Info-IBMPC@B.ISI.EDU (Info-IBMPC Digest) (08/23/86)
Info-IBMPC Digest Friday, August 22, 1986 Volume 5 : Issue 78
This Week's Editor: Richard Gillmann
Today's Topics:
UULINK (3 msgs)
Wire Wrap Boards (2 msgs)
Display-Write File Recovery
DSI-0 Board
Conversion from Basic to ASCII
Lotus 123-1A EGA Driver Available
3278/79 Cards and Structured Fields
Basic and Dos 3.2
IBM Technical Reference Guide List
Zenith Data Systems BBS
Professional FORTRAN
three-part-harmony.bas / shaped note hymn
Golden Bow System's VCache
3Comm Networks
Today's Queries:
Small Agricultural Programs Wanted
Zilog 8530 and X.25 Queries
Foreign Language Fonts Wanted
Turbo Pascal Compiler Limit Problem
Turbo Pascal BBS Code Wanted
Multi-Tasking Query
Intermittent 1701 error
Word Perfect 4.1 Problem
EGA Screen Dump Wanted
Disk Problems
Problem with Hercules in a Compaq
MSKERMIT query
Why does IBM async hold up CD?
----------------------------------------------------------------------
Date: Wed, 23-Apr-86 21:23:49 PDT
From: bcsaic!asymet!library@uw-june.arpa (mailing list readers)
Subject: Lauren's MSDOS uucp
To: uw-june!usc-isib.arpa!info-ibmpc@UNKNOWN!asymet
I'm sending this mail from a PC/AT with Lauren's UUCP workalike (UULINK).
I can't give a real review because we use such a small fraction of the
program's functionality. I can tell you that it runs smoothly and without
tinkering for polling one remote site and exchanging mail and (small amounts
of!) Usenet news. This is more than can be said for some versions ATT
distributed.
Support: I have little data. After getting installation instructions I
have had almost no occasion to ask Lauren anything. It took over a week
to get hold of Lauren to ask about installation procedures, but Lauren
tells me that my experience was exceptional.
Anyone who would like to hear more should send me direct mail.
Fred Wamsley bcsaic!asymet!fred@uw-june
------------------------------
Date: Fri, 25-Apr-86 16:02:58 EDT
From: David Farber <farber%pcpond.pc.udel.edu@Louie.UDEL.EDU>
Subject: UULINK
To: info-ibmpc%usc-isib.arpa@louie.udel.EDU
In past Info-IBMPC Digests there has been reference to Lauren
Weinstein's UULINK package. I have been using it for the past 3+
months from my AT at home. It works in my environment into a MMDF
channel on one of the Udel systems and then onto the internet. To say
I have been pleased with it is an understatment. It allows me to
survive the combined world of the VAX and DOS. I run it in background
under Desqview with some accessories my son put together to notify me
in forground when mail has arrived. Its all real fun. I have tried
uulink on the HP Vectra, the Toshiba T1100 etc and it works just fine
there also.
Dave
------------------------------
Date: 20 Aug 1986 13:41:12 PDT
Subject: UULINK Information
From: Billy <BRACKENRIDGE@B.ISI.EDU> (and GILLMANN@B.ISI.EDU)
Lauren Weinstein's UUCP program for PCs is available as a commercial
product. It's called UULINK(tm) and allows you to put your PC on
Usenet using the UUCP protocol. For details, contact Lauren
Weinstein at Vortex Technology, (213) 390-3920. Please don't use the
net to make inquiries about this product.
------------------------------
Date: 18 Aug 86 14:26 EDT
From: Shoots.wbst@Xerox.COM
Subject: Wire Wrap Board Query
To: INFO-IBMPC@B.ISI.EDU
Where can I buy PC/XT compatible, full length wire wrap boards?
(I already know about I.B.M.'s boards, and have seen boards from
JDR Microdevices, San Jose, CA. These are both basically "perf"
boards, with no wire wrap pins in them.) I would like to find some boards
which have been staked with wire wrap pins 0.1" apart, in columns
spaced 0.3" apart, or something similar.
Thanks,
Jim Shoots
------------------------------
Date: 20 Aug 1986 12:47:31 PDT
Subject: Re: Wire Wrap Boards
From: Billy <BRACKENRIDGE@B.ISI.EDU>
To: Shoots.wbst@XEROX.COM
cc: info-ibmpc@B.ISI.EDU
I bought wire-wrap prototype boards from Priority One Electronics in
Chatsworth (although they have other So. Cal. branches). Of course the
wire wrap pins are on the sockets not on the board. Priority One has
a selection of sockets in various pin lengths as well as pre cut and
color coded wires etc.
