Post by Paul[]
Post by g***@aol.com I should do something similar, but the 'problem' is that when I get a
new system (laptop), the 'installation' (read: initialization) procedure
of the Windows (now Windows 11) on the disk [1] does not give the
choice/chance to partition the disk.
I haven't played with 11 much but doesn't it have "Disk Management"
that lets you juggle partitions? You should be able to shrink C: and
add another simple volume in that vacated space
Yes, you can do it after the fact, i.e. after the initial Windows
'installation'/initialization, but not before/during that procedure.
After the fact, you can indeed do it with the stock Disk Management
(which does have a 'Shrink Volume...' function) or with third-party
partitioning software, which is probably more flexible and user-friendly.
Maybe one day I'll do it, or maybe not! :-)
Yes, some feature for handling partitions has been built into the OS since 7.
In the 7 version, it was unable to shrink C: below a point because of
"immovable files", which were usually about half way up the original
size of C:, though (a) repeating it after a reboot _sometimes_ allowed
further reduction and (b) third-party software (e. g. EaseUS) did not
have this problem. I don't know if this limitation remained in the
in-OS partition utility in future versions.
The NTFS file system, places some sort of metadata at the 50% point on the disk.
In Linux, something there, used to place certain structures at the
1/3rd and 2/3rds points on their partitions.
These represent "value statements". The positions selected, have
something to do with file system performance. Someone believed,
long ago, that putting a certain structure at the 50% mark, gave
the best average performance.
I can see that, following certain perceptions, and with a disc with
files fairly well distributed (not necessarily fragmented), it could be
a perception that, on average, putting something that's accessed a lot
at the half-way point means the head will have less far to move from any
random position than if that something was near the start. I can't think
what the 1/3 and 2/3 justification is - maybe some combination of the
above with discs being faster near the start.
Such assumptions do assume very specific things about the way the data
on the disc is used - or, perhaps, assume very specifically that it is
random, which for any given user probably isn't quite the case.
Post by PaulYou will notice in the SSD era, this no longer matters, and the value
statements mean nothing now.
Indeed.
Post by PaulThus, the behavior of the utility is incorrect for SSDs, and
remains a value statement for HDD. If your seek time is zero,
then nobody cares where a structure is stored.
Just about any utility worth its salt, does not give a fig newton
(I've not come across that expression with the "newton" part in it
before.)
Post by Paulfor the 50% rule. The metadata gets moved, as required, at the
time a partition manipulation is carried out. I can use GParted
on Linux, and squeeze and squish all I want, and no "rules" are
applied to what I do.
Same with I think any of the third-party ones available to a 7 user (I
currently have EaseUS, though I think I only used it near the start of
my usage of this drive): within the practical limits, of course, of not
being able to squish below the size of the files currently on the
partition. (I don't know if it insists on a teeny bit extra for C: so
you don't leave yourself with an unusable system; I haven't tried, not
having any reason to.) They usually require a reboot (in practice two,
though the process is automated; I think the in-between case is not
normal Windows, so that they _can_ move the unmovable).
Post by PaulThe same goes for defragmenters like Raxco PerfectDisk.
Microsoft did not write the original defragmenter API. A third
party wrote it, and Microsoft bought it, and put it in the OS.
This is the feature that, say, JKDefrag would call, to move
a 64KB block or the like. The defragmenter API is "power-safe",
which means there is a reduced probability of destruction
if the power goes off, while you were using the defragmenter API.
It should not be using any cache path, neither System Read Cache
nor System Write Cache, are allowed to be used (if they did,
defragmentation would "fly like the wind").
In the same way that it's often far quicker to move (which is of course
really a copy then delete) all to another disc and then back, than to do
a defrag.
Post by PaulThe defragmenter API is busy, during a "disk shrink". Open the
Optimization dialog and open Disk Management, at the same time,
so both windows are open. Now, make a request to shrink a partition in DM.
What you should see, in the corner of your eye, is the
Optimize window wakes up and "something is going on". This is the
selective movement of materials. When DM shrinks a partition,
it calls Optimize to do the deed (move the materials out of the way).
I'm not going to play with my partition sizes anyway (I'm quite content
to take your word for the above), but where/what is the "Optimization
dialog" in Windows 7? Sounds like it might be an interesting thing.
Post by PaulBut even so, there are a number of things that certain OS utilities
will not request movement. Whereas the defrag API is technically
capable of doing a lot of things. There is a disconnect between
what Microsoft deems moveable, versus what the defrag API can
actually do.
Yes, I thought the "unmovable files" were system files or such. I
suppose what is "metadata" and what "system files" is just semantics.
Post by PaulNow, you have some idea, what ingredients fit into the "why won't
my partition shrink further in Disk Management" thing. It's not
a technical issue. It may be a value statement, such as
"put a drink coaster under that drink, you cretin".
We must have doilies and throw cushions, and a hassock you
cannot put your feet on. The hassock sits immobile in the
livingroom, serving no worldly purpose :-)
You see this thing ? Don't put your feet on it, or mom will be pissed.
Well, that's like the metadata at the 50% point. It's to remain
in the livingroom, even if it never demonstrates a function.
https://i.etsystatic.com/5275497/r/il/8cc62b/3822959334/il_794xN.3822959
334_33s3.jpg
(I have inherited something similar; mine is rectangular - about 2:1 -
whereas yours, hard to tell from the pic., looks as if it might be
square. I don't remember any direction not to sit on it [I don't think
anyone ever put their feet on it so that didn't come up].)
Post by PaulThe purpose of a hassock, as it turns out, it's a "land mine". If
an intruder enters the living room, they trip over the hassock
and are "dispatched" by the fall. That's why the hassock must
remain at 50%.
Paul
Whereas in the computing sense, it's just what you called a "value
statement", not a trap.
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)***@T+H+Sh0!:`)DNAf
Q. How much is 2 + 2?
A. Thank you so much for asking your question.
Are you still having this problem? I'll be delighted to help you. Please
restate the problem twice and include your Windows version along with
all error logs.
- Mayayana in alt.windows7.general, 2018-11-1
