Info-IBMPC@USC-ISIB.ARPA (Info-IBMPC Digest) (02/08/86)
Info-IBMPC Digest Saturday, 8 February 1986 Volume 5 : Issue 19
This Week's Editor: Richard Nelson
Today's Topics:
ARC - Freeware File Compression
AT 1.2Mb Drive Failure Not Maxell Media
UNIX|STAT Ordering Information
More About the PC RT (Than You Ever Wanted to Know...)
Beware of 'MVDIR.BAT' (& all self-documenting code)
Microsoft Quick-Basic 1.00 Known Bugs
Undocumented Microsoft LINK Option: /E
Microsoft Assembler
Alsys Ada on AT
Windows with Mouse; Norton Editor
The CHMOD Woes
XTREE Review
Plotting Overhead Transparencies
Hard Disk Optimizer
AT to VAX Serial Problem
XENIX outb() Function Not Found
Parallel Transfer Rate of AT Query
Columbia BIOS ROM Query
Film Digitizing Query
DBASE III+ Query
Digitizing Tablets Query
Apple Laserwriter with MS-DOS Query
----------------------------------------------------------------------
Date: 7 Feb 1986 17:33:53 PST
Subject: ARC - Freeware File Compression
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
We will shortly be getting a new version of Hack for our lending
library. For those that may have forgotten Hack is sort of a super
Rogue (a screen oriented adventure game). I understand from Tim
Iverson (iverson%cory@BERKELEY) the source code is some 280K. Koji
will let the world know when we have transferred the files.
The only reason I mention this before the fact is that we are getting
the file in ARC format. I have seen several messages in the digest
discussing differences between ARC 5.0 format and ARC 4.0 format. I
didn't know what these messages were referring to until now.
ARC is a freeware program from System Enhancement Associates which
compresses files of any kind (by one of four compression techniques)
and stores the compressed files in one large archive file. It is just
the ticket for something like Hack which involves dozens of files and
hundreds of thousands of characters.
If anyone is interested in this program, I have placed the file
ARC500.DOC in our info-ibmpc lending library. INFO-IBMPC still
maintains its policy of not distributing freeware, and all program
submissions must be in clear text source code.
------------------------------
Date: Thu, 6 Feb 86 8:50:37 EST
From: "Lawrence D. Sher" <sher@BBNJ.ARPA>
To: Info-IBMPC@usc-isib.ARPA
Subject: AT 1.2Mb Drive Failure Not Maxell Media
In Vol. 4 Issue 105 of this Digest (14-Sep-85), I reported that 2 of 10 Maxell
floppies (high capacity type MD2-HD for the PC AT) had failed. After two more
failed, I tried the Dysan equivalent, whereupon one of those failed. I called
Maxell and explained what had happened on the chance that they would see some
self-interest in understanding the problem. They were extra-ordinarily
helpful, and I must say that were every company half as helpful, we would have
a much more serene Digest to read. Witness:
They offered to send two of my failed diskettes to their lab for detailed
examination. I sent the diskettes and (after a month) they replied with a
telephoned, then a written(!!!), report including micrographs and oscilloscope
tracings. The report shows that the head of my seldom-used IBM high-density
diskette drive is canted so as to scratch track 0 into track heaven. The two
failed samples (one Maxell, one Dysan) show identical scratches on the 200x
photographs.
There are several lessons for us all:
1. A new diskette drive can have an insidious diskette-killing disease.
2. Media failures should not automatically be blamed on a "bad batch".
3. Maxell has shown itself to be a paragon of an end-user-oriented company.
------------------------------
Date: Wed, 5 Feb 86 10:01:36 est
From: "Bennett E. Todd III" <ecsvax!bet%mcnc.csnet@CSNET-RELAY.ARPA>
To: info-ibmpc@usc-isib.ARPA
Subject: UNIX|STAT Ordering Information
Turns out the announcement of the availability of UNIX|STAT for MS-DOS
went out early in November of 1985, and I cannot find machine-readable
for it anywhere, so I will just retype the crucial information:
Requirements to run UNIX|STAT:
MS-DOS
at least 96K RAM
floppy drives
Of course, more memory and a hard disk make things (substantially) nicer
to use, but isn't that always the case?
