1. cluster size
Does anyone know how I can programmatically find out what is the minimum
cluster size used on a drive? I'm under the impression that it can vary,
since a small file (little bigger than a Kb) on one drive takes up 4096
bytes, and the same file on another drive takes up 8192 bytes. Both drives
indicate FAT32 from fdisk.
And am I right in thinking that it is cluster size rather than simply file
size that relates to how much of a drive is used up? In other words, if you
have a whole bunch of small files, each actually takes up (at least) one
cluster, so more of your drive may be used up than would seem so from just
adding up file sizes?
Dan Moyer
2. Re: cluster size
On 26 Jan 2002, at 21:15, Dan Moyer wrote:
>
> Does anyone know how I can programmatically find out what is the minimum
> cluster size used on a drive? I'm under the impression that it can vary,
> since a small file (little bigger than a Kb) on one drive takes up 4096
> bytes, and the same file on another drive takes up 8192 bytes. Both drives
> indicate FAT32 from fdisk.
I think Euman did a include to get all the drive stats.
> And am I right in thinking that it is cluster size rather than simply file
> size that relates to how much of a drive is used up? In other words, if you
> have a whole bunch of small files, each actually takes up (at least) one
> cluster, so more of your drive may be used up than would seem so from just
> adding up file sizes?
Correct. The dos 8.3 filename takes up one slot, 9.3 takes up one more
slot, and each 7chars over that will count as another filename slot too under
fat32.
Kat
3. Re: cluster size
- Posted by vern at lvp.eastlink.ca
Jan 27, 2002
----- Original Message -----
From: "Kat" <gertie at PELL.NET>
To: "EUforum" <EUforum at topica.com>
Subject: Re: cluster size
>
> On 26 Jan 2002, at 21:15, Dan Moyer wrote:
>
> >
> > Does anyone know how I can programmatically find out what is the minimum
> > cluster size used on a drive? I'm under the impression that it can
vary,
> > since a small file (little bigger than a Kb) on one drive takes up 4096
> > bytes, and the same file on another drive takes up 8192 bytes. Both
drives
> > indicate FAT32 from fdisk.
>
> I think Euman did a include to get all the drive stats.
>
> > And am I right in thinking that it is cluster size rather than simply
file
> > size that relates to how much of a drive is used up? In other words, if
you
> > have a whole bunch of small files, each actually takes up (at least) one
> > cluster, so more of your drive may be used up than would seem so from
just
> > adding up file sizes?
>
> Correct. The dos 8.3 filename takes up one slot, 9.3 takes up one more
> slot, and each 7chars over that will count as another filename slot too
under
> fat32.
>
> Kat
>
>
One reason, I believe, is that the 1st FAT32 drive is the native drive and
the second drive is a FAT32b partition (extended DOS and Logical Drive), the
extra size being additional logical making, an unfortunate side effect of
DOS' single sized cluster size, fortunately the solution is to install a
Unix os and format cluster at 1024 :)
>
>
>
4. Re: cluster size
On 27 Jan 2002, at 8:09, vern at lvp.eastlink.ca wrote:
>
>
> ----- Original Message -----
> From: "Kat" <gertie at PELL.NET>
> To: "EUforum" <EUforum at topica.com>
> Sent: Sunday, January 27, 2002 2:04 AM
> Subject: Re: cluster size
>
>
> > On 26 Jan 2002, at 21:15, Dan Moyer wrote:
> >
> > >
> > > Does anyone know how I can programmatically find out what is the minimum
> > > cluster size used on a drive? I'm under the impression that it can
> vary,
> > > since a small file (little bigger than a Kb) on one drive takes up 4096
> > > bytes, and the same file on another drive takes up 8192 bytes. Both
> drives
> > > indicate FAT32 from fdisk.
> >
> > I think Euman did a include to get all the drive stats.
> >
> > > And am I right in thinking that it is cluster size rather than simply
> file
> > > size that relates to how much of a drive is used up? In other words, if
> you
> > > have a whole bunch of small files, each actually takes up (at least) one
> > > cluster, so more of your drive may be used up than would seem so from
> just
> > > adding up file sizes?
