hicks@WALKER-EMH.ARPA (Gregory Hicks COMFLEACTS) (03/04/88)
Info-IBMPC Digest Tue, 2 Mar 88 Volume 7 : Issue 20
This Week's Editor: Gregory Hicks -- Chinhae Korea <hicks@walker-emh.arpa>
Today's Topics:
DOS 3.3 and DISKPARM
Disappearing COMMAND.COM when doing Backup/Restore
Caution for those considering MSC 5.0
Compaq Portable II hard disk
FLUSHOT3.ARC protection against Trojans now available
FORTRAN to C translator
Microsoft C / PS/2-80 incompatibilty
Turbo C vs Quick C (3 msgs)
Protecting system configurations
Today's Queries:
An elusive QUATTRO bug
Excelan Card on Toshiba AT T-3200
Double buffering CGA with Microsoft C 5.0 graphics.
MSDOS 3.31 and Support of Large Hard Disks
Help Needed Fixing an Corvus Omnidrive crash
LISPs for the IBM PC
Splitting a big physical disk into smaller logical disks
SPSS-PC+ system file format
Info-IBMPC Lending Library is available from:
Bitnet via server at CCUC; and from SIMTEL20.ARPA (see file
PD1:<msdos>files.idx for listing of source files)
SIMTEL20.ARPA can now be accessed access from BITNET is via
LISTSERV@RPICICGE.BITNET using LISTSERV Commands
INFO-IBMPC BBS Phone Numbers: (213) 827-2635 and (213) 827-2515
----------------------------------------------------------------------
Date: Sun Feb 21 18:10:06 1988
From: Gregory Hicks <hicks@walker-emh.arpa>
Subject: DOS 3.3 and DISKPARM
> Info-IBMPC Digest Fri, 29 Jan 88 Volume 7 : Issue 7
>
> From: Tim Margush <R1TMARG@AKRONVM.bitnet@cunyvm.cuny.edu>
> Subject: dos 3.3 and diskparm
>
> Is it true that the DISKPARM command is not available in DOS 3.3? If so,
> what is the best way to configure a system with 1 20M, 1 360K 5 1/4 and 1
> 720K 3 1/2 drive. Can the DRIVESYS command be used to force the assignment
> of B: to the 3 1/2 inch drive??? Attempts to do this have resulted in as-
> signing D:. Any help would be appreciated.
The DRIVPARM command (there is no "DISKPARM") of MS-DOS v3.30 does not
work. A fix is in the works, and should be ready by march. If you have the
MS-DOS v3.30 Packaged Product, call Microsoft at 1-206-882-8089 and tell
them your problem, and that you got this information via the Info-IBMPC
Digest. Users of any other OEM version of DOS should contact their OEM,
since they'll have this fix as well. IBM does not use the DRIVPARM
command, so users of IBM PC-DOS don't have to worry about this.
The only workaround that I know of is to use DRIVER.SYS (there is no
"DRIVESYS"). This will create a new logical drive, instead of modifying an
existing physical drive (which is what DRIVPARM does), but the end results
are very close. If you *must* have it named B:, try using the ASSIGN
command.
Hope this helps.
Disclaimer: MY OPINIONS ARE MY *OWN*, NOT THOSE OF MY EMLPOYER.