Send a check or money order for $15 in US funds (no personal notes or
invoices) made out to Gary Perlman. Include the return mailing address
in *exactly* the following format:
Gary Perlman
School of Information Technology
Wang Institute of Graduate Studies
Tyngsboro, MA 01879 USA
Distribution is via US Mail *only*. You can also write him for more
information.
-Bennett Todd
Bennett Todd -- Duke Computation Center, Durham, NC 27706-7756; (919) 684-3695
UUCP: ...{decvax,seismo,philabs,ihnp4,akgua}!mcnc!ecsvax!duccpc!bet
------------------------------
Date: Wed, 5 Feb 86 22:03:21 est
From: "John R. Levine, P.O.Box 349, Cambridge MA 02238-0349 (617-494-1400)" <ima!johnl%cca-unix.arpa@cca-unix.arpa>
Subject: More About the PC RT (Than You Ever Wanted to Know...)
To: info-ibmpc%usc-isib.arpa.ima!net.micro.pc@cca-unix.arpa
I just got a copy of the PC RT "technology book" which contains 157 pages
all about the RT's hardware and software, including the VRM and AIX but
not 4.2BSD. Order SA23-1057, "IBM RT Personal Computer Technology" from
your local IBM branch office.
John Levine, ima!johnl, Levine@YALE
------------------------------
Date: Thu, 06 Feb 86 02:46:28 EST
From: James H. Coombs <JAZBO%BROWNVM.BITNET@WISCVM.WISC.EDU>
To: info-ibmpc@usc-isib.ARPA
Subject: Beware of 'MVDIR.BAT' (& all self-documenting code)
Info-IBMPC Digest V5 #15 (01/28/86) carried a notice that gave
directions for moving "an entire directory." The procedure involves
creating a 'MVDIR.BAT', which invokes the public domain 'MV.DIR'
(presumably the one by Harry G. McGavran, Jr., version 3.00).
While I'm sure that the author of the notice has had lots of success
with this technique, one of my colleagues destroyed our hard disk
directory on the first try. Don't ask me how. All files were moved
into one directory, which ended up with bad links--an error message that
I have never seen before. When I restored the original directory
structure, I ended up with cross-linked files and truncated entries. I
restored a couple of the files but lost a directory. CHKDSK recovered a
lot of chains that were unrecognizable, of course. Finally, I can't
boot off the disk. I found that COMMAND.COM was degraded and restored
that. Still no help. It will be quicker to just reformat the damned
thing and start from scratch.
So far, 'MVDIR.BAT' has cost me two hours. If I'm lucky, I can get
everything set up again in another two hours, plus diddling this and
that for several weeks. I hope no one else has shared this experience,
and I am writing in the hope that no one else will.
Finally, I have to say that I am not pleased to find an uncommented,
unexplained BAT file in INFO-IBMPC. I don't have the time to work
through it to figure out what it is supposed to do, and I am sure as
hell not going to try it out. I suppose that the author might feel that
the code is "self-documenting"; well, that is why I would never have
tried the damned thing in the first place. Unfortunately, my
colleague is not a programmer and has not been taught to recognize and
revile something that is not well-thought out.
For those who do not have technical backgrounds, I strongly encourage
that you do not try out public domain software unless you understand
how it works or have had it "certified" by a local expert.