------------------------------
Date: Mon, 18 Aug 86 18:22:57 edt
From: Mike Ciaraldi <ciaraldi@rochester.arpa>
To: info-ibmpc@usc-isib.arpa
Subject: Display-Write File Recovery
Thanks to the various people who sent me answers to my recent query
about DW3 file recovery. Most suggested using the "recover" command,
which did not work in this case, but which they report worked
fine for them. I came up with a partial solution, which I offer here.
I tried poking around in the disk file (which DW3 rejected with the
message "Disk Error") with a disk editor, the Ultra Utilities. I
found nothing recognizable. So, I sent the file up to a Unix (tm)
system using Kermit, and ran it through the "tr" and "dd" programs.
First I converted EBCDIC to ASCII. Looking at the file I saw chunks
of text separated by garbage. Then I converted all control characters
except CR and LF to asterisks. I downloaded the resulting file to the
PC using Kermit again. Then I started up DW3 with a new blank
document and used the "Get ASCII" command to import an ASCII text
file.
I could now manipulate the file with DW3. I did the "Repage" command,
and discovered that the last several pages were total gibberish.
I deleted those, and wound up with a 30-page document, presentable except
for the interspersed blocks of asterisks and garbage characters.
I was prepared to chop those out, when I realized that the blocks of text
broke at arbitrary places (middle of sentences, middle of words, even), and
that they were not in the same order as the original document!
At last the light dawned! I was actually looking at what might be
parts of several files, interspersed. I mentioned this to someone
else, who said he had heard that DW3 manipulates the FAT (File Access
Table) and directory on the disk DIRECTLY. If the program was
interrupted abnormally, it might leave all your files in an
inconsistent state (actually, this is possible with any program, but
only if you do something like turning off the machine while the
program is still running, so DOS doesn't have the chance to tidy
things up, as it does when exiting a program nurmally). DW3's direct
manipulations run the risk of causing much greater damage, since i it
might manipulate several files without writing the FAT and directory
info back to the disk. The clincher was when I mentioned to the lady
who originally brought the problem to me, "You didn't turn off the
computer while DisplayWrite was running, did you?", and she replied,
"Maybe THAT's what I did..."
The upshot of the whole thing is that some of the information on
the disk was recoverable, which is good news for some people,
I suppose.
The moral is that all users should be instructed to:
1) Make backup copies of all important files, and
2) Exit all programs normally, returning to the DOS "A>" prompt
before removing a disk and/or powering down.
Hope this helps someone!
Mike Ciaraldi
ciaraldi@rochester
seismo!rochester!ciaraldi
------------------------------
Date: 18 Aug 86 00:21:00 EDT
From: DAVE THOMAS <DAVE_THOMAS%CARLETON.BITNET@WISCVM.WISC.EDU>
To: <info-ibmpc@isi.ARPA>
Subject: DSI-0 Board
For those interested , we have purchased a couple of the DSI-020
boards as described in Byte July/August 86 We bought the 12.5Mhz
versions with 1MB. They have a DSI-780 {(20mhz and 4MB) for Sept!
We bought the SVS Pascal and C compilers. I plugged the board into
my old Zenith 150 and up it came! It made my old PC useable as
a 68K development system. The Pascal compiler(very standard a la ISO)
loads and compiles just like Turbo. Everything we have tried works
great. The board has a few jumpers, but looks very well done.
They provide great support via Bix(Dsi.32bit).
------------------------------
Date: Mon, 18 Aug 86 17:25:52 EDT
From: Edward_Vielmetti%UMich-MTS.Mailnet@MIT-MULTICS.ARPA
To: info-ibmpc@USC-ISIB.ARPA,
deba%umdc.bitnet@MITVMA.BITNET
Subject: Re: Conversion from Basic to ASCII
There is a program written by Vern Buerg called "Readbas" that takes
a tokenized BASIC file and spits out a reuglar ASCII file.
This is what you want, I think. How to get it to you? I don't have
sources, so I can't post to Info-IBMPC, but it is available on the
following public bulletin boards:
wipcus bbs 'readbas.arc' (313) 663-1835
vor rbbs 'readbas.lbr' (415) 994-2944 (the author's bbs)
The first one is in Michigan (Ann Arbor), the other is in California.
