9 Sep
2020
9 Sep
'20
12:40 p.m.
> Date and time of Command.com and any other DOS
files will identify the
> version number.
On Wed, 9 Sep 2020, Jules Richardson via cctalk wrote:
I've got 11/26/85 on command.com.
That would be MS-DOS version 3.10
Best known for having the "network redirector"? (which was also the
loophole that permitted MSCDEX to have CD-ROM (bigger than 32MB, BUT DOS
didn't know that it was local - thought that it was on a network
somewhere))
1.00 August 1981
1.10 May 1982 double sided drives
1.25 MS-DOS
2.00 March 1983 HDDs, hierarchical directories, file handle API
2.10 October 1983 slowed down disk access for the PCJr Qumetrak 142
2.11 MS-DOS heavily customized by OEMs
3.00 August 1984 AT/5170, 1.2MB floppy,
3.10 Nov 1985 network redirector?
3.20 January 1986 included PC-DOS 720K 3.5" floppy (some OEM MS-DOS
versions had 3.5" since MS-DOS 2.11)
3.30 April 1987 PS/2
3.31 MS-DOS heavily customized by OEMs, >32MB HDD
4.00 PC-DOS HDD >32MB "not compatible with Norton"
4.01
5.00 May 1991 DOS version in the displayed date
6.00 March 1993 massive inclusion of aftermarket add-ons
6.10 IBM's Version
6.20 October 1993 reliability improvements! (such as changing SMARTDRV)
6.21 removal of infringing compression
6.22 addition of non-infringing compression
DIR /A? or
DIR /A:H
will let you see the hidden files (presumably IO.SYS and MSDOS.SYS; PC-DOS
had IBMBIO.COM and IBMDOS.COM instead)
The /a switch isn't part of the dir
command in whatever 3.3 version I have. However, I just wrote a DOS 6.22 boot disk and that
shows three hidden files
on the hard disk:
MSDOS.SYS (11/18/85)
MIO.SYS (11/18/85)
SD.INI
I'm guessing that the 'M' in MIO.SYS is "Mitsubishi" and the OS is
tailored
in some way to the machine.
TYPEing "sd.ini" indicates that it's related to Norton Speed Disk - not a
program I'm familiar with, but as that appears to be a defragmenter, it might
hint at why my drive isn't bootable and seems to be having problems with
various executables - I wonder if the directory structure "looks sane", but
file contents have been completely mangled.
COULD BE.
HOPE NOT.
Just like, . . .
when DOS 1.10/1.25 came out, there were disasters when people ran CHKDSK
1.0 on a 1.10 disk, even if running the CHKDSK on DOS 1.10. Chkdsk
proceeded to "FIX" the disk (cf. Merck Veterinary manual)
THAT disaster resulted in later versions:
1) CHKDSK beagan to require "/F" ("fix"), AND requiring confirmation
before "repairs"
2) DOS utilities ("External commands") wouldn't run on any version of DOS
other than what they were bundled with.
THAT resulted eventuaally in release (V5.00?) of SETVER.
In my assembly language class, I showed them how to use DEBUG to patch
LINK and EXE2BIN to stop caring about DOS version, and assigned them to go
home and do it. (comment out the conditional jump after Int21h, Fn 30h)
Ah, but another classic example, . . .
Norton fUtilities wasn't compatible with DOS 4.00 or 4.01 >32MB HDD.
It needed changes to comply with the 3.31 and newer changes to HDD
The press didn't phrase it that way!
They said, "DOS 4 is broken, it isn't compatible with Norton! DOS needs
to be fixed to work with Norton!"
Obviously, a new version of Norton fUtilities was called for to go with
the new OS.
MICROS~1 learned the hard way that major changes called for meeting with
certain software vendors and giving them advanced warning, so that they
could come out with new versions at same time as new OS release.
It is probably
completely CGA compatible, unless you invoke of of its other
modes.
Maybe. Assuming that the hard drive's directory contents are intact,
and
therefore there's nothing too unusual on the drive, I've perhaps got nothing
to lose by reformatting - the only issue is that MIO.SYS, how it differs from
Microsoft's, whether it's corrupt or not, and how - if it is intact - I can
migrate it over post-install. Of course maybe speed disk (if that is even the cause
of my problems) has
some kind of backup / undo aspect, but I doubt it.
"and UNDO the
IMPROVEMENTS??!?"
--
Grumpy Ol' Fred cisin at xenosoft.com