------------------------------
Date: Wed, 5 Feb 86 05:50:30 PST
From: Ya'akov_Miles%UBC.MAILNET@MIT-MULTICS.ARPA
To: info-ibmpc@USC-ISIB.ARPA
Subject: Microsoft Quick-Basic 1.00 Known Bugs
List of Known Bugs in Microsoft Quick-Basic Compiler Ver 1.0
------------------------------------------------------------
I currently have Microsoft Quick-Basic Compiler 1.00 in my possession, and have
placed an order for Microsoft Quick-Basic 2.00 with JMG International. I found
the following bugs in Ver 1.00 and would like to know if they have been fixed:
o BASCOM /C:size accepted, but not documented. Does it work?
o BASCOM /V checks each STATEMENT for events. Docs say /V checks each LINE
o BASCOM /W does not allow ON KEY GOSUB/ON COM GOSUB/... etc. at run-time
o BREAK sent to (opened) COM1: or COM2: makes QUICK-BASIC program to crash
Please reply to this message...
------------------------------
Date: Wed, 5 Feb 86 06:07:10 PST
From: Ya'akov_Miles%UBC.MAILNET@MIT-MULTICS.ARPA
To: INFO-IBMPC@USC-ISIB.ARPA
Subject: Undocumented Microsoft LINK Option: /E
There exists an undocumented(?) switch to Microsoft LINK.EXE ver 3.XX,
which will cause an automatic compaction during binding. This process
will eliminate storage for uninitialized arrays from the .EXE file
produced by the linker, reducing the .EXE file size by up to 300 percent!
To use this feature, specify the /E option to the command line, eg
LINK myprog/E;
should work. For example, PCKERMIT.EXE ver 2.27 was 80K when linked normally
but shrunk down to 33K when linked with the /E option...
------------------------------
From: dob%ihmax.UUCP@brl
Subject: Microsoft Assembler
Date: Sat, 1-Feb-86 06:41:31 PST
ReSent-Date: 6 Feb 1986 14:18:37 PST
ReSent-From: Richard Gillmann <GILLMANN@USC-ISIB.ARPA>
ReSent-To: info-IBMPC@USC-ISIB.ARPA
> >I think that the following should assemble without problems:
> >PUBLIC foo
> >foo PROC NEAR
> > mov [si],OFFSET bar
> >foo ENDP
...
> > error 35: Operand must have size
>
> The only work-around I found was:
> mov [si],Word Ptr (OFFSET bar)
> which looks funny, but assembles correctly. The parentheses are
> necessary, as:
...
Won't this work?
mov word ptr [si],OFFSET bar
Daniel M. O'Brien (ihnp4!ihmax!dob)
AT&T Bell Laboratories
IH 4A-257, x 4782
Naperville-Wheaton Road
Naperville, IL 60566
------------------------------
Date: Fri, 7 Feb 86 14:37:52 PST
From: larry@Jpl-VLSI.ARPA
Subject: Alsys Ada on AT
To: info-ibmpc@isib.arpa
I received info on a full-Ada compiler for the PC-AT today. Alsys Inc. is
the vendor. It's headed by Jean Ichbiach, the Ada design-team leader. The
compiler looks very interesting; I intend to give it a close look at the
SIGAda/AdaJUG conference the last week of this month in L.A. Alsys says they
will be demoing their AT, Sun, and possibly Apollo versions of the compiler.
The host is an AT with a hard disk, DOS 3.x, and at least 512K of memory.
The compiler comes with a ~3 MB memory board that fits in a full slot. You
run under PC-DOS, using your own editor to create source files. Source
files are stored in a library created and maintained by an Alsys-proprietary
library manager. You could, of course, edit your files on an XT or any
other system that creates ASCII files you could copy to your AT. Possibly
even the library manager would run on an XT or PC compatible. Compilation,
however, has to be done on the AT.
The Alsys linker binds into the load module a run-time executive which handles
dynamic memory allocation/freeing. If you want to create programs which do
parallel processing with Ada tasks, you would bind a larger alternate run-time
executive which also implements tasking . The resulting module will run on a
PC or compatible under MS-DOS 2.x. If you want to do any floating point the
target must also have an 8087/80287 chip. No mention is made of a symbolic
debugger of any kind, a serious lack if there really is none.