If anyone is interested in putting these binaries in a place where
they are accessable via anonymous FTP, contact me and I will FTP
files to you.
Edward Vielmetti, Computing Center Microgroup, University of Michigan
arpa: emv%UMich-MTS.Mailnet@MIT-Multics.ARPA
uucp: ihnp4!umich!umix!emv
new-style: emv@umich.cc.umich.edu
emv@umix.cc.umich.edu
or on Mike Myers' m-net public access unix, (313) 994-6333
------------------------------
Date: 19 August 86 10:18 EDT
From: T7Y%CORNELLC.BITNET@WISCVM.WISC.EDU
Subject: Lotus 123-1A EGA Driver Available
To: INFO-IBMPC@B.ISI.EDU
Check out the Boston Computer Society BBS (area code 617 237-3750).
Enter the File Section and then File Area 16 - Spreadsheets. In a few
days there should be a file named 123EGA.ARC in that directory.
123EGA.ARC contains the screen drivers that 1-2-3 v1a needs for EGA
use along with a batch file that will load them onto your disk(s).
You will also need a copy of ARC51.EXE to decode the file (should also
be available on the BBS).
Tom Abdella
Tufts University
------------------------------
Date: Tue, 19 Aug 86 17:11 EDT
Subject: Re: 3278/79 Cards and Structured Fields
From: Michael_Evans_S-E-Banken%QZCOM.MAILNET@MIT-MULTICS.ARPA
To: Info-IBMPC Digest <Info-IBMPC@B.ISI.EDU>
> I would like to transfer data to and from an IBM mainframe using the
> IBM 3278/79 card and 'OEM data' structured fields.
> Unfortunately, neither of the available programming interfaces
> supports this. Doe anyone know how to program the card directly to
> send and receive this types of structure fields?
I cannot say exactly how you can do this but can maybe supply some
pointers.
As far as I know their are two requirements for a PC to process
structured fields:
1) The port on the 3274 controller must be defined as a DFT
(? Distributed Function Terminal) port. The controller then allows
more of the SNA session (eg the structured fields) through to terminal
than it would for a vanilla 3278/9.
2) The 3278/9 card/driver must be able to take care of all this extra
information. I am not sure whether the old IBM 3278/9 card could do
this but the IBM card combined with the PC-3270 software which was
announced on April 2 definitely CAN support a DFT connection (at least
at some point in the future when the right software is delivered). The
expensive 3270-PC can use structured fields (with release 2 or later
of the control program). Therefore it would seem to be possible. I do
not remember if this uses the same 3278/9 card as the above.
I would suggest that you get as much information as possible about the
3270 products in the April 2 announcement as see what fits your
configuration.
------------------------------
Date: Wed 20 Aug 86 16:20:25-EDT
From: David.Zubrow <David.Zubrow@C.CS.CMU.EDU>
Subject: Basic and Dos 3.2
To: info-ibmpc@B.ISI.EDU
I recently installed Dos 3.2 on a new AT and also copied the Basic
interpreters onto my system. Now whenever I run Basic, it starts
filling the screen with an endless stream of a seemingly arbitrarily
chosen character. Hitting 'break' stops the character output and
basic seems to run properly from then on. Anyone have any ideas on
how to fix this problem?
Thanks in advance,
Dave Zubrow
------------------------------
Date: Tue, 19 Aug 1986 15:39 EDT
From: Ben Yalow <YBMCU%CUNYVM.BITNET@WISCVM.WISC.EDU>
Subject: Re: IBM Technical Reference Guide List
To: Deba Patnaik <DEBA%UMDC.BITNET@WISCVM.WISC.EDU>
cc: <info-ibmpc@usc-isib.arpa>
The best bet for getting a list of all the PC technical reference stuff
is an IBM booklet called the IBM Technical Directory Vol II. It
contains a list of all the books/reference materials that you can
get from IBM, along with their prices. The date on mine is 4/86.
The book says you can get more copies of the Technical Directory by
writing or calling IBM at:
IBM Technical Directory
PO Box 2009
Racine, WI 53404
1-800-426-7282
They say you should allow 1-2 weeks for delivery on call-in orders, 2-4
weeks for mail-in.