> >
> > Correct. The dos 8.3 filename takes up one slot, 9.3 takes up one more
> > slot, and each 7chars over that will count as another filename slot too
> under
> > fat32.
> >
> > Kat
> >
> >
> One reason, I believe, is that the 1st FAT32 drive is the native drive and
> the second drive is a FAT32b partition (extended DOS and Logical Drive), the
But that is not what he said, he said 2nd drive. I have more than one drive in
all my computers.
Kat
> extra size being additional logical making, an unfortunate side effect of DOS'
> single sized cluster size, fortunately the solution is to install a Unix os
> and
> format cluster at 1024 :)
>
> >
>
>
>
5. Re: cluster size
It doesn't matter what patition the drive is on. It depends on what the
cluster size is set to when it is formatted. The drive is seperated into
blocks of 512 bytes. When it is changed to FAT32, you can set it to 8, 16,
32, 64 or 128 blocks per cluster. When you create a file, it takes up at
least one of these clusters. The filename does take up space in the FAT
tables, but that is pre-allocated. In Win95-OSR2 and Win98 there is a
program that you can use to change from FAT16 to FAT32 and you can choose
the cluster size. Unfortunately, you ca't change it later on unless you
reformat the drive or use non-destructive methods.
>From: Kat <gertie at PELL.NET>
>Reply-To: EUforum at topica.com
>To: EUforum <EUforum at topica.com>
>Subject: Re: cluster size
>Date: Sun, 27 Jan 2002 14:06:43 -0600
>
>
>On 27 Jan 2002, at 8:09, vern at lvp.eastlink.ca wrote:
>
> >
> > ----- Original Message -----
> > From: "Kat" <gertie at PELL.NET>
> > To: "EUforum" <EUforum at topica.com>
> > Sent: Sunday, January 27, 2002 2:04 AM
> > Subject: Re: cluster size
> >
> >
> > > On 26 Jan 2002, at 21:15, Dan Moyer wrote:
> > >
> > > >
> > > > Does anyone know how I can programmatically find out what is the
>minimum
> > > > cluster size used on a drive? I'm under the impression that it can
> > vary,
> > > > since a small file (little bigger than a Kb) on one drive takes up
>4096
> > > > bytes, and the same file on another drive takes up 8192 bytes. Both
> > drives
> > > > indicate FAT32 from fdisk.
> > >
> > > I think Euman did a include to get all the drive stats.
> > >
> > > > And am I right in thinking that it is cluster size rather than
>simply
> > file
> > > > size that relates to how much of a drive is used up? In other
>words, if
> > you
> > > > have a whole bunch of small files, each actually takes up (at least)
>one
> > > > cluster, so more of your drive may be used up than would seem so
>from
> > just
> > > > adding up file sizes?
> > >
> > > Correct. The dos 8.3 filename takes up one slot, 9.3 takes up one
>more
> > > slot, and each 7chars over that will count as another filename slot
>too
> > under
> > > fat32.
> > >
> > > Kat
> > >
> > >
> > One reason, I believe, is that the 1st FAT32 drive is the native drive
>and
> > the second drive is a FAT32b partition (extended DOS and Logical Drive),
>the
>
>But that is not what he said, he said 2nd drive. I have more than one drive
>in
>all my computers.
>
>Kat
>
>
> > extra size being additional logical making, an unfortunate side effect
>of DOS'
> > single sized cluster size, fortunately the solution is to install a Unix
>os and
> > format cluster at 1024 :)
> >
> > >
>
>
>
6. Re: cluster size
On 27 Jan 2002, at 21:00, Elliott Sales de Andrade wrote:
>
> It doesn't matter what patition the drive is on.
But you said:
> > > One reason, I believe, is that the 1st FAT32 drive is the native drive
> >and
> > > the second drive is a FAT32b partition (extended DOS and Logical Drive),
Which as you now say, has nothing to do with the question or the
explaination. Especially since the original question was about 2 different
drives, not two partitions.