If the target is an AT with more than 512K of memory you can have it run in
virtual mode (this is NOT virtual memory, of course, but the 16 MByte direct
memory addressing feature of the 80286). There's no need to recompile to have
your program run in real or virtual mode; I assume this is a linker option.
Alsys claims the object code is about as compact and fast as C code and give
a few benchmarks to support that. The size of the run-time executive isn't
given, however. My guess is that the executable is probably a good deal
bulkier than the C executable if it does parallel processing, due to the
tasking part of the alternate run-time exec. (The dynamic-memory management
code in the executive should be comparable to the C code in size.)
Alsys says they'll soon pre-validate the AT compiler system under the Ada
verification suite 1.6 as a preliminary to getting an official validation from
DoD. They say the total cost including the ~3MB memory will be under $3000.
If the compile and execution speed and load-module size isn't too bad, my
section will probably buy a copy. That combined with Alsys' computer-aided
instruction "Lessons on Ada" ought to make a good educational tool.
After the conference I'll pass on any interesting info.
Larry @ JPL-VLSI.arpa
------------------------------
Date: Fri, 7 Feb 86 08:22:38 PST
From: walton%Deimos@Hamlet.Caltech.Edu
Subject: Windows with Mouse; Norton Editor
To: info-ibmpc@Hamlet.Caltech.Edu
I just got a Microsoft Mouse for use with Windows and my Zenith Z-151. I
have seen the future of DOS, and it is good (to misquote PC magazine). I
have come to agree with those who have said that Windows without a mouse is
not a good thing. Windows with a mouse is rapidly becoming indispensable
to me. Now that a means exists for making any program RAM-resident,
perhaps we can hope to see Borland Lightning for Windows...
On another topic--I just ordered my copy of The Norton Editor. It was
reviewed positively in PC magazine, saying essentially the ads were
correct--namely it is fast, great for Pascal and C, and can do simple word
processing. PC griped a bit about the fairly limited printer support, but
whatta ya want for 50 bucks? I would be interested in corresponding with
other users of this editor on tricks and customization.
Steve Walton
walton%deimos@hamlet.cit.edu
swalton@caltech.bitnet
------------------------------
Date: 6 Feb 1986 02:59:43 PST
Subject: The CHMOD Woes
From: Craig Milo Rogers <ROGERS@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
CHMOD (PC-DOS interrupt function 43H) negligently trashes
register AX on a non-error return from subfunction 00H (Get File
Attributes). My DOS Technical Reference says "AX Error codes if
carry flag set". I interpreted that to mean no change if carry
flag not set. I was wrong.
Craig Milo Rogers
------------------------------
Date: Fri, 7 Feb 86 08:22:15 PST
From: walton%Deimos@Hamlet.Caltech.Edu
Subject: XTREE Review
To: INFO-IBMPC@Hamlet.Caltech.Edu
[Forwarded for Art Chodos, who is not on the ARPAnet.]
"I am not on the network but I do read the PCINFO documents put out on
PCPUBLIC by Educational Computing. [Submitter's note: there is an AT
running Xenix on campus which is our central repository for PD software,
FreeWare, and ShareWare.] In PCINFO8 there were a lot of comments about
writing programs to change file attributes and rename subdirectories. An
extremely powerful, inexpensive program exists which does all that and
more and it seems to me that it would save a lot of people considerable
effort. The program is XTREE by Executive Systems, 15300 Ventura Blvd.,
Suite 305, Sherman Oaks, CA 91403 800-634-5545 or 818-990-3457 $49.95
list. It allows you to see the tree structure of your disk and examine
the files within directories or subdirectories or it will show you ALL the
files on your disk. Files may be sorted by name, date, size, or
extension. Directories and subdirectories may be created, deleted or
renamed. File display can include attributes, date of creation, size, and
so on. Attributes can be changed. Files can be tagged so that only
tagged files will be acted upon. DOS can be run from within XTREE. I
could not get along without it. I would appreciate it if you could make
this information available on the network. I have no connection with
Executive Systems--I'm just a satisfied user."
