internet Logs, but this ginormous file lies there like a malevolent
toad. I'm going to squash it and see what happens!

Regards.

Tony.
 

galen

Distinguished
May 24, 2004
1,879
0
19,780
Archived from groups: microsoft.public.windowsxp.help_and_support (More info?)

In news:39ac0ec1.0504261159.f3b3eec@posting.google.com,
Tony <tonybocymru@evemail.net> had this to say:

My reply is at the bottom of your sent message:

> Probably addressed these queries to the wrong group, as I'm only
> interested in the *why* rather than the *how*. Conventional wisdom
> says put the swapfile on a separate partition, but I've not seen
> anyone explain why. My own surmise is that
> when on the OS partition it 'floats', changing size and position, so
> windows has to 'search' [don't you hate anthropomorphic words applied
> to a PC?] for it whereas, in a fixed location it 'knows' and can head
> straight there. But that's pure speculation on my part.

Why? Because the drive can only be accessed at a certain speed, a limited
speed. Putting the pagefile on a separate drive enables the drive that
contains the data to operate more quickly as the pagefile is being accessed
from a separate drive and the pagefile is accessed more frequently that the
disk so you'd want it on the faster drive if at all possible. Thus is is
faster because instead of the data transer slowing while accessing BOTH the
data and the pagefile it's able to grab them both from different drives at
the same time.

Putting the pagefile on a separate partition on the same drive isn't going
to do much other than keep the drive uncluttered. Deleting the pagefile is
just going to result in it being rebuilt. Turning off the page file, even
with a full 4 GB of RAM, is still not a great idea as some applications
"need" the pagefile, heck - at least one even creates it's own... It's a
good idea to set the max page file to 1 1/5 times the RAM you have and
putting it on another drive on a separate controller. That being said I'm
not too sure where "they" get the math that says that 1 1/5 times the RAM is
good...

Another 'tweak' is to set the pagefile at the same max and minimum values
thus eliminating the CPU cycles that would be used by Windows to determine
the usage. I've no proof that there's any benefit in that on today's
computers. The few cycles it uses on today's CPUs are likely to be
nanoseconds in length and probably not even noticed by even the most picky
among us.

Galen
--

"My mind rebels at stagnation. Give me problems, give me work, give me
the most abstruse cryptogram or the most intricate analysis, and I am
in my own proper atmosphere. I can dispense then with artificial
stimulants. But I abhor the dull routine of existence. I crave for
mental exaltation." -- Sherlock Holmes
 
G

Guest

Guest
Archived from groups: microsoft.public.windowsxp.help_and_support (More info?)

Galen,

Thanks again, but not convinced. Accessing the information *at the same
time* as the request is made is too much like having two people yelling
simultaneously into opposite ears. There must be a time lag for any
processing to take place. Whatever the merits of having the swap on a
separate disc I cannot see much of an improvement over the head moving
a hundredth of a millimetre to access an adjacent partition for its
information. But I could be wrong - I often am <g>

Haven't got around to creating a truly separate partition yet,
contenting myself with having a single pagefile of about 1.5GB (I have
1 GB of RAM) on an adjacent drive not likely to be otherwise accessed
by the OS during its fulminations. But a brief and unsuccessful foray
into LINUX demonstrated some of the virtues of a separate partition
although, of course, that OS is nowhere near as bloated as Windows
anyway. Had I been prepared to splash out the cash for new compatible
modem and printer to fully explore it I suspect I would not be here
asking questions.

Tony.
 

galen

Distinguished
May 24, 2004
1,879
0
19,780
Archived from groups: microsoft.public.windowsxp.help_and_support (More info?)

In news:1114629917.678679.142900@o13g2000cwo.googlegroups.com,
tonybocymru@evemail.net <tonybocymru@evemail.net> had this to say:

My reply is at the bottom of your sent message:

> Galen,
>
> Thanks again, but not convinced. Accessing the information *at the
> same time* as the request is made is too much like having two people
> yelling simultaneously into opposite ears. There must be a time lag
> for any processing to take place. Whatever the merits of having the
> swap on a separate disc I cannot see much of an improvement over the
> head moving a hundredth of a millimetre to access an adjacent
> partition for its information. But I could be wrong - I often am <g>
>
> Haven't got around to creating a truly separate partition yet,
> contenting myself with having a single pagefile of about 1.5GB (I have
> 1 GB of RAM) on an adjacent drive not likely to be otherwise accessed
> by the OS during its fulminations. But a brief and unsuccessful foray
> into LINUX demonstrated some of the virtues of a separate partition
> although, of course, that OS is nowhere near as bloated as Windows
> anyway. Had I been prepared to splash out the cash for new compatible
> modem and printer to fully explore it I suspect I would not be here
> asking questions.
>
> Tony.

Each IDE channel is limited to a certain amount of bandwidth while the CPU
is capable of processing much more than what's coming from a single channel.
Thus it's not really much like yelling at the same time it's more like
someone IMing you when you're talking on the telephone. Great improvement?
No, not really, that's just the reasoning behind the tweak. I personally
don't bother with it though I do put it on it's own specially made partition
just to keep the root drive nice and tidy. I'd move the hiberfil.sys to that
drive if I could only figure out a way.

Galen
--

"My mind rebels at stagnation. Give me problems, give me work, give me
the most abstruse cryptogram or the most intricate analysis, and I am
in my own proper atmosphere. I can dispense then with artificial
stimulants. But I abhor the dull routine of existence. I crave for
mental exaltation." -- Sherlock Holmes