Kat
> It depends on what the
> cluster size is set to when it is formatted. The drive is seperated into
> blocks of 512 bytes. When it is changed to FAT32, you can set it to 8, 16, 32,
> 64 or 128 blocks per cluster. When you create a file, it takes up at least one
> of these clusters. The filename does take up space in the FAT tables, but that
> is pre-allocated. In Win95-OSR2 and Win98 there is a program that you can use
> to
> change from FAT16 to FAT32 and you can choose the cluster size. Unfortunately,
> you ca't change it later on unless you reformat the drive or use
> non-destructive
> methods.
>
>
> >From: Kat <gertie at PELL.NET>
> >Reply-To: EUforum at topica.com
> >To: EUforum <EUforum at topica.com>
> >Subject: Re: cluster size
> >Date: Sun, 27 Jan 2002 14:06:43 -0600
> >
> >
> >On 27 Jan 2002, at 8:09, vern at lvp.eastlink.ca wrote:
> >
> > >
> > > ----- Original Message -----
> > > From: "Kat" <gertie at PELL.NET>
> > > To: "EUforum" <EUforum at topica.com>
> > > Sent: Sunday, January 27, 2002 2:04 AM
> > > Subject: Re: cluster size
> > >
> > >
> > > > On 26 Jan 2002, at 21:15, Dan Moyer wrote:
> > > >
> > > > >
> > > > > Does anyone know how I can programmatically find out what is the
> >minimum
> > > > > cluster size used on a drive? I'm under the impression that it can
> > > vary,
> > > > > since a small file (little bigger than a Kb) on one drive takes up
> >4096
> > > > > bytes, and the same file on another drive takes up 8192 bytes. Both
> > > drives
> > > > > indicate FAT32 from fdisk.
> > > >
> > > > I think Euman did a include to get all the drive stats.
> > > >
> > > > > And am I right in thinking that it is cluster size rather than
> >simply
> > > file
> > > > > size that relates to how much of a drive is used up? In other
> >words, if
> > > you
> > > > > have a whole bunch of small files, each actually takes up (at least)
> >one
> > > > > cluster, so more of your drive may be used up than would seem so
> >from
> > > just
> > > > > adding up file sizes?
> > > >
> > > > Correct. The dos 8.3 filename takes up one slot, 9.3 takes up one
> >more
> > > > slot, and each 7chars over that will count as another filename slot
> >too
> > > under
> > > > fat32.
> > > >
> > > > Kat
> > > >
> > > >
> > > One reason, I believe, is that the 1st FAT32 drive is the native drive
> >and
> > > the second drive is a FAT32b partition (extended DOS and Logical Drive),
> >the
> >
> >But that is not what he said, he said 2nd drive. I have more than one drive
> >in
> >all my computers.
> >
> >Kat
> >
> >
> > > extra size being additional logical making, an unfortunate side effect
> >of DOS'
> > > single sized cluster size, fortunately the solution is to install a Unix
> >os and
> > > format cluster at 1024 :)
> > >
> > > >
>
>
>
7. Re: cluster size
>>It doesn't matter what patition the drive is on.
>
>But you said:
>
>>>>One reason, I believe, is that the 1st FAT32 drive is the native
>>>> >>>>drive and the second drive is a FAT32b partition (extended DOS and
>>>> >>>>Logical Drive),
I never said that. That message was from vern at lvp.eastlink.ca.
>From: Kat <gertie at PELL.NET>
>Reply-To: EUforum at topica.com
>To: EUforum <EUforum at topica.com>
>Subject: Re: cluster size
>Date: Sun, 27 Jan 2002 20:12:19 -0600
>
>
>On 27 Jan 2002, at 21:00, Elliott Sales de Andrade wrote:
>
> >
> > It doesn't matter what patition the drive is on.
>
>But you said:
>
> > > > One reason, I believe, is that the 1st FAT32 drive is the native
>drive
> > >and
> > > > the second drive is a FAT32b partition (extended DOS and Logical
>Drive),
>
>Which as you now say, has nothing to do with the question or the
>explaination. Especially since the original question was about 2 different
>drives, not two partitions.