Steve Walton
walton%deimos@hamlet.cit.edu
swalton@caltech.bitnet
------------------------------
From: Roy Stehle <stehle@sri-tsc>
Date: 7 Feb 1986 0838-PST (Friday)
To: Bohnsa@wharton-10
Cc: Info-Ibmpc@usc-isib, Stehle@sri-tsc
Subject: Plotting Overhead Transparencies
Advanced Graphics Software offers SlideWrite. I have tried it with an HP
plotter and found it to be a fairly good product. They can provide you
with a trial disk (which is how I tried it).
If you get the trial disk, you will probably get a coupon worth $30 off the
list price of $225. They offer a 30 day money back guarantee.
SlideWrite offers: 2 Fonts (Modern, Roman)
3 Weights (Light, Medium, Bold)
5 Sizes (Tiny, Small, Medium, Large, Giant)
4 Styles (Italic, Color, Underline, Outline)
4 Symbols (Dot, Box, Triangle, Checkmark)
4 Rules (Dotted, Thin, Medium, Heavy)
Graphics (Boxes, Circles, Arrows, etc)
What You See Is What You Get
Contact: Advanced Graphics Software Inc.
333 W. Maude Avenue, Suite 105
Sunnyvale, CA 94086
(408) 749-8620
------------------------------
Date: Fri, 7 Feb 86 13:10:13 cst
From: knutson@huey.UTEXAS.EDU (Jim Knutson)
To: info-ibmpc@usc-isib.ARPA
Subject: Hard Disk Optimizer
About the cheapest optimizer you can get is the BACKUP/RESTORE
programs. Just backup your hard disk, zero it out and restore all the
files. They will all have contiguous allocations. This, however, does
not move the most frequently accessed files together (to reduce seek
time) but then I don't know of any optimizers that do this automatically
either.
Jim Knutson
ARPA: knutson@ngp.UTEXAS.EDU
UUCP: {ihnp4,seismo,kpno,ctvax}!ut-sally!ut-ngp!knutson
Phone: (512) 471-3241
------------------------------
Date: 7 Feb 1986 09:30-EST
Subject: AT to VAX Serial Problem
From: MHARRIS@A.BBN.COM
To: Info-IBMPC@USC-ISIB.ARPA
Cc: jpg@SRN-VAX.BBN.COM, MHarris@A.BBN.COM
Perhaps there's something I don't understand (likely), and perhaps this has
already been discussed here (although I don't recall seeing it), but:
I have a vanilla AT with AST Advantage card (2 serial ports) and need to
make a serial line connection directly (not via modem) to a vanilla Vax 750.
Until recently, I've had this connection running successfully by converting
to current-loop at the Vax end and converting back to EIA at the AT end. But
I just moved the AT closer to the Vax and removed the current-loop stuff; now
it doesn't work. On poking around with a 'scope, I discovered that the AT
serial card is putting out 0 to +6 volt signal levels; the RS232 standard
calls for something like -3 to +3 volt levels. The current-loop converters
were being friendly and making the right thing happen, but apparently the
Vax serial port hardware won't deal with it. So:
1. Do I understand correctly what's going on?
2. Same problem if I use IBM serial/parallel adapters?
3. Any solution other than going back to current-loop?
Many thanks.
-- Michael Harris MHarris @ BBNA
BBN Labs
617-497-3794
------------------------------
Date: Tue, 4 Feb 86 11:02:58 est
From: rochester!kodak!gardos@seismo.CSS.GOV (tom gardos)
Subject: XENIX outb() Function Not Found
To: rochester!seismo!usc-isib.arpa!info-ibmpc@seismo.CSS.GOV
I am trying to write a device driver in C for the IBM PC AT running XENIX.