-----------------------------
Date: Sun Feb 21 18:10:06 1988
From: Gregory Hicks <hicks@walker-emh.arpa>
Subject: Disappearing COMMAND.COM when doing Backup/Restore
> Info-IBMPC Digest Fri, 29 Jan 88 Volume 7 : Issue 7
>
> Date: 25 Jan 88 14:20:47 GMT
> From: K368181%CZHRZU1A.BITNET@CUNYVM.CUNY.EDU
>
> On my IBM XT I must maintain three versions of PC-DOS Version 3.3 for nor-
> mal use Version 3.1 to use the hercules graphic/basic overlay Version 3.2
> with a few programs that dislike version 3.3 Originally, I prepared three
> start-up diskettes, containing the two hidden files, command.com of the
> respective version and config.sys and autoexec.bat as required. containing
> path /dos31, path /dos32 or path /dos33. To speed up the booting, I in-
> cluded a shell command in the config.sys file and I set conspec to
> c:/dos31/command.com in the autoexec file. This way the respective version
> of the required command.com file was read from the hard disk from the
> respective subdirectory. This is noticably faster,
>
> Everything worked as expected, until I used backup/restore on my hard
> disk. This removed all three versions of command.com, one each from every
> subdirectory /dos31, /dos32 and /dos33. I am bringing this to your atten-
> tion because i think this is of general interest in two cases. When using
> the IBM PC Network Program V1.2, the current command.com is also stored in
> subdirectory /apps/dos3-30 and missing after restore. When using IBM OS/2
> V1.0 this useage of shell and comspec is the suggested method for booting
> DOS on an OS/2 hard disk, required for using programs that do not like OS/2
> (communications).
>
> Thank you for your help. Hippolyt
RESTORE does not restore system files: IO.SYS, MS-DOS.SYS, and
COMMAND.COM. I believe that IBM's PC-DOS RESTORE does the same, but with
IBMBIO.COM, IBMDOS.COM, and COMMAND.COM. To quote from the manual,
"RESTORE cannot restore the system files. Use the SYS command to restore
these files." This is unfortunate for a situation like yours, where you
are keeping many copies of DOS on the system. The only workaround I can
think of at the moment (if you continue to use BACKUP/RESTORE) is to have
a batch file that will rename these files after you restore them. (Ugly, I
know, but then again, I've had use to rename COMMAND.COM to SH.COM on
occasion...)
Sorry about the work-around, but I hope this helps.
------------------------------
Date: 23 Feb 88 00:29:12 GMT
From: Henry Spencer <henry@utzoo.uucp>
Subject: Caution for those considering MSC 5.0
> >... isn't all global data implicitly
> >initialized to zero if not otherwise specified?
>
> It's true on "many", but not all systems...
This is like saying that 2+2 == 4 is true on many but not all systems.
2+2 == 4 and implicit initialization of globals to zero are both
guaranteed facts in correct implementations of C. Your own environment
will determine how interested you are in buying and using incorrect
implementations. Personally I think it's a waste of money and time.
------------------------------
Date: Mon, 22 Feb 88 10:37:51 PST
From: swillett%plutonium.CChem.Berkeley.EDU@jade.berkeley.edu (Steve Willett)
Subject: Compaq Portable II hard disk
To: sydney%acorn@live-oak.lcs.mit.edu
Just saw your note in Info-IBMPC Digest. I have, sitting at home on my
desk, a Compaq Portable II with 44 Mbytes of hard drive in it. The system
was assembled for me by a small outfit in San Rafael, CA, called Applied
Computer Technology. I don't have their address or phone here, but I
could get the info for you if need be. The area code for San Rafael is
415 if you want to go through information.
The drive is, I believe, a MicroScience 50 Mb drive, which formatted to 44
Mb in the Compaq. It has a 28 ms access time and is formatted as two
logical drives (C: at 33 Mb and D: at 11 Mb). I have had the system for
over a year with absolutely no trouble.
I do believe that the installation was a shoe horn operation - when I
bought the machine from them I asked them what the largest drive was that
they could put in and the engineer said "Well there is an interesting
drive I think I could squeeze in - I've always wanted to try!" He did,
and I have been very pleased with the results.
If you want more info from me reply by the route of your choice:
Steve Willett
1739 Ward St.
Berkeley, CA 94703
(415) 849-3025 (Home - and an answering machine)
(415) 643-9544 (Work - sometimes!)
(415) 642-2057 (Work - sometimes, and another *#$^&^!! machine)
swillett@plutonium.cchem.berkeley.edu
------------------------------
Date: Sun, 21 Feb 1988 16:53 MST
From: Keith Petersen <W8SDZ@SIMTEL20.ARPA>
Subject: FLUSHOT3.ARC protection against Trojans now available
Now available VIA STANDARD ANONYMOUS FTP from simtel20...