>
>Kat
>
> > It depends on what the
> > cluster size is set to when it is formatted. The drive is seperated into
> > blocks of 512 bytes. When it is changed to FAT32, you can set it to 8,
>16, 32,
> > 64 or 128 blocks per cluster. When you create a file, it takes up at
>least one
> > of these clusters. The filename does take up space in the FAT tables,
>but that
> > is pre-allocated. In Win95-OSR2 and Win98 there is a program that you
>can use to
> > change from FAT16 to FAT32 and you can choose the cluster size.
>Unfortunately,
> > you ca't change it later on unless you reformat the drive or use
>non-destructive
> > methods.
> >
> >
> > >From: Kat <gertie at PELL.NET>
> > >Reply-To: EUforum at topica.com
> > >To: EUforum <EUforum at topica.com>
> > >Subject: Re: cluster size
> > >Date: Sun, 27 Jan 2002 14:06:43 -0600
> > >
> > >
> > >On 27 Jan 2002, at 8:09, vern at lvp.eastlink.ca wrote:
> > >
> > > >
> > > > ----- Original Message -----
> > > > From: "Kat" <gertie at PELL.NET>
> > > > To: "EUforum" <EUforum at topica.com>
> > > > Sent: Sunday, January 27, 2002 2:04 AM
> > > > Subject: Re: cluster size
> > > >
> > > >
> > > > > On 26 Jan 2002, at 21:15, Dan Moyer wrote:
> > > > >
> > > > > >
> > > > > > Does anyone know how I can programmatically find out what is the
> > >minimum
> > > > > > cluster size used on a drive? I'm under the impression that it
>can
> > > > vary,
> > > > > > since a small file (little bigger than a Kb) on one drive takes
>up
> > >4096
> > > > > > bytes, and the same file on another drive takes up 8192 bytes.
>Both
> > > > drives
> > > > > > indicate FAT32 from fdisk.
> > > > >
> > > > > I think Euman did a include to get all the drive stats.
> > > > >
> > > > > > And am I right in thinking that it is cluster size rather than
> > >simply
> > > > file
> > > > > > size that relates to how much of a drive is used up? In other
> > >words, if
> > > > you
> > > > > > have a whole bunch of small files, each actually takes up (at
>least)
> > >one
> > > > > > cluster, so more of your drive may be used up than would seem so
> > >from
> > > > just
> > > > > > adding up file sizes?
> > > > >
> > > > > Correct. The dos 8.3 filename takes up one slot, 9.3 takes up one
> > >more
> > > > > slot, and each 7chars over that will count as another filename
>slot
> > >too
> > > > under
> > > > > fat32.
> > > > >
> > > > > Kat
<snip>
>
>
>
8. Re: cluster size
Sorry, I didn't explain in enough detail. Although I was speaking in
particular about two different drives, the second one *is* partitioned,
which from Vern & Elliott's explanations now seems to be the cause of the
different minimum cluster sizes.
But independent of the cause, & unrelated to how to *change* the minimum
cluster size, what I need to be able to do is *programmatically* find out
*what* the minimum cluster size on any specified drive in any users system
is.
Reason is, I'm trying to finish up a "what all's taking up so much space on
my drive!" utility I had started once before & couldn't get to work quite
right, which now does work except for knowing the actual minimum cluster
size for any drive.
Dan Moyer
----- Original Message -----
From: "Kat" <gertie at PELL.NET>
To: "EUforum" <EUforum at topica.com>
Sent: Sunday, January 27, 2002 6:12 PM
Subject: Re: cluster size
>
> On 27 Jan 2002, at 21:00, Elliott Sales de Andrade wrote:
>
> >
> > It doesn't matter what patition the drive is on.
>
> But you said:
>
> > > > One reason, I believe, is that the 1st FAT32 drive is the native
drive
> > >and
> > > > the second drive is a FAT32b partition (extended DOS and Logical
Drive),
>
> Which as you now say, has nothing to do with the question or the
> explaination. Especially since the original question was about 2 different
> drives, not two partitions.
