Info-IBMPC@WSMR-SIMTEL20.ARMY.MIL ("Info-IBMPC Digest") (02/24/90)
Info-IBMPC Digest Sat, 24 Feb 90 Volume 90 : Issue 30
Today's Editor:
Gregory Hicks - Chinhae Korea <GHICKS@WSMR-Simtel20.Army.Mil>
Today's Topics:
8087 Emulation
Bogus Version of SCANV58
Bug in PKZIP 1.02???
COM1: pin-outs
Comments on CRUNCH10 and MS Kermit 3.0
Problem with Microsoft QC 2.0 (2 msgs)
UNARCing files Downloaded from SIMTEL20
UNARCing files Downloaded from SIMTEL20 (Solution)
Write lock for floppies
Today's Queries:
File txfer problems
get PC to auto call home when reboot
Graphics Terminal Emulation Programs
RLL controller for an OPTIMA70 hard disk
Send Replies or notes for publication to: <INFO-IBMPC@WSMR-SIMTEL20.ARMY.MIL>
Send requests of an administrative nature (addition to, deletion from the
distribution list, et al) to: <INFO-IBMPC-REQUEST@WSMR-SIMTEL20.ARMY.MIL>
The Lending Library is available from: WSMR-SIMTEL20.ARMY.MIL (see file
PD1:<MSDOS.FILEDOCS>AAAREAD.ME details on file directories and
descriptions.)
Archives of past issues of the Info-IBMPC Digest are available by FTP only
from WSMR-SIMTEL20.ARMY.MIL in directory PD2:<ARCHIVES.IBMPC>.
WSMR-SIMTEL20.ARMY.MIL can be accessed using LISTSERV commands from BITNET
via LISTSERV@NDSUVM1, LISTSERV@RPIECS, LISTSERV@FINTUVM and in Europe from
EARN TRICKLE servers. Send commands to TRICKLE@<host-name> (example:
TRICKLE@TREARN). The following TRICKLE servers are presently available:
AWIWUW11 (Austria), BANUFS11 (Belgium), DKTC11 (Denmark), DB0FUB11 or
DTUZDV1 (Germany), IMIPOLI (Italy), EB0UB011 (Spain) TAUNIVM (Israel) and
TREARN (Turkey).
If you are unable to access SIMTEL20 via Internet FTP or through one of
the BITNET/EARN file servers, most MSDOS SIMTEL20 files, including the
PC-Blue collection, are available for downloading on the Detroit Download
Central network at 313-885-3956. DDC is a networked system with multiple
lines that support 300, 1200, 2400, and 9600 bps (HST). This system is a
subscription system with an average hourly cost of 17 cents per hour. It
is also accessable on Telenet via PC Pursuit and on Tymnet via StarLink
outdial. New files uploaded to WSMR-SIMTEL20 are usually available on DDC
within 24 hours.
----------------------------------------------------------------------
Date: Wed, 21 Feb 90 22:18:58 KST
From: Gregory Hicks - Chinhae Korea <comfleact@taegu-emh1.army.mil>
Subject: 8087 Emulation
As much as I hate to tell you this, I don't believe that the 8087 can be
simulated if the program REQUIRES the chip. The 8087 instructions are
placed inline with the program and there is no way an 8087 simulator can
intercept these instructions and leave the 8088 ones alone.
If you're DEVELOPING a program, most compilers come with 8087 emulation
software in the library.
I've forwarded your message. Let's see what develops.
Best,
Gregory Hicks
------------------------------
Date: Wed, 21 Feb 90 13:12:36 PST
From: portal!cup.portal.com!Alan_J_Roberts@SUN.COM
Subject: Bogus Version of SCANV58
This is a forward from John McAfee:
We have received reports of a SCANV58.ZIP file that contains a bogus
VALIDATE program. The program is an EXE file (the real validate is a COM
file) and is 46,167 bytes long versus 6,485 for the original VALIDATE.
There have been reports of system damage from the use of this program and
under no circumstances should it be used.
The authorized version 58 of scan is 42,977 bytes long, has a creation
date of 2-15-90 and the validation checks should be: Method 1: 2F16 and
Method 2: 1C57.
