>In various replies to problems I've pointed out that various programs
>refuse to run under any version of DOS later than 3.3 because of bugs in
>MS-DOS 4.0. I'm curious about what the bugs actually were. Anyone know ?
One of the 'best' bugs in all versions of M$-DO$ 4.0x - as well as the
April version of M$-DO$ 5.0 - was the infamous CHKDSK bug...
IF you had File Allocation Tables of the size of the maximum number of
sectors - 255 IIRC - and had lost allocation units and ran CHKDSK /F -
remember that the /F does *not* stand for "fix" - instead of writing a
single backup copy of FAT 1, ie. FAT 2, it would write lots of them...
Bye, bye, Root DIRectory... and quite a bit of your initial data ! ;-|
Rather a blast, don't you think ? ;-|
>Marcus Houlden marcus_houlden (at) hotmail (dot) com
>Read this in a newsgroup? Please reply to the group.
Also many fairly low level hard disk drive utilities of the M$-DO$ 3.3
era did not understand Compaq DOS 3.31 / M$ DO$ 4.0x BIGDOS partitions
(obviously since they'd only just been 'invented' ;-) and trashed them
too. I remember setting up a whole bunch of really K001 demo's to show
the other tech's that they needed to update their software, *now* !
M$-DO$ 4.0x was also a pretty memory hungry beast compared to previous
versions... That was when most of the memory managers got released. No
coincidence there.
Are We having fun yet ? ;-)
S.NAPper
-- -
Stuff the B's - Show Me the Honey !
Well, that bit about having to load SHARE to get FCB support working on
large (>32 meg) drives was decidedly weird.
DOS 4 had a number of other "improvements" that went away in DOS 5. Most
of them probably seemed like good ideas at the time. Like the ability to
put disk buffers in expanded memory, or moving the IFS API out to a
separate file. I'm not sure that any of these could honestly be called
"bugs," but they were peculiar design decisions.
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own