>
> Kat
>
> > It depends on what the
> > cluster size is set to when it is formatted. The drive is seperated into
> > blocks of 512 bytes. When it is changed to FAT32, you can set it to 8,
16, 32,
> > 64 or 128 blocks per cluster. When you create a file, it takes up at
least one
> > of these clusters. The filename does take up space in the FAT tables,
but that
> > is pre-allocated. In Win95-OSR2 and Win98 there is a program that you
can use to
> > change from FAT16 to FAT32 and you can choose the cluster size.
Unfortunately,
> > you ca't change it later on unless you reformat the drive or use
non-destructive
> > methods.
> >
> >
> > >From: Kat <gertie at PELL.NET>
> > >Reply-To: EUforum at topica.com
> > >To: EUforum <EUforum at topica.com>
> > >Subject: Re: cluster size
> > >Date: Sun, 27 Jan 2002 14:06:43 -0600
> > >
> > >
> > >On 27 Jan 2002, at 8:09, vern at lvp.eastlink.ca wrote:
> > >
> > > >
> > > > ----- Original Message -----
> > > > From: "Kat" <gertie at PELL.NET>
> > > > To: "EUforum" <EUforum at topica.com>
> > > > Sent: Sunday, January 27, 2002 2:04 AM
> > > > Subject: Re: cluster size
> > > >
> > > >
> > > > > On 26 Jan 2002, at 21:15, Dan Moyer wrote:
> > > > >
> > > > > >
> > > > > > Does anyone know how I can programmatically find out what is the
> > >minimum
> > > > > > cluster size used on a drive? I'm under the impression that it
can
> > > > vary,
> > > > > > since a small file (little bigger than a Kb) on one drive takes
up
> > >4096
> > > > > > bytes, and the same file on another drive takes up 8192 bytes.
Both
> > > > drives
> > > > > > indicate FAT32 from fdisk.
> > > > >
> > > > > I think Euman did a include to get all the drive stats.
> > > > >
> > > > > > And am I right in thinking that it is cluster size rather than
> > >simply
> > > > file
> > > > > > size that relates to how much of a drive is used up? In other
> > >words, if
> > > > you
> > > > > > have a whole bunch of small files, each actually takes up (at
least)
> > >one
> > > > > > cluster, so more of your drive may be used up than would seem so
> > >from
> > > > just
> > > > > > adding up file sizes?
> > > > >
> > > > > Correct. The dos 8.3 filename takes up one slot, 9.3 takes up one
> > >more
> > > > > slot, and each 7chars over that will count as another filename
slot
> > >too
> > > > under
> > > > > fat32.
> > > > >
> > > > > Kat
> > > > >
> > > > >
> > > > One reason, I believe, is that the 1st FAT32 drive is the native
drive
> > >and
> > > > the second drive is a FAT32b partition (extended DOS and Logical
Drive),
> > >the
> > >
> > >But that is not what he said, he said 2nd drive. I have more than one
drive in
> > >all my computers.
> > >
> > >Kat
> > >
> > >
> > > > extra size being additional logical making, an unfortunate side
effect
> > >of DOS'
> > > > single sized cluster size, fortunately the solution is to install a
Unix
> > >os and
> > > > format cluster at 1024 :)
> > > >
> > > > >
>
>
>
9. Re: cluster size
On 27 Jan 2002, at 20:52, Dan Moyer wrote:
>
> Sorry, I didn't explain in enough detail. Although I was speaking in
> particular about two different drives, the second one *is* partitioned,
> which from Vern & Elliott's explanations now seems to be the cause of the
> different minimum cluster sizes.
Ok, sorry, i didn't understand.
> But independent of the cause, & unrelated to how to *change* the minimum
> cluster size, what I need to be able to do is *programmatically* find out
> *what* the minimum cluster size on any specified drive in any users system
> is.
>
> Reason is, I'm trying to finish up a "what all's taking up so much space on my
> drive!" utility I had started once before & couldn't get to work quite right,
> which now does work except for knowing the actual minimum cluster size for any
> drive.