If you come accross the bogus version and have information about where
it came from, then please contact me at 408 988 3832. The bogus validate
program appears to be identical to a program uploaded to HomeBase on
2-19-90 by a person usinmg the name Richard Levey. The documentation for
the program contained a copyright by Richard Levy, but there was no phone
number or any other contact information provided.
As you can imagine, we are very anxious to find this person and if
anyone has any information then please call.
John McAfee
408 988 3832 (voice)
408 970 9727 (FAX)
408 988 4004 (BBS)
------------------------------
Date: Tue, 20 Feb 90 19:41:28 PST
From: nyet%gap.caltech.edu@Tybalt.Caltech.Edu (Nye T. Liu)
Subject: Bug in PKZIP 1.02???
In comp.sys.ibm.pc.digest you write:
>I think I've found a bug in PKZIP/UNZIP 1.02! Whenever I run the -t
>option to test the archive, for each file in the archive I get this
>message:
>PKUNZIP: Warning! can't open: NUL
When you say : can't open : NUL, does is really say "NUL"? Since "NUL" is
MS-DOS's null device, I can only assume either you are using pkunzip
through a batch file which is trying to pipe output into the the NUL
device improperly, you have a corrupted version of pkzip (I've had no
problems running the -t option on any IBM i've used..), or the pkzip
program itself is trying to use the NUL device and for some reason NUL
does not exist on your computer.
Does anything else that you have use the NUL device succesfully? If so,
that's not the problem, and something else is wrong. If you don't know
how to use NUL, try something along the lines of: C:>echo hello > NUL
(spaces before and after the ">" are VERY important) If nothing echos, NUL
works properly (it has piped the Standard output "hello" into space..)
If it doesn't work, you are in big trouble..
------------------------------
Date: Wed, 21 Feb 90 9:08:27 EST
From: Bill Clay (CSD) <wpclay@BRL.MIL>
Subject: COM1: pin-outs
In V90, Issue 17, Tom Claydon asked about the pin-outs on the COM1:
connector on a Kaypro.
Tom,
Unless Kaypro provides unique drivers for devices attached to COM1:,
the following should work:
PIN CONFIGURATION FOR IBM PC 9 PIN CONNECTOR TO RS232 DB-25 CONNECTOR This
configuration makes the IBM PC 9 pin connector plug in compatible with the
RS232 connector on a DCE device. For a DTE device, RS232 pins 2 and 3
should be swapped and RS232 pins 8 and 20 should be swapped.
IBM CONNECTOR RS232 CONNECTOR
PIN NUMBER PIN NUMBER
1 (DCD) in 8
2 (RXD) out 3
3 (TXD) in 2
4 (DTR) out 20
5 (GND) 7
6 (DSR) in 6 *
7 (RTS) out 4 *
8 (CTS) in 5 *
9 (RI) out 22 *
* Pin is not needed for most devices. Check device manuals.
Note 1: Modems, acoustic couplers and line drivers are almost universally
DCE. There is no standard, however, for printers. Some are DCE, some are
DTE, check the manual.
Note 2: Pin 5 on the 9 pin connector must always be connected to pin 7 on
the 25 pin connector. This is signal ground and is required for proper
operation.
Note 3: Pins 2 and 3 on the 25 pin connector are always used for
communications and pins 7 is always used for signal ground, beyond that
there are a lot of devices that use non-standard pin assignments for flow
control. Again, check the manual.
Note 4: Don't assume that if a device refers to a pin as TXD (Transmit
Data) that that is the pin used to output characters. Some devices (note
above) refer to pin 3 as TXD even though that is the pin on which it is
receiving data.
Hope this helps.
-- Bill Clay
------------------------------
Date: Wed, 21 Feb 90 10:49:00 EST
From: <GILL%QUCDNAST.bitnet@ugw.utcs.utoronto.ca>
Subject: Comments on CRUNCH10 and MS Kermit 3.0
CRUNCH10 is a utility that was posted to SIMTEL in January. I gave
it a whirl because it promised to be useful. Well, it almost was what I
was looking for. CRUNCH10 compresses files, which automatically
uncompress upon running them. For a saving in disk space, you lose a
little bit in start-up time. However, I found that you lose a lot. A 70k
file was compressed by about 12k in about 1 minute. Uncompressing took
about 15 seconds - far too long to make this programme useful, unless one
happens to be in dire need of hard disk space.