Filename Type Bytes CRC
Directory PD1:<MSDOS.DSKUTL>
FLUSHOT3.ARC.1 BINARY 9088 0546H
This is an update to FLUSHOT2, announced previously, a program to
protect your COMMAND.COM, FAT, boot sector, NOVRAM, etc., against
Trojan horses.
The following is a list of its contents. If you get this file from
other sources be sure to check the CRCs to make sure they match this
listing. (more comments on this below).
Name Length Stowage SF Size now Date Time CRC
============ ======== ======== ==== ======== ========= ====== ====
DISCLAIM.TXT 640 Crunched 27% 469 10 Feb 88 0:00a 4B65
FLUSHOT2.INF 2176 Crunched 38% 1361 10 Feb 88 0:00a C709
FLUSHOT3.COM 2363 Crunched 20% 1912 10 Feb 88 0:00a 6C75
FLUSHOT3.INF 2432 Crunched 34% 1627 10 Feb 88 0:00a 2DA1
FLU_SHOT.DOC 4045 Crunched 43% 2322 10 Feb 88 0:00a 4C6A
REGISTER.TXT 2816 Crunched 57% 1220 10 Feb 88 0:00a 15D8
==== ======== ==== ========
Total 6 14472 39% 8911
--from file FLUWARN2.TXT--
FURTHER INFORMATION ON "FLUSHOT" FILES!
The program FLU-SHOT.arc has had "some" copies "BUGGED". Instead of
protecting you from the so called "COMMAND VIRUS"...they actually erase
Command interpreters and files!
I spoke to the author today. A new version of Flushot has been released
called FLUSHOT3.ARC. USE ONLY THIS VERSION ! Other versions "M A Y" have
been tampered with and be Trojan. To get a working copy of the "GOOD"
FLUSHOT PROGRAM call 212-889-6438 and download FLUSHOT3.ARC.
If you have a copy of any other release...please check it out
C A R E F U L L Y !
FLUSHOT3 is an excellent program....and has been installed on my Board
with no problems. I was NOT so lucky with the original version I received.
It had been tampered with and....erased over 15 files and Command.com
interpreters from my system and the systems of 5 other users. Some had to
completely re-format!
There "may" be other so-called "cures" that ( in actuality ) are Trojan.
ALWAYS test these programs before installation and BE SURE they are OK!
WARNING: ON tampered versions of FLU-SHOT, most Bomb programs detect no
problems with the program...I know because I ran three of them before
installing to my system. After my disaster...I looked at the program using
various utility files. I can still detect nothing out of the ordinary.
However, looking at command.com (after the installation) you will note
"garbage" at the end of your current command.com file. If you see this
"DO NOT RE-BOOT YOUR SYSTEM! TAKE YOUR ORIGINAL DOS BOOT DISK and COPY
COMMAND.COM OVER THE OLD VERSION ON YOUR HARD-DISK! I did not do this
and....... you know the rest.
Leonard Lee..Sysop
VoiceQuest RBBS
(601) 638-3390
---end included text--
Good advice. The file FLUSHOT3.ARC on SIMTEL20 came direct from Ross
Greenberg, the author of FLUSHOT. I downloaded it myself from his BBS.
--Keith Petersen
Arpa: W8SDZ@SIMTEL20.ARPA
Uucp: {decwrl,harvard,lll-crg,ucbvax,uunet,uw-beaver}!simtel20.arpa!w8sdz
------------------------------
Date: Mon, 22 Feb 88 08:45:02 PST
From: herman@marlin.nosc.mil (John W. Herman)
Subject: FORTRAN to C translator
FORTRIX (FORTRAN to C) 1-800-FORTRIX
FORTRIX (FORTRAN to ADA)
COBLIX (COBOL to C) 1-800-COBLIX
Within New York (914)368-3000
Rapitech Systems, Inc.
