DEFRAG A:/

  • Thread starter Thread starter Johann Bäcker
  • Start date Start date
J

Johann Bäcker

Unknown said:
No version of Windows ever defragged a floppy. There absolutely no need
for that.

____________________________________________

Windows for Workgroups 3.11 AND Windows95 IF you . . upgraded from WfW
3.11. You could also DEFRAG a:\ or c:\ from a . . . DOS Prompt on boot.
(Remember FAST working DOS?)

RTF files are retained on the floppy. Additions, deletions and text changes
are made therefore I - - - - DO NEED to defrag drive a:\.

Yes, I have been moving (batch) everything to the HD then back (batch again)
to the Floppy. Moving 195 files, one at a time would be . . . . . time
consuming.

It would appear that "Unknown" is making a decision for me.
 
Johann Bäcker said in news:[email protected]:
That is your opinion; sis.

Okay, stop the name calling. Think about what you are trying to do.
You are trying to arrange the files so their sectors are contiguous.
Other than using a defragger that has to use whatever freespace is
available to repeatedly move clusters around to get them contiguous, how
else might that be accomplished, especially with removable media?

(clock ticks away while you ponder)

So what do you think happened when you first copied the files onto a
newly formatted floppy?

(some more time ticks by)

Copy the files off the floppy into a temp folder on your hard drive,
format the floppy, and copy the files back. That will actually take
less time than a defragger slugging away with limited freespace trying
to move one cluster at a time. When you copy the files, they get copied
one at a time so each gets whatever is the next available cluster on the
target media. Since you just formatted it, the next available cluster
is immediately after the previously used cluster. Only if there was
very little defragmentation and also lots of freespace on the floppy
would defragging it be faster than moving the files (twice) with an
intervening format (because you wouldn't be moving unfragmented files).
Put the following in a batch (.bat) file, or just run the asterisked
commands (which are NOT to be included on the lines in the .bat file):

echo off
cls
echo Insert floppy into drive A: (hit a key when ready) ...
pause > nul
if exist c:\fd$temp rd /s /q c:\fd$temp
* md c:\fd$temp
if exist c:\fd$temp goto Continue
echo ERROR! Cannot create holding folder (c:\fd$temp) - ABORTED.
goto EndBatch
:Continue
echo Moving files off floppy ...
* xcopy /e a:\* c:\fd$temp > nul
echo Format the floppy ...
* format a: /q
echo Moving files back onto floppy ...
* xcopy /e c:\fd$temp\* a: > nul
* rd /s /q c:\fd$temp
:EndBatch
echo -DONE-

You take the marbles out of the wet sand, wipe the sand so it is level,
and then start poking the marbles back into the wet sand one color at a
time.
 
He said he already did that. Defragging a floppy makes it many times quicker. Of course editing RTF files on a floppy is asking for trouble, Word will crash one day throwing away your work..
 
David Candy said in news:[email protected]:
He said he already did that. Defragging a floppy makes it many times
quicker. Of course editing RTF files on a floppy is asking for
trouble, Word will crash one day throwing away your work..

Since xcopy is being used, they are not getting transferred "one at a
time". Xcopy uses a cache which can span reading/writing across
multiple files "at a time" per read/write action.

WD: How Word for Windows Uses Temporary Files
http://support.microsoft.com/?id=211632

According to this article, the temp files go under wherever is your temp
directory (i.e., %temp%). This would be on the hard disk under your
profile path. Word crashing would leave behind the ~<file> that Word
notes on restart as to whether or not you want to use it to recover you
changes made up to that point. However, as per the section titled
"Saved Files (Same Directory as the Saved File):", you will need as much
freespace on the disk (floppy in this case) as the file you are editing
(because there will be 2 copies of that file on the floppy when you go
to save it). I would think Word would simply report an error rather
than trash the floppy (i.e., "not enough space" error).

The gotcha is when you cut and paste within Word for a doc on a floppy.
Every one creates a temp file on that floppy. I've seen this when
editing a .doc file on the desktop. While editing the doc, I'll see
lots of these temp files appear on the desktop and increase in number as
I continue doing cut-paste operations within the document. Word also
has an option to save a backup of your file every N minutes, so you
might need as much freespace as twice the size of the file you are
editing, one for the backup and one for the new tilde-named file before
Word then deletes the original file along with the backup.

As I recall, there can also be problems with a link to a file that you
edited on a floppy that would then be included in the history portion of
the Start -> Documents menu, like the floppy getting accessed (and an
access error if there is no floppy in the drive pointed to by the MRU
link to the file).