Ok, look to the end of function disk_free_space() in Craig Gilbert's diskutil.e
in http://www.rapideuphoria.com/diskutil.zip It's dos. Also, look in function
GetDiskFreeSpace() in win32lib v0.15c.
Kat
10. Re: cluster size
Thanks Kat,
But now I'm running into a problem: when I modify Craig Gilbert's
disk_free_space() to return cluster_size (took me an embarrassingly long
time to figure out it was just bytes_per_sector * sectors_per_cluster!), I
get a return of 32K, for both a 3gig drive and a logical drive of 10gig,
which actually seem to have 4 & 8k cluster sizes respectively. Since FAT16
will format for 32K per cluster for anything above 1 or 2 gig (I think?),
I'm wondering if maybe Craig's function is not correct for FAT32? His demo
does discern that my drives are FAT32, but the free space sizes computed are
maxed at 2gig, not what they really are, which seems to further suggest not
compatible with FAT32.
Any thoughts?
Dan Moyer
----- Original Message -----
From: "Kat" <gertie at PELL.NET>
To: "EUforum" <EUforum at topica.com>
Sent: Sunday, January 27, 2002 10:05 PM
Subject: Re: cluster size
>
> On 27 Jan 2002, at 20:52, Dan Moyer wrote:
>
> >
> > Sorry, I didn't explain in enough detail. Although I was speaking in
> > particular about two different drives, the second one *is* partitioned,
> > which from Vern & Elliott's explanations now seems to be the cause of
the
> > different minimum cluster sizes.
>
> Ok, sorry, i didn't understand.
>
> > But independent of the cause, & unrelated to how to *change* the minimum
> > cluster size, what I need to be able to do is *programmatically* find
out
> > *what* the minimum cluster size on any specified drive in any users
system
> > is.
> >
> > Reason is, I'm trying to finish up a "what all's taking up so much space
on my
> > drive!" utility I had started once before & couldn't get to work quite
right,
> > which now does work except for knowing the actual minimum cluster size
for any
> > drive.
>
> Ok, look to the end of function disk_free_space() in Craig Gilbert's
diskutil.e
> in http://www.rapideuphoria.com/diskutil.zip It's dos. Also, look in
function
> GetDiskFreeSpace() in win32lib v0.15c.
>
> Kat
>
>
>
>
11. Re: cluster size
On 28 Jan 2002, at 20:19, Dan Moyer wrote:
>
> Thanks Kat,
>
> But now I'm running into a problem: when I modify Craig Gilbert's
> disk_free_space() to return cluster_size (took me an embarrassingly long
> time to figure out it was just bytes_per_sector * sectors_per_cluster!), I
> get a return of 32K, for both a 3gig drive and a logical drive of 10gig,
> which actually seem to have 4 & 8k cluster sizes respectively. Since FAT16
> will
> format for 32K per cluster for anything above 1 or 2 gig (I think?), I'm
> wondering if maybe Craig's function is not correct for FAT32? His demo does
> discern that my drives are FAT32, but the free space sizes computed are maxed
> at
> 2gig, not what they really are, which seems to further suggest not compatible
> with FAT32.
>
> Any thoughts?
Yes, Craig's utility is dos. As you know, dos is limited to the 2Gig size. The
other file i mentioned uses the windows api thru win32lib, and should return
good data on the big drives. YMMV tho. What i didn't check is if the newer
win32libs have the same function as the 0.15c version.
Kat
> Dan Moyer
>
>
> ----- Original Message -----
> From: "Kat" <gertie at PELL.NET>
> To: "EUforum" <EUforum at topica.com>
> Sent: Sunday, January 27, 2002 10:05 PM
> Subject: Re: cluster size
>
>
> > On 27 Jan 2002, at 20:52, Dan Moyer wrote:
> >
> > >
> > > Sorry, I didn't explain in enough detail. Although I was speaking in
> > > particular about two different drives, the second one *is* partitioned,
> > > which from Vern & Elliott's explanations now seems to be the cause of
> the
> > > different minimum cluster sizes.
> >
> > Ok, sorry, i didn't understand.