Montebello Corporate Park
Suffern, NY 10901
I have no experience with this product. I never heard of them until last
Saturday when the card pack came in the mail.
------------------------------
Date: 24 Feb 88 01:07:43 GMT
From: Clayton Cramer <cramer@optilink.uucp>
Subject: Microsoft C / PS/2-80 incompatibilty
>
> We've just been trying out Microsoft C version 4.00 on a PS/2 model 80.
> Every time we try to compile we get an error message
>
> "Fatal error 13, cannot open <file>.C".
>
> We're using PC-DOS 3.30, but running MSC and DOS 3.3 on a standard PC
> works OK, so that's not the problem (I guess).
>
> Anyone got any ideas how to fix this, or what to do?
>
> All hints appreciated.
> ---
> Bjorn Heimir Bjornsson Internet: bjornb@rhi.hi.is
> University of Iceland UUCP: ..!mcvax!hafro!rhi!bjornb
If you had asked last week, I would have had no idea. Fortunately, I just
bought myself an AT clone, tried my first compile, and had a brief attack
of panic. ("Oh no, it's not even compatible with the Microsoft C
compiler!")
Check your CONFIG.SYS file -- if you don't have a FILES= line, the default
number of files won't be enough for the C compiler to work. Try FILES=32.
Clayton E. Cramer
88 0:00a 2DA1
FLU_SHOT
------------------------------
Date: Tue, 23 Feb 1988 10:41 EST
From: ejs%acorn@oak.lcs.mit.edu
Subject: Turbo C vs Quick C
In reference to Jim Vallino's comments about incorrect memory models
mathing causing unpredictable errors: Very well put. The blame is not
really to be placed on the C compilers either -- DOS is the culprit. With
OS/2 and the use of the protected-mode of the Intel processors these
problems should go away. Without memory protection, there is nothing that
can be done to protect against errant programs trashing system data
structures. So please, blame DOS and switch to OS/2.
------------------------------
Date: 22 Feb 88 15:29:34 GMT
From: Peter Knoppers <knop@dutesta.uucp>
Subject: Turbo C vs Quick C
Why, oh why don't the .obj files in MSDOS contain some bits telling the
linker whether a function in the .obj file expects to be called with a FAR
or a NEAR call. This can prevent accidentally linking modules compiled for
different models.
Peter Knoppers, Delft Univ. of Technology
...!mcvax!dutrun!dutesta!knop
or
knop@dutesta.UUCP
------------------------------
Date: 19 Feb 88 11:08:00 GMT
From: Ian Phillipps <igp@camcon.uucp>
Subject: Turbo C vs Quick C
From article <389@lscvax.UUCP>, by ram@lscvax.UUCP (Ric Messier):
> I've finally talked myself into picking up a C compiler and thought I
> had even decided on which one to get. The price was just right and so [...]
>
> I can pick up both Microsoft's Quick C and Borland's Turbo C for $53/ea but
> I am not sure which is the better compiler. I have asked everyone I know
> but ...
I've used Turbo 1.0 and MSC 4.0. The latter got squeezed off my disk a
while back. The Turbo editor is limited as an editor, but does have remap
- you could tart it up if need be with a keyboard remap program.
> inline assembler code to DOS and BIOS interrupt calls within the code,
Yes - the compiler will generate IN OUT INT and you have direct register
access, if you like that sort of thing.
>
> Anyway, to the point. I want to know what kind of package each of them
> is, what kind of graphics capabilities, support of the Kernighan/Ritchie
Both go WAY beyond K & R. I think Borland is a bit ahead of MS on some
things; as I haven't read the draft ANSI standard, I dont know if
declarations like
main( int argc, char **argv ) { /* hello world */ }
are in it - Borland takes them, MSC doesn't. MSC 5.0 "totally compatible
with Quick C" barfs on #pragma - guess how I found that out!
There are irritating differences with the header files (memory.h vs mem.h)
but not much serious unless you use the Turbo extensions in non-macro
form. (If you use inp or outp, Turbo will do them in-line via macros).
