Paul said:
Thanks for the link. I found it interesting.
I find this quote are especially silly from a WXP point of view:
"Long before you hit the theoretical maximum volume size, you will
reach the practical limits."
It's not really all that silly. Many folks regularly run against the
4GB file size barrier, we see posts to that effect regularly in these
groups. And the 65,535 objects per directory limit is one that is
easily and often busted by some large applications or by folks who keep
and inventory a large number of files. The 65,535 file limit applies
when using the 8.3 naming convention, using Long File Names (LFN)
significantly reduces the number of available entries as objects with 13
or more characters will use 3 (three) or more directory entries, this
applies to FAT or FAT32, but not NTFS. FAT32 folders often conk out at
20 to 25 thousand files when LFN are used, if you use really long file
names the folder can conk out at even fewer entries. This can leave
folks puzzled when they receive messages telling them that they can no
longer create files on their nearly empty drives.
I don't see any problems using MY 1.5 GB FAT32 USB drive; it has only
a few thousand files, mostly video files, none greater than 2 GB.
Your drive is certainly bigger than 1.5 GB? Even in the Windows 95 era
a 1.5GB drive was on the smallish side...
The article also states: For a 32GB FAT32 drive, it takes 4 megabytes
of disk I/O to compute the amount of free space. I assume this
means my 1.5 TB drive would require 1500GB/32GB times as much I/O, or
about 187 megabytes, which goes pretty fast with USB 2.0.
I don't know about the disk I/O math that you use or the 187MB value
that you advance but don't let the USB specifications fool you, USB
drives are still slower that IDE drives (let alone SATA) and generally
operations on large FAT32 volumes are slower than on NTFS volumes.
My thoughts are, the engineers crippled Window's formatting
capability of FAT32 systems because they could not envision that
computer speeds would increase or memory prices decrease in the
future like they had in the past.
No, not at all! When Windows 2000 was released memory capacities where
quickly increasing and prices were rapidly falling along with the
increased capacities, the same goes with disk sizes. The trend was
clearly established and I don't think that the engineers at Microsoft
were that stupid or that they were in the dark enough to not know what
the trends were. I mean, after all, Microsoft is the largest software
company in the world with nearly 100,000 employees and according to
reliable figures about 90% of the computers out there run on Microsoft
operating systems. I just don't buy the notion that Microsoft didn't
know any better, the facts are that FAT32 has limits and performance
issues that where just not suitable for modern operating systems, these
issues were especially not acceptable for a modern operating systems
like Windows 2000 which was primarily designed for business use. In my
opinion they decided to draw a line somewhere for a good reason.
I'm glad Norton Ghost's Gdisk.exe which runs under DOS, partitions
and formats my big fat32 drive to my liking. I see no downside to
fat32.
If FAT32 suits your needs better and if you are happy with it then by
all means use it! But when compared to NTFS it does have many downsides
and by far NTFS is a superior file system. NTFS is the native file
system for NT operating systems and unless they have compelling reasons
to do otherwise most users should use NTFS.
John
