Partion Magic - Slow when resizing partition

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

I use PM a fair amount and it is not too fast, I never looked into why
as it does the job well. Besides, for some reason, wanting one
partition, why delete the 8G partion? Now there is NO SYSTEM if it was
set up normally. I think it might have been quicker to use the "Merge
Partitions" option.
I personally like a smallish 1st partition as I image it often and
thus is a lot quicker, usually a few minutes. Also if somthing bad
happens to the system the entire drive is usually NOT lost, as most of
my drives are over 160G.

On Fri, 11 Feb 2005 09:00:03 +0000, Doughboy <anon@invalid.com> wrote:

>I'm fixing a friend's PC, and one of the things I'm doing is altering
>the partitions on his second HDD using Partition Magic 8 (DOS).
>
>The situation was that this HDD had two partitions (both NTFS), one
>Primary of about 8GB and the other Logical of about 30GB. He wanted
>this changed to just one 38GB partition.
>
>So I loaded up PM8 in DOS and deleted the first partition, then moved
>the free space into the Extended partition area, and then resized the
>Logical partition to use the free space in front (or to the left) of
>it.
>
>The first two operations take no time, but the resizing is incredibly
>slow. It had been running for over an hour and only reached 20% when I
>left. (I'm a bit annoyed with myself, 'cos after PM8 had started
>resizing and I saw how long it was going to take, I remembered that
>there were large files/folders that I could have safely deleted from
>Partition Two as we have the install discs, and much of the other data
>I could have probably temporarily moved to another drive. I am correct
>in assuming that the less actual data on the partition, the quicker
>the resize process will be, aren't I?)
>
>Anyway, I've got a couple of questions about PM8.
>
>a) whilst performing the resize operation, PM8 says something like
>'moving Partition down by 8GB and resizing to 38GB'. Shouldn't this be
>'moving Partition up by 8GB' as the only space is in front of the
>partition?.
>
>b) Couldn't it just leave the data on Partition Two where it is, and
>write a new FAT to the relevant place on the newly resized partition?
>If the FAT uses absolute values to point to files/folders this should
>be very easy, but if it uses relative values then it would involve
>examining each FAT entry, calculating the new relative value and
>writing this to where the FAT will be located once the partition has
>been resized. Even so, wouldn't this be quicker?
>
>c) is there any alternative to PM8 that can perform such functions a
>lot quicker, or are they inherently this slow?
>
>Doughboy
>Power corrupts. Absolute power is kind of neat.

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

Previously Doughboy <anon@invalid.com> wrote:
> I'm fixing a friend's PC, and one of the things I'm doing is altering
> the partitions on his second HDD using Partition Magic 8 (DOS).

> The situation was that this HDD had two partitions (both NTFS), one
> Primary of about 8GB and the other Logical of about 30GB. He wanted
> this changed to just one 38GB partition.

> So I loaded up PM8 in DOS and deleted the first partition, then moved
> the free space into the Extended partition area, and then resized the
> Logical partition to use the free space in front (or to the left) of
> it.

> The first two operations take no time, but the resizing is incredibly
> slow. It had been running for over an hour and only reached 20% when I
> left. (I'm a bit annoyed with myself, 'cos after PM8 had started
> resizing and I saw how long it was going to take, I remembered that
> there were large files/folders that I could have safely deleted from
> Partition Two as we have the install discs, and much of the other data
> I could have probably temporarily moved to another drive. I am correct
> in assuming that the less actual data on the partition, the quicker
> the resize process will be, aren't I?)

> Anyway, I've got a couple of questions about PM8.

> a) whilst performing the resize operation, PM8 says something like
> 'moving Partition down by 8GB and resizing to 38GB'. Shouldn't this be
> 'moving Partition up by 8GB' as the only space is in front of the
> partition?.

No, this is correct. Most resizing can only bo done at the end of
the partition (otherwise it gets far more complicated and risky),
so the partition is moved down, then extended at the end.