> release of Turbo C. I am kind of partial to pulldown menus which are
> included in the Quick C package and I don't particularly enjoy going to
> the Turbo editor sometimes, though I realize that I can change the
> commands to whatever suits me, there are just some things I want to do
> that it can't.
If you like pull down menus, youll LOVE Turbo C. There are lots of
graphics in Turbo V 1.5 - I haven't got that yet.
... all views purely personal ...
UUCP: ...!ukc!camcon!igp | Cambridge Consultants Ltd | Ian Phillipps
or: igp@camcon.uucp | Science Park, Milton Road |-----------------
Phone: +44 223 358855 | Cambridge CB4 4DW, England |
------------------------------
Date: Mon, 22 Feb 88 22:09:16 EST
From: Russell Nelson <nelson@clutx.clarkson.edu>
Subject: Protecting system configurations
I, too, have been looking for a way to do this with the IBM-PC, and my
conclusion is that a modified BIOS is needed. Even Zenith's BIOS, which
includes everything but the kitchen sink, doesn't include a "don't boot
from anything but the hard disk, and don't do Ctrl-Break either" mode.
Obviously, any software-only scheme is doomed to failure, as Greg pointed
out.
-russ
------------------------------
Date: Mon, 22 Feb 88 19:30:19 EST
From: Joe Morris (jcmorris@mitre.arpa) <jcmorris@mitre.arpa>
Subject: An elusive QUATTRO bug
I've encountered an elusive bug in Borland's QUATTRO product; the trouble
is that I can't reliably recreate it even if I follow identical keystroke
sequences beginning with DOS boot. I'm posting this in the hopes that
someone else may be seeing the problem so that I can stop worrying that
it's something I've done on my own...and yes, I've reloaded the code from
the original disks.
Take the following spreadsheet:
A1: The number -.1
B1: The number 0
C1: @SUM(A1..B1)
D1: @IF((C1<>0),@ERR,"zero")
As entered, cell D1 properly displays the error flag. If the bug is
awake, something doesn't happen properly when the spreadsheet is stored,
because when the file is loaded again D1 displays the string "zero".
Touching any of the cells on which D1 depends (or D1 itself) causes the
display to change to the error flag. Borland's help desk hasn't been able
to recreate the error.
This error makes me question the validity of any retrieved spreadsheet
which contains an @IF function. I've seen the error on a AT running DOS
3.3 without a math chip, and a PC-1 running DOS 3.1 and an 8087.
Since Borland doesn't have a way to command a total recalculation, my
"fix" was to move the entire spreadsheet on top of itself (e.g., move
A1..D1 to A1).
The situation stinks of an uninitialized variable somewhere, but I would
expect that that should be "uninitialized" the same way after a cold boot.
Ideas, anyone?
------------------------------
Date: 22 FEB 88 08:45:16
From: Z3000JD%AWITUW01.BITNET@CUNYVM.CUNY.EDU
Subject: Excelan Card on Toshiba AT T-3200
I have a Thoshiba AT, Type T3200 (1 Mbyte, no extended Memory) and want to
install a Excelan Ethernet Adapter Card into the long slot of the AT. The
Card has Product-Name EXOS 225.
The Card can be configured must be configured to use a RAM-page of 64
kbyte to communicate with the Software running in the AT. It can be
configured to be at address 80000, 90000, A0000 or B0000. Interrupt
address and IO-Address range must be configured too.
I have tried all four memory-addresses with the following results:
80000 AT boots successfull, Card not accessible
90000 AT boots successfull, Card not accessible
A0000 During Selfttest, I get a PARITY ERROR 2
B0000 'boot successfull' but screen not usable.
Cases 8,9 and B are understandable, because in this cases there is an
overlap with the RAM or the display-RAM.
I do not understand case A0000 which should work, because their should be
no memory etc. All works fine on a XT-clone.
Does anyone has an idea how to solve the problem?