The program requires use of the functions out() and outb(), however the C
compiler cannot find them. Has anyone had this problem? Does anyone know where
the functions are defined?
Please respond by mail and I will summarize the responses. Thanks.
Tom Gardos
------------------------------
Date: 5 Feb 86 15:00:45 PST (Wednesday)
From: mwilliams.es@Xerox.COM
Subject: Parallel Transfer Rate of AT Query
To: <INFO-IBMPC@USC-ISIB.ARPA>
cc: mwilliams.es@Xerox.COM
Hello,
I am looking for the data transmission rate for the IBM AT when running
in parallel mode. I am not interested in its burst rate but the rate
for a document > 20,000 bytes. If anyone knows, I would greatly
appreciate it. Thanx
Mitch
------------------------------
From: bwong2%amovax.uucp@UCI.EDU
To: ucivax!INFO-IBMPC%USC-ISIB@amovax
Subject: Columbia BIOS ROM Query
Date: 05 Feb 86 15:46:54 PST (Wed)
A friend of mine happens to have a Columbia computer. He
wanted to upgrade the system to the new version of WordPerfect but
was unable to do so because he can only run an old version of DOS
because he has a outdated BIOS ROM. Does anyone know of a compatible
ROM that he could buy? I thought that a company called Phoenix sold
one but then again I am not sure. Many Tanks.
Bradley Wong
UUCP: ucbvax!ucivax!amovax!bwong2
------------------------------
Date: Thu, 6 Feb 1986 12:26 EST
From: Reggie Aurora <AURORA%SBBIOVM.BITNET@WISCVM.WISC.EDU>
To: <info-ibmpc%usc-isib@wiscvm.arpa>
Subject: Film Digitizing Query
Hello:
We are interested in setting up a scanning/digitizing system from
which we can obtain intensity and positional (X,Y coordinates) off film
negatives. Any one who has set up a system that uses a video camera or
alternatively a scanner type of system and can provide us with information
or specifications, would be of great help. If this is of sufficient interest
to other netlanders I will post to the network. Thanx Much!
-- Reggie.
------------------------------
Date: Thu, 06 Feb 86 14:00:10 CST
From: C449499%UMCVMB.BITNET@WISCVM.WISC.EDU (Randy Davis)
To: INFO-IBMPC@USC-ISIB.ARPA
Subject: DBASE III+ Query
Has anybody seen an actual working copy of this new
'wonderful' update? Is it really worth shelling out
another $100? If all I get is the speeded up indexing
time, I might as well buy QuickIndex by Fox & Geller.
Randy Davis
University of Missouri - Columbia
------------------------------
Date: Thu 6 Feb 86 21:25:13-PST
From: Brian Reilly <CON.REILLY@SU-GSB-HOW.ARPA>
Subject: Digitizing Tablets Query
To: info-ibmpc@USC-ISIB.ARPA
A friend of mine is interested in purchasing a digitizing
tablet for his IBM-XT. If anyone has any experience with such
products, or can recommend one, please send me a note and I will
forward it on to him. I'm not on this list, so please send directly
to me. Thanks.
- Brian Reilly
con.reilly@su-gsb-how.arpa
------------------------------
To: info-ibmpc@usc-isib.csnet-relay
From: tcr@mhuxi.btl
Date: Fri 7 Feb EST 1986 09:39
Subject: Apple Laserwriter with MS-DOS Query
We are obtaining a Apple Laserwriter and would like to be able to use it
with both the Macintosh and with MS-DOS. Switching from one to
another electrically is no problem, but what we need is a printer
driver for MS-DOS. If it could be used to print simple ASCII text,
that would be good. If it could emulate a printer like the Diablo,
that would be even better.
Has this been done before and if not, any suggestions?
Tom Russell
AT & T Bell Labs, Murray Hill, NJ
(201) 582-7578
Usenet: ihnp4!mhuxi!tcr
CSnet: tcr.mhuxi@btl
------------------------------
End of Info-IBMPC Digest
************************
-------