Time: Tue Jun 03 23:56:23 1997
	by primenet.com (8.8.5/8.8.5) with SMTP id WAA18820
	for [address in tool bar]; Tue, 3 Jun 1997 22:20:42 -0700 (MST)
Delivered-To: liberty-and-justice-outgoing@majordomo.pobox.com
Date: Tue, 03 Jun 1997 23:55:19 -0700
To: liberty-and-justice@pobox.com
From: Paul Andrew Mitchell [address in tool bar]
Subject: L&J: SLF: FAT16 problem: demonstration and proof
References: <3.0.1.16.19970603041137.2b5f5d34@pop.primenet.com>

Grier et al., 

I must confess to having made an 
important error here.  On a typical
Western Digital 2 GB drive, the
maximum cluster size is 32 KB,
not 64 KB.  I had confused the
average, with the maximum.  

The average waste is 0.5 x 32 KB = 16 KB
with Windows 3.11 and no HPFS.
 
I failed to read the fine manual (RTFM),
before writing what I did.  My apologies.

But, with that said, the methodology was
still correct, as is yours below.

Thanks for the alternate route.

/s/ Paul Mitchell
http://www.supremelaw.com



At 11:51 PM 6/3/97 -0400, you wrote:
>Paul Andrew Mitchell wrote:
>> 
>> Dear Friends,
>> 
>> Several people have asked me for a safe
>> method of demonstrating the problem with
>> FAT16 partitions.  I happened to stumble
>> upon the perfect test, when I was doing
>> general maintenance last weekend.  I was
>> deleting unwanted files in the CACHE
>> directory utilized by NETSCAPE to store
>> web files which are downloaded from the
>> Internet.  I did a DOS DIR command, both
>> before and after deleting all those files.
>> The DIR command shows the sum of the
>> individual file sizes, in bytes, and
>> also the amount of space available within
>> that partition.  The discrepancy between
>> the sum, and the difference between space
>> available before the deletion, and space
>> available after  the deletion, tells the
>> story.
>>
>>...<snip>
>> 
>> Now, do the arithmetic:
>> 
>> free bytes AFTER  deletion:   244,088,832
>> free bytes BEFORE deletion:  -239,140,864
>>                              ------------
>>                 difference:     4,947,968
>> bytes used BEFORE deletion:      -751,202  (see DIR output)
>>                              ------------
>>                 difference:     4,196,766
>>                  / # files:         / 146
>>                              ------------
>> TOTAL WASTED SPACE, FAT16:         28,745 bytes per file
>>                              ============
>> 
>> So, if you have 1,000 files with FAT16,
>> you are wasting 1,000 x 28,745 = 28.7 MB!
>> 
>> Your numbers may vary;  the average will be
>> closer to a mean waste per file of 32K
>> (0.50 x 64K).
>
>Only if you are using 64K clusters on NT (the maximum
>cluster size prior to NT v4 was 32K).
>
>You don't really need to do a demonstration 
>to prove that large clusters waste disk space.
>Just realize that:
>   1) disk space is allocated in chunks called clusters
>   2) on average, each file will only fill up only half
>      of the last cluster allocated to it
>   3) run the numbers on your system
>      a) type "chkdsk" at a DOS prompt, and it will 
>         report the number of bytes per "allocation unit" 
>         (cluster)
>      b) type "dir c:\ /s" at a DOS prompt, and it will
>         tell you the total number of files on your
>         c: drive
>      c) multiply the total number of files by half of
>         the cluster size to get an estimate of the
>         dead (wasted) space on your c: drive
>
>grier
>
>=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>Unsub info - send e-mail to majordomo@majordomo.pobox.com, with
>"unsubscribe liberty-and-justice" in the body (not the subject)
>Liberty-and-Justice list-owner is Mike Goldman <whig@pobox.com>
>
>

========================================================================
Paul Andrew, Mitchell, B.A., M.S.    : Counselor at Law, federal witness
email:       [address in tool bar]   : Eudora Pro 3.0.2 on Intel 586 CPU
web site:  http://www.supremelaw.com : library & law school registration
ship to: c/o 2509 N. Campbell, #1776 : this is free speech,  at its best
             Tucson, Arizona state   : state zone,  not the federal zone
             Postal Zone 85719/tdc   : USPS delays first class  w/o this
========================================================================

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Unsub info - send e-mail to majordomo@majordomo.pobox.com, with
"unsubscribe liberty-and-justice" in the body (not the subject)
Liberty-and-Justice list-owner is Mike Goldman <whig@pobox.com>

      


Return to Table of Contents for

Supreme Law School:   E-mail