(This is only for the stuff you buy from them; it doesn't include the
PC Seminar proceedings)
------------------------------
Date: 21-Aug-1986 0246
From: fulton%donner.DEC@decwrl.DEC.COM (Cathy Fulton -- CXO Technical Training)
To: info-ibmpc@usc-isib.ARPA, fulton%donner.DEC@decwrl.DEC.COM
Subject: Zenith Data Systems BBS
I have available the entire contents of the Zenith Data Systems Bulletin
Board System. The ZDS BBS has a wealth of information not only specific to
users of ZDS products, but also for users of MS-DOS and applications in
general. There is a lot of information on both software and hardware topics
such as compatibility, performance, and specific operational problems. There
is a large history of questions and answers; many of the answers are very
detailed and often provide code patches and solutions.
The information in the ZDS BBS is freely available: when you log in, you are
not even asked to provide a name or password. The reason I am offering the
contents of the BBS is that it is very large and therefore requires a
substantial amount of download time.
I have the contents of the BBS available in uuencoded, ARCed form. In this
form, the BBS comprises about 385K. I have split this 385K into 6 files,
none of which is greater than 64K.
If anyone is interested in receiving this, please send mail.
- Cathy
uucp: ...decvax!decwrl!dec-rhea!dec-comet!fulton
ARPA: fulton%comet.DEC@decwrl
------------------------------
Date: 20 August 86 08:59-PST
From: DEP%SLACVM.BITNET@WISCVM.WISC.EDU
To: INFO-IBMPC@USC-ISIB.ARPA
Subject: Professional FORTRAN
This is in response to Mark Callaghan at the University of Western
Australia, concerning his complaint about bugs in IBM's Professional
FORTRAN. If his institution has a volume purchase agreement with IBM,
his designated "technical representative" may obtain all current
updates to PC products. I know, becuase I have, in fact, obtained
updates to the professional FORTRAN for SLAC.
I don't what he can do about getting these updates otherwise. It's hard
to believe that a bona fide purchaser doesn't have some way to obtain
these.
------------------------------
From: ted@imsvax.UUCP (Ted Holden)
Subject: three-part-harmony.bas / shaped note hymn
Date: 24 Jul 86 02:01:41 GMT
Most microcomputers are doomed to spend their lives in mundane
pursuits, if not as instruments of outright anti-culture, e.g.
spread-sheets, data-bases, pac-man, global-thermonuclear-war etc. This
doesn't mean they can't be put to better use; every PC, AT etc. out
there in the land has a voice which, while uninteresting in itself,
becomes interesting in combination with others of its kind, much like
recorders (the antique flute variety). Not surprisingly, recorder duets
and trios sound quite good on two or three PCs. Using Basica or GW
Basic insures that the PCs involved don't even have to be running at the
same MHZ rating; basic goes by the system clock.
The following are two soprano (or soprano and tenor, your
choice) lines and a bass line for "Babylon is Falling", my pick of the
so-called "shaped note" hymns from colonial days in America, when
itinerant singing masters went from town to town teaching singing
classes for a month or so at inns, schools etc., using a system of
shaped notes in which notes of different duration were represented as
different geometric shapes.
Simply cut out the three segments of basic "play" statements,
read each into a seperate PC, have three people each hit F2 at the
same time, and enjoy. If you want to really get serious about it, you
might try a slightly deeper speaker on the PC with the bass line.
10 PLAY "l4 o4 g g b g8 b8 >c8 <a8>d d d <g g b8 a8 g b b8 a8 b2
20 PLAY "g g b g8 b8 >c8 <a8 >d d d <g g b8 a8 g b b8 a8 b2
30 PLAY "b. a8 g g >c <b b g b. a8 g g >c <b b2
40 PLAY "l8 g. g16 g g b4 b g >c <a a a >d4 <a4 g. g16 g g b4 g b b4 b a b2
50 PLAY "l8 g. g16 g g b4 b g >c <a a a >d4 <a4 g. g16 g g b4 g b b4 b a b2
100 PLAY "o4 l4 b b g8 f+8e a a8 f+8 g a b b g8 f+8 e8 g8 f+ f+ e2
110 PLAY "o4 l4 b b g8 f+8e a a8 f+8 g a b b g8 f+8 e8 g8 f+ f+ e2
120 PLAY " b. >c8 d8 <b8 >d e f+ e8 d8 <b b. >c8 d8 <b8 >d e f+ e2
130 PLAY "<B8. B16 B8 B8 L8 G4 G E A4 A F+ G4 A4 B. B16 B B G4 E G F+4 F+4 e2
135 PLAY "B8. B16 B8 B8 L8 G4 G E A4 A F+ G4 A4 B. B16 B B G4 E G F+4 F+4 e2
200 PLAY "l4 o4 e e <b >e d d <b >d g e e e <b b >e2
210 PLAY "l4 o4 e e <b >e d d <b >d g e e e <b b >e2
220 PLAY " G. A8 B G E <B >E E G. A8 B G E <B >E2