> >
> > > But independent of the cause, & unrelated to how to *change* the minimum
> > > cluster size, what I need to be able to do is *programmatically* find
> out
> > > *what* the minimum cluster size on any specified drive in any users
> system
> > > is.
> > >
> > > Reason is, I'm trying to finish up a "what all's taking up so much space
> on my
> > > drive!" utility I had started once before & couldn't get to work quite
> right,
> > > which now does work except for knowing the actual minimum cluster size
> for any
> > > drive.
> >
> > Ok, look to the end of function disk_free_space() in Craig Gilbert's
> diskutil.e
> > in http://www.rapideuphoria.com/diskutil.zip It's dos. Also, look in
> function
> > GetDiskFreeSpace() in win32lib v0.15c.
> >
> > Kat
> >
> >
>
>
>
12. Re: cluster size
Kat,
Oh, ok, I forgot or didn't realize dos was limited to 2gig, & didn't find
win32lib v0.15c on my drive, & GetDiskFreeSpace wasn't in the later
versions, but maybe there is something similar (or I missed it). I'll keep
looking, thanks.
Dan
----- Original Message -----
From: "Kat" <gertie at PELL.NET>
To: "EUforum" <EUforum at topica.com>
Sent: Monday, January 28, 2002 9:08 PM
Subject: Re: cluster size
>
> On 28 Jan 2002, at 20:19, Dan Moyer wrote:
>
> >
> > Thanks Kat,
> >
> > But now I'm running into a problem: when I modify Craig Gilbert's
> > disk_free_space() to return cluster_size (took me an embarrassingly long
> > time to figure out it was just bytes_per_sector * sectors_per_cluster!),
I
> > get a return of 32K, for both a 3gig drive and a logical drive of 10gig,
> > which actually seem to have 4 & 8k cluster sizes respectively. Since
FAT16 will
> > format for 32K per cluster for anything above 1 or 2 gig (I think?), I'm
> > wondering if maybe Craig's function is not correct for FAT32? His demo
does
> > discern that my drives are FAT32, but the free space sizes computed are
maxed at
> > 2gig, not what they really are, which seems to further suggest not
compatible
> > with FAT32.
> >
> > Any thoughts?
>
> Yes, Craig's utility is dos. As you know, dos is limited to the 2Gig size.
The
> other file i mentioned uses the windows api thru win32lib, and should
return
> good data on the big drives. YMMV tho. What i didn't check is if the newer
> win32libs have the same function as the 0.15c version.
>
> Kat
>
> > Dan Moyer
> >
> >
> > ----- Original Message -----
> > From: "Kat" <gertie at PELL.NET>
> > To: "EUforum" <EUforum at topica.com>
> > Sent: Sunday, January 27, 2002 10:05 PM
> > Subject: Re: cluster size
> >
> >
> > > On 27 Jan 2002, at 20:52, Dan Moyer wrote:
> > >
> > > >
> > > > Sorry, I didn't explain in enough detail. Although I was speaking
in
> > > > particular about two different drives, the second one *is*
partitioned,
> > > > which from Vern & Elliott's explanations now seems to be the cause
of
> > the
> > > > different minimum cluster sizes.
> > >
> > > Ok, sorry, i didn't understand.
> > >
> > > > But independent of the cause, & unrelated to how to *change* the
minimum
> > > > cluster size, what I need to be able to do is *programmatically*
find
> > out
> > > > *what* the minimum cluster size on any specified drive in any users
> > system
> > > > is.
> > > >
> > > > Reason is, I'm trying to finish up a "what all's taking up so much
space
> > on my
> > > > drive!" utility I had started once before & couldn't get to work
quite
> > right,
> > > > which now does work except for knowing the actual minimum cluster
size
> > for any
> > > > drive.
> > >
> > > Ok, look to the end of function disk_free_space() in Craig Gilbert's
> > diskutil.e
> > > in http://www.rapideuphoria.com/diskutil.zip It's dos. Also, look in
> > function
> > > GetDiskFreeSpace() in win32lib v0.15c.
> > >
> > > Kat
> > >
> > >
>
>
>