The reason that I had high hopes for this utility was due to the
existence of a similar utility available for the Amiga. That one is
called Power Packer (written by Nico Francois). It takes a long time to
pack files (like 10 minutes!), but the reduction in file size is so great,
that loading and decoding the shortened file is NOT appreciably longer
than just running the uncompressed file. An amazing utility, and
especially useful for those 300k programmes which are starting to appear
everywhere. I have been looking for a PC equivalent ever since. Does
anyone know of such a beast? Now that you know such a thing exists (PC
hackers and Amigoids don't have much of an overlap, in my experience), is
anyone interested in attempting to write something like this? Maybe a
port from the Amiga utility would be a starting point?
One other problem I have experienced. I have a keyboard buffer
routine written that gives me 150 type-ahead keystrokes, instead of the 15
DOS so graciously bestows on its users. However, several programmes that
I have conflict with this TSR - namely MS-Kermit 3.0 (only upon
connecting) and MS QuickBasic 3.0. Does anyone have a keyboard buffer
programme that doesn't interfere with anything else?
I would appreciate answers sent to me - I read this list rather
infrequently. Thank you in advance. I'll post a summary if one is
warranted.
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
| Arnold Gill |
| Queen's University at Kingston |
| BITNET: gill@qucdnast |
| INTERNET: gill@bill.phy.queensu.ca |
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
------------------------------
Date: Wed Feb 21 10:23:43 1990
From: microsoft!toddw@beaver.cs.washington.edu
Subject: Problem with Microsoft QC 2.0
In Info-IBMPC Digest V90 #26, David Zielke <zielke@phy.duke.edu> asked:
> I am using Microsoft QC 2.0 and having horrible failures trying to use
>the assert directive! I tried compiling the sample code provided in the
>help utility and even that fails... Is this a known problem or am I doing
>something stupid?
Not sure where you are having problems, but I'll review the steps.
Obviously, as you already know assert() is invaluable in debugging
unexpected circumstances in a program. Here's a check-list for its use:
1. the macro NDEBUG must NOT be defined
2. the assert condition must be false. Try the following
#include <assert.h>
#include <stdio.h>
main() {
assert(0);
}
This should print an assertion when run.
3. Double check your environment, in particular:
INCLUDE should point to the location of the QuickC 2 include files.
If you own more than one compiler you will need include files for each.
You can check this from within the environment under options.environment
in the INCLUDE edit field.
LIB and PATH should also point first to the ms qc2 files first.
make sure the first link.exe on your path is the qc link.exe.
4. If the above fail, try the sample program above from the command
line (e.g. qcl assert.c) If that fails, send me a message listing the
contents of your environment. (e.g. set > mymessage.txt) and the
date/time stamps of your include files and LIB files (combined libs are
named in the form ?LIBC?.LIB where the first ? represents the memory
model and the last ? the math options (emulator, 8087, altmath).
thanks,
Todd Warren
C program manager.
------------------------------
Date: Wed, 21 Feb 90 17:46:16 EST
From: zielke@phy.duke.edu (David Zielke)
Subject: Problem with Microsoft QC 2.0
Mr. Warren,
In working through the suggestions you made I seem to have found the
problem. I use the
SET CL=\AL \FPi87 \Za
command line variable to get QC to use a large memory model, use the
floating point chip and to hold to the ANSI standard for C. If I remove
the \Za from the CL variable it compiles fine, otherwise it fails. Is this to
be expected. Does assert somehow not follow the ANSI standard?
At any rate, my programs are now compiling and I have at least found
that I have a problem in one of my routines from the use of the assert
command. A variable which should never be more than 4 is at 18932! Silly
programming error which would have taken forever to find. Thank you much
for your interest in this problem.