> b) Couldn't it just leave the data on Partition Two where it is, and
> write a new FAT to the relevant place on the newly resized partition?
> If the FAT uses absolute values to point to files/folders this should
> be very easy, but if it uses relative values then it would involve
> examining each FAT entry, calculating the new relative value and
> writing this to where the FAT will be located once the partition has
> been resized. Even so, wouldn't this be quicker?

Yes, but that results in a very strange layout and be more difficult.
PM is crash-prone as it is, adding more complex operations will not
help.

> c) is there any alternative to PM8 that can perform such functions a
> lot quicker, or are they inherently this slow?

They are slow. But my peersonal opinion is that PM takes much more
time than needed. I use GNU parted, which is free and boots from one
floppy (or as programm under Linux). It does not have a GUI and
it will do only one operation at a time, but I find it to work well
and relatively fast. It is a bit scary, because it does not ask
fro confirmation on operations. It does check whether they are
doable. Its homepage is here:

http://www.gnu.org/software/parted/parted.html

Arno
--
For email address: lastname AT tik DOT ee DOT ethz DOT ch
GnuPG: ID:1E25338F FP:0C30 5782 9D93 F785 E79C 0296 797F 6B50 1E25 338F
"The more corrupt the state, the more numerous the laws" - Tacitus

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

I've just been looking at PM8 under XP (PMagicNT.exe) and I'm curious
whether either of the two methods below would have been substantially
quicker.

a)
1) Delete Partition One (8GB Primary)
2) Convert Partition Two (30GB Logical in Extended Partition) to
Primary.
3) Resize 30GB Primary created in step 2 to 38GB.

or

b)
1) Convert Partition Two to Primary
2) Merge Partition Two with Partition One.


--
Power corrupts. Absolute power is kind of neat.

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

On Fri, 11 Feb 2005 10:59:44 GMT, IDIDIT <VISCH@EVASION.COM> wrote:

>
>I use PM a fair amount and it is not too fast, I never looked into why
>as it does the job well. Besides, for some reason, wanting one
>partition, why delete the 8G partion? Now there is NO SYSTEM if it was
>set up normally.

My friend has two HDD's in his PC and this is the second one which he
just uses for data. His OS is on the other drive.

>I personally like a smallish 1st partition as I image it often and
>thus is a lot quicker, usually a few minutes. Also if somthing bad
>happens to the system the entire drive is usually NOT lost, as most of
>my drives are over 160G.

Yeah I agree that it's nice to keep the OS partition as small as
possible because it makes it easier and quicker to Ghost and restore
it. For this reason, I keep my swapfile and all tempfiles on another
partition. However, my friend's not too computer savvy and he'll
probably just accept the defaults and install stuff to Program Files,
so I'd be asking for trouble if I make his C: partition too small.


--
Power corrupts. Absolute power is kind of neat.

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

On 11 Feb 2005 12:36:59 GMT, Arno Wagner <me@privacy.net> wrote:

>Previously Doughboy <anon@invalid.com> wrote:
>> a) whilst performing the resize operation, PM8 says something like
>> 'moving Partition down by 8GB and resizing to 38GB'. Shouldn't this be
>> 'moving Partition up by 8GB' as the only space is in front of the
>> partition?.
>
>No, this is correct. Most resizing can only bo done at the end of
>the partition (otherwise it gets far more complicated and risky),
>so the partition is moved down, then extended at the end.

Maybe I'm having a blond moment, but I still don't understand this. I
deleted Partition One and Partition Two occupied all the space on the
drive from the end of Partition One to the end of the drive, so where
exactly can it move the partition to?

>> b) Couldn't it just leave the data on Partition Two where it is, and
>> write a new FAT to the relevant place on the newly resized partition?
>> If the FAT uses absolute values to point to files/folders this should
>> be very easy, but if it uses relative values then it would involve
>> examining each FAT entry, calculating the new relative value and
>> writing this to where the FAT will be located once the partition has
>> been resized. Even so, wouldn't this be quicker?
>
>Yes, but that results in a very strange layout and be more difficult.
>PM is crash-prone as it is, adding more complex operations will not
>help.