The smart way is to copy the file from the floppy onto the hard drive
and edit that copy. You can use Save As to save your changes back
directly onto the floppy (but then the hard disk copy doesn't get
updated unless you follow with a Save). "The creation of temporary
files in the working folder, is a principle reason why Word documents
corrupt when loaded from or saved to floppy disc. If there is
insufficient space to accommodate the required temporary files, the
document will be corrupted and you may not be able to recover it. Use
floppy to transport documents, but the hard drive for opening and saving
files." (http://snipurl.com/6mhd). Users yanking out the floppy before
closing Word is another source of corruption.
 
Defragging 1.44 megabytes makes a floppy many times quicker?? Absolute
hogwash.
He said he already did that. Defragging a floppy makes it many times quicker.
Of course editing RTF files on a floppy is asking for trouble, Word will crash
one day throwing away your work..
 
Shows how much you know. Obviously you're a bit young to remember floppy only machines. Disk caching negates a lot of the effects of fragmenting. Dos had two caches, Buffers and smartdrv. But seeking is slow on floppies and each fragment incurs a substantial time delay (though the floppy spins for 2 seconds after last access in case you want to read more to overcome latency).

So saving a 40K word processing file takes 40 clusters. This could be 40 seeks.

A floppy seek time is about 100mS. It spins at 360rpm giving a high latency. So just positioing the heads could take a max of 4 secs compared to 0.1 for a contigous file, though track to track seek would change the ratio. This gets even slower if there are no buffers or caching as the fat needs to be reread (fats are kept in memory in Windows).
 
You showed your ignorance. Floppies do not have clusters. They have tracks and
sectors.
Shows how much you know. Obviously you're a bit young to remember floppy only
machines. Disk caching negates a lot of the effects of fragmenting. Dos had
two caches, Buffers and smartdrv. But seeking is slow on floppies and each
fragment incurs a substantial time delay (though the floppy spins for 2
seconds after last access in case you want to read more to overcome latency).

So saving a 40K word processing file takes 40 clusters. This could be 40
seeks.

A floppy seek time is about 100mS. It spins at 360rpm giving a high latency.
So just positioing the heads could take a max of 4 secs compared to 0.1 for a
contigous file, though track to track seek would change the ratio. This gets
even slower if there are no buffers or caching as the fat needs to be reread
(fats are kept in memory in Windows).
 
Floppies have one or two sectors (360K, 720K, 2.88Mb) per cluster.Floppies differ from hard disk in that they don't have a MBR, only a Dos Boot Sector.
 
Unknown said in news:[email protected]:
You showed your ignorance. Floppies do not have clusters. They have
tracks and sectors.

If you don't know, don't spew. You showed YOUR ignorance. Go read the
following:

FAT Type and Cluster Size Depends on Logical Drive Size
http://support.microsoft.com/?id=67321

The Default Cluster Size for the NTFS and FAT File Systems
http://support.microsoft.com/?id=314878

The PC Guide: Summary of Floppy Disk Types and Specifications
http://www.pcguide.com/ref/fdd/format.htm

In the case of a 1.44MB, 3.5-inch floppy diskette, 1 cluster = 1 sector.
That is not true for other capacity diskettes and drives. So, yeah, you
got lucky despite your ignorance that the allocation on this type of
floppy for a cluster is the same size as a sector. It certainly doesn't
prove you knew anything about what you claimed to know, especially since
you generalized your claim for all floppies. Ah, the joy of ignorance,
the lack of complexity, and the ease it affords while awaiting death.

"It is wiser to remain silent and let others ponder your ignorance than
to open your mouth and remove all doubt. If you don't know, don't
profess to know."
 
This is from the programmer of MSDos ver 1 explaining sectors/clusters. So the very first Dos (8" [200mm] 4 x 128 b sectors = 1 cluster

Now take a standard 8-inch single-density floppy disk that has 2002 sectors of 128 bytes. To minimize internal fragmentation, choose the smallest possible allocation-unit size of one sector. Two thousand allocation units will require 3000 bytes (24 sectors) per FAT, or 48 sectors for two FATs. If the average file size is 16K bytes (128 sectors), the disk will be full when there are 16 files on it. Waste due to internal fragmentation would be

16 files x 64 bytes per file = 1024 bytes (8 sectors)

Far more space is occupied by the FATs on the disk than is wasted by internal fragmentation!

To provide maximum usable data space on the disk, both internal fragmentation and FAT size must be considered because both consume data area. The standard MS-DOS format for 8-inch single-density disks strikes a balance by using four sectors per allocation unit.
 
Back
Top