Oh, I would appreciate knowing if there is a way to force ANSI
compatability and still use assert. Also, does the QC compiler support
the defined variables which return line number and module name? If I
remember correctly they are like __LINE__ or __MODULE__ but I am not sure
of the names. There are some advanced versions of assert.h which expand
as macro's which do fancy things like printing out line number and
subroutine name.
Thank you again for your help, from a satisfied owner!
David M. Zielke
zielke@physics.phy.duke.edu
zielke@cs.duke.edu
zielke@ccf3.nrl.navy.mil
------------------------------
Date: Tue, 20 Feb 90 19:44:39 PST
From: nyet%gap.caltech.edu@Tybalt.Caltech.Edu (Nye T. Liu)
Subject: UNARCing files Downloaded from SIMTEL20
In comp.sys.ibm.pc.digest you write:
>I too have problems unarcing files downloaded from Simtel. First I ftp a
>file to a vax/unix. I test it there with arc t and everything is OK. Then
>I use kermit to transfer the arced file to my pc. Here I try to apply
>pkxarc and I get a message that the crc check fails. Am I missing
>something? Oscar M.
The only thing I can imagine isn't working is you not setting file type to
binary on kermit, or you not setting file type to tenex on simtel
Either that, or you're version of your unarcer is to old..
Other than that..????
------------------------------
Date: 22 Feb 90 10:01:00 GMT-9:00
From: 1962cg_lgmq@kadena-emh.af.mil
Subject: UNARCing files Downloaded from SIMTEL20 (Solution)
Several people mentioned having problems UNARCing files downloaded from
SIMTEL20... I used to have a problem, here's how I resolved it. At least
once a month I download SIMIBM.IDX and import it into a DBASE viewing
program. Before downloading a file I extract the following information
from the DBASE program:
1. Directory Name
2. File Name <FILE>
3. File Type (7 or 8 Bit)
4. File Size <AFile> (Big Ticket Item)
I generate a script file with ASPECT or some other script generator.
Within the script file I compare the number of bytes transmited by the
SIMTEL computer <TFILE> with the item 4 (Actual File Size <AFile>).
Through experiance noticed as long as TFILE doesn't exceeded the size of
AFILE by more than 6Bits the FILE UNARCs okay on my PC.
IF (TFile > (AFile + 6Bits))
THEN GET FILE
ELSE GET NEXT FILE
An exact match (TFile = AFile) is ideal but not always possible. No, I
don't have an explanation for the 6Bit window.
I hope the above is both clear and helpful.
CRAIG L. SMITH
------------------------------
Date: Wed, 21 Feb 90 9:13:08 EST
From: Bill Clay (CSD) <wpclay@BRL.MIL>
Subject: Write lock for floppies
TSgt Frank Starr asked about write lock programs for floppies to
force folks to return his disks.
I think you are mistaken in your assumption that if you can make the
floppies read only to others, they will bring them back. In the first
place, as long as your users can read them, they can use them as the back
up copy of whatever you gave them. Secondly, I think most users have a
lot of floppies laying around that never get used.
I would suggest that the solution to your problem is to require your
users to supply you with a disk as a condition on getting one from you.
Good luck.
-- Bill Clay
------------------------------
Date: Thu Feb 22 00:00:22 1990
From: johnboyd@ocdis01.af.mil (John Boyd)
Subject: File txfer problems
I seem to be having difficulty transferring binary files. Here's how the
process goes:
I get into the FTP site OK, and set the proper file type. When I compare
the file size from the sending host to the size it occupies on my
receiving host, they match. I then use ST220 (for which we have not been
issued manuals at the working level) to transfer the file from the
receiving unix machine to my desktop Zenith Z-248 using Kermit. This is
where the problem arises. The file size no longer matches. It's larger
on my PC after the transfer, and any unarcing indicates that the file is
corrupt. However, text file transfers are always OK. I have pretty much
narrowed the problem down to some sort of incompatibility between the
Kermit at my end, and the UNIX machine; (I think).
Any Ideas?? All help appreciated.