I did think it might be a bit complicated having files dotted all over
the place (although wouldn't a subsequent defrag take care of this?)
However I do take your point about not making things any more
complicated than the already are  

>> c) is there any alternative to PM8 that can perform such functions a
>> lot quicker, or are they inherently this slow?
>
>They are slow. But my peersonal opinion is that PM takes much more
>time than needed. I use GNU parted, which is free and boots from one
>floppy (or as programm under Linux). It does not have a GUI and
>it will do only one operation at a time, but I find it to work well
>and relatively fast. It is a bit scary, because it does not ask
>fro confirmation on operations. It does check whether they are
>doable. Its homepage is here:
>
>http://www.gnu.org/software/parted/parted.html
>
>Arno

Thanks for the info, I'll check parted out (well, when i fix my floppy
drive anyway!)

Doughboy
--
Power corrupts. Absolute power is kind of neat.

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

> Maybe I'm having a blond moment, but I still don't understand this. I
> deleted Partition One and Partition Two occupied all the space on the
> drive from the end of Partition One to the end of the drive, so where
> exactly can it move the partition to?

1. primary 8 Gb - Logical 30 Gb (this is what you had)
2. Unallocated 8Gb - Logical 30 Gb (now move 'down' / to the left to get
to step 3)
3. Logical 30 Gb - Unallocated 8 Gb (now resize into unallocated space)
4. xxxxxxxxx Logical 38 Gb xxxxxxxx

--
Joep

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

On Fri, 11 Feb 2005 17:23:28 +0100, "Joep" <available@request.nl>
wrote:

>> Maybe I'm having a blond moment, but I still don't understand this. I
>> deleted Partition One and Partition Two occupied all the space on the
>> drive from the end of Partition One to the end of the drive, so where
>> exactly can it move the partition to?
>
>1. primary 8 Gb - Logical 30 Gb (this is what you had)
>2. Unallocated 8Gb - Logical 30 Gb (now move 'down' / to the left to get
>to step 3)
>3. Logical 30 Gb - Unallocated 8 Gb (now resize into unallocated space)
>4. xxxxxxxxx Logical 38 Gb xxxxxxxx

Ah, it all makes sense now. I was thinking 'move down' meant towards
the end of the drive (to the right).

Cheers for clearing that up for me.

Doughboy
--
Power corrupts. Absolute power is kind of neat.

Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

Previously Doughboy <anon@invalid.com> wrote:
> On Fri, 11 Feb 2005 17:23:28 +0100, "Joep" <available@request.nl>
> wrote:

>>> Maybe I'm having a blond moment, but I still don't understand this. I
>>> deleted Partition One and Partition Two occupied all the space on the
>>> drive from the end of Partition One to the end of the drive, so where
>>> exactly can it move the partition to?
>>
>>1. primary 8 Gb - Logical 30 Gb (this is what you had)
>>2. Unallocated 8Gb - Logical 30 Gb (now move 'down' / to the left to get
>>to step 3)
>>3. Logical 30 Gb - Unallocated 8 Gb (now resize into unallocated space)
>>4. xxxxxxxxx Logical 38 Gb xxxxxxxx

> Ah, it all makes sense now. I was thinking 'move down' meant towards
> the end of the drive (to the right).

Aha, of course. Never suspected that. The usual way to do memory
(also disk memory) maps is to put the beginning on the bottom
and the end on the top. But it is not standardised or anything.

> Cheers for clearing that up for me.

And cheers to you for actually understandig what happens and
only being a bit confused about terminology.

Arno
--
For email address: lastname AT tik DOT ee DOT ethz DOT ch
GnuPG: ID:1E25338F FP:0C30 5782 9D93 F785 E79C 0296 797F 6B50 1E25 338F
"The more corrupt the state, the more numerous the laws" - Tacitus