Johannes Demel, Technical University Vienna,
Z3000JD@AWITUW01.BITNET
P.S. please respond directly to me.
------------------------------
Date: Wed, 17 Feb 1988 11:55 EST
From: <JWK%MCOIARC.bitnet@cunyvm.cuny.edu>
Subject: Double buffering CGA with Microsoft C 5.0 graphics.
Has anyone used Microsoft C 5.0 or Quick C graphics to perform double
buffering on a CGA compatible display? I need to simulate a hidden
graphics page in RAM and convince the Microsoft graphics routines to write
to it in place of the standard CGA address to perform common double
buffered animation. This could be done in BASICA by changing the written
to page address, anyone with details of how do to it in MSC 5.0? Thanks
in advance.
Joe Klingler
Image Analysis Research Center
Medical College of Ohio
bitnet address: <jwk@mcoiarc>
------------------------------
Date: Sun, 21 Feb 88 12:10:18 MST
From: Villy G Madsen <VMADSEN@UALTAVM.bitnet@cunyvm.cuny.edu>
Subject: MSDOS 3.31 and Support of Large Hard Disks
Does anyone have detailed information about the changes that were made to
allow support of >32 MByte disk drives with this release. As far as I can
determine, changes would have to be made to INTR 25 and 26, and DOS
service 1Ch, and the layout of the BPB tables....I need to know what the
changes are...
Thanks... Villy
------------------------------
Date: 22 February 1988 16:45:09 CST
From: Bill Rogers <ROGERS@UIUCVMD.bitnet@cunyvm.cuny.edu>
Subject: Help Needed Fixing an Corvus Omnidrive crash
My justification for sending to this list is there is a larege number of
crazy IBM-PC users needing help. Our Corvus Network HAD a 45 Mb Omnidrive.
The little fellow started making noises like it was grinding coffee beans
one day and that was the end of that drive on that server.
The service people tell me $1200+ to replace the drive. Fat chance, no
netbios compatiablity, no file sharing, damn slow when you really get down
to brass tacks. Well, I opened up the coffin and low and behold there is
the plainest looking Rodime 5.25 hard drive inside. The only note is
"build level 7" on the drive. Here's what could win you a turkey this
christmas:
1. Does anyone know if I might be so bold as to slip a $659 40 Mb into
it's place and laugh at the service people.
2. If #1 is not possible, does anyone have any comment on Corvus's PC-NOS
product as the only thing that might run on their board.
Help bring a doorstop back to life, Thanks,
Bill
------------------------------
Date: Tue, 23 Feb 88 15:09:46 PST
From: Brian Leverich <leverich@rand-unix.ARPA>
Subject: LISPs for the IBM PC
Can anyone with hands-on experience provide me with some comments on LISPs
that run on PCs under MSDOS? The slick vendors' literature has a Shannon
measure just about epsilon away from zero. :-(
I'm particularly interested in how large the data spaces of the LISP are
(how many atoms, DTPRs, etc., can be crammed into memory) and speed of the
compiled code, especially at handling real number arithmetic. Tnx.
------------------------------
Date: Mon, 22 Feb 88 10:41:09 SET
From: "K.Keyte" <ESC1332%ESOC.BITNET@CUNYVM.CUNY.EDU>
Subject: Splitting a big physical disk into smaller logical disks
How do I split a 40MB disk into say three smaller logical drives, say
D:, E:, and F: of 20MB, 10MB, and 10MB respectively? FDISK???
How does the OS (MS-DOS 3.2) know about the drive set-up?
Please help!
Karl (ESC1332@ESOC.BITNET)
------------------------------
Date: Tue, 23 Feb 1988 12:15:53 CET
From: AHA21%DK0RRZK0.BITNET@CUNYVM.CUNY.EDU
Subject: SPSS-PC+ system file format
Has anyone information concerning the format of SPSS-PC+ system files?
This information is required for interface routines to other statistical
programs. Thanks in advance.
R.Schnell
------------------------------
************************
End of Info-IBMPC Digest
-------