[I tried to duplicate this problem and succeeded. To satisify SIMTEL20, I
had to declare TYPE 8 L (my FTP process understands this as TENEX type)
before I could start the transfer. Once the ARC file was at my host (runs
Unix BTW), I used Kermit to transfer the new file. As long as I told
Kermit SET FILE TYPE BINARY, the file transferred. I tried transferring
the file without this and ended up with a MUCH larger file that filed the
PKXARC CRC check and that the file was corrupt. The NAME of the archived
file was there, but it couldn't be unarchived past the first few hundred
bytes. The command SET FILE TYPE BINARY or SET FILE TYPE TEXT should be a
standard one and should be issued '...just to make sure that...' the
program is set the way YOU want it to be. gph]
------------------------------
Date: 21 Feb 90 16:11 -0800
From: Thomas Wong <twong@civil.ubc.ca>
Subject: get PC to auto call home when reboot
A colleague of mine has a problem.
He is doing research where a computer must constantly be running to
monitor and control his experiment for periods of several months. His
problem is the lab where he works occassionally gets blackouts. These
blackouts are short so the computer would be up again in under an hour. He
needs to be notified right away if the blackout occurs so that he can come
in, fix things up when power comes back up, and continue the process. He
is able to do this if he comes in within several minutes of the power
coming back up and the process has gone too far. It would be nice if he's
notified when the blackout occurs but it's more practical if during the
reboot, the computer runs a program that would notify him at home.
Therefore, one possibility is to put a program call in his autoexec.bat
file.
I suggested the answering machine that I've been hearing about that plugs
into a slot in a PC. One of its feature is to keep a track of a schedule
so that it can call a phone number with a certain message at a certain
time. So I thought it might be able to call him everytime the computer
reboots and the program is executed. But this option might be too
expensive. Does anybody know how much this card/answering machine costs?
And if this setup might work?
My other idea was to run a communication program in his autoexec.bat which
accepts login scripts. Basically what this does is when the computer
reboots, and it executes autoexec, which executes the comm program, which
runs the the login scripts to call his home and just beeps to try to get a
carrier detect. So when he hears a beep, he knows he better start running.
It's a lot cheaper. Suggestions? Can anybody recommend a communication
program that takes login scripts?
Does anyone have any other ideas?
Thank you in advance.
Thomas.
/*----------------------------------------------------------------------*/
/* Thomas Y. K. Wong INTERNET: thomas_wong@civil.ubc.ca */
/* Civil Eng., U.B.C BITNET: thomas_wong%civil.ubc.ca@ubcmtsg */
/* Van, B.C., Canada UUCP: ...!van-bc!civil.ubc.ca!thomas_wong */
/*----------------------------------------------------------------------*/
------------------------------
Date: Wed, 21 Feb 90 14:46:05 PST
From: theo@ouzo.berkeley.edu (Theologos Kelessoglou)
Subject: Graphics Terminal Emulation Programs
I am looking for PC programs emulating GRAPHICS terminals (Tektronix
etc.). Any information on such programs available in the public domain or
commercialy will be appreciated.
Please respond by personal mail info the Digest as I do not follow this
newsgroup. Thank you.
-Theo Kelessoglou
theo@ic.Berkeley.EDU
------------------------------
Date: Wed, 21 Feb 90 17:01:45 CET
From: SVAGHI%ESASTSP.BITNET@CUNYVM.CUNY.EDU
Subject: RLL controller for an OPTIMA70 hard disk
A friend of mine has bought an OPTIMA70 hard disk distributed by a
Company called Core International to be used in his AT-clone.
He has not been able to format the hard disk and he has been told that,
instead of the normal control card, he should use an RLL-type control
card.
[He should use the RLL card if he wants to get the increased storage
capacity. Unless I'm mistaken, an RLL drive is the same as a normal drive
EXCEPT the RLL drive is manufactured to tighter specifications due to the
touchiness of the encoding algorithm. gph]
Such card costs here in Holland the equivalent of approximately $750.
He would like to hear about any mail dealer in the States who could
provide an RLL control card for the OPTIMA70 hard disk, and an indication
of its price there.
Any pointer will be appreciated.
Best regards,
Sergio Vaghi SVAGHI@ESASTSP.BITNET
------------------------------
End of Info-IBMPC Digest
************************
-------