230 PLAY "e8. e16 e8 e8 <b > e8 e8 d d8 c8 <b >d e8. e18 e8 e8 e e8 e8 <b b >e2
235 PLAY "e8. e16 e8 e8 <b > e8 e8 d d8 c8 <b >d e8. e18 e8 e8 e e8 e8 <b b >e2
------------------------------
From: connery@bnrmtv.UUCP (Glenn Connery)
Subject: Golden Bow System's VCache
Date: 18 Aug 86 19:26:55 GMT
Just a couple of days ago I received the above disk caching software and
have been running it on my PC/AT. This is a write-through disk cacher
which can utilize regular memory, LIM/EMS expanded memory or PC/AT extended
memory for the cache. I have an AST Advantage board with 1.5M on it. After
rounding out the regular memory to 640K, this leaves me with 1408K to
play with. Allocating the entire thing to a disk cache instead of to the
DOS VDisk seems like the best idea from the few tests I've done. You avoid
all of the special case code in synonyms and batch files, shorten your
autoexec file, and avoid having to run special load batch files to load
say your microsoft c compiler, libraries and include files onto ram disk.
It won't be quite as fast the first time you try something, but upon
subsequent uses it appears to be just a little more efficient than VDISK.
Plus you get the benefit of caching of data files as well.
Currently I have the system set up for 1M of VCache and 384K of VDISK,
which I use just for the TMP files created by the various passes of
the Microsoft C compiler. This seems to be a marginal trade off, but
pays off a little for large compiles. For those of you who are
interested it has sped up my whole system quite noticably, except for
the compile time on Microsoft "C", which hasn't changed much, i.e.
it's still abominably slow.
Anyway, at this point I strongly recommend the product for those with LIM
or extended memory boards. Its available from Golden Bow Systems, P.O.
Box 3039, San Diego, California 92103, (619)298-9349 for $80.00 +TAX +S&H.
Glenn Connery, Bell Northern Research, Mountain View, CA
{hplabs,amdahl,3comvax}!bnrmtv!connery
------------------------------
Date: 22 Aug 1986 14:33:52 PDT
Subject: 3Comm Networks
From: Billy <BRACKENRIDGE@B.ISI.EDU>
To: info-ibmpc@B.ISI.EDU
cc: lotto@HARVARD.HARVARD.EDU
I thought I'd add my two cents to Gerald Lotto's network survey.
We run both 3Comm Ethershare and 3+ software here at USC/ISI. As was
pointed out in the earlier message 3+ code uses the Microsoft
redirector and IBM netbios interface spec to provide a file system.
The earlier Ethershare code just emulated remote disks at the BIOS level.
The remote disk emulation was difficult from a systems operation
standpoint in that we had to completely back up all disks every week
even if only one file changed. Now with the new file system we back
up only the files that have been changed.
We have several applications which use Dbase and we would like to
make them accessible to multiple users. 3+ implements the DOS 3 file
locking features which makes this all possible.
Performance has improved although that was never much of a problem as
our PC applications aren't particularly disk intensive.
The bad news is that 3+ is a memory pig. My operating system takes
more than 260K now. The SUN network file system provides the same
functionality as the 3Comm code in about 100K less code. Also 3Comm
masks off interrupts for longer than a 9600 baud character time. As
all the PCs in the institute are used as terminal emulators to talk
to time sharing systems this rules out using 3Comm 3+ for all but
experimental purposes.
The new code also appears less robust than the older ethershare
code. Our ethernet has a lot of traffic on it and sometimes things
get completely out of hand. In these situations the old
ethershare code would indicate a disk error code and I would
pound on the R for "retry" key until my file got read or written. The
3+ code either gives up or locks up in those situations. Since I got
a new ethernet transceiver this problem has disappeared, but the
point remains that the newer code is not as robust as the older
simpler code in the presence of net errors.
I have been able to communicate these problems to 3Comm. Their
customer service is a little bizarre in that they have an automated
telephone routing system, but it works. Networks are complicated
beasts, and given the size of the task they are doing a pretty good
job answering telephone problems and complaints.
We also have the SUN network file system on order and I will send in
my comments on that system after we have a chance to try it.
We also run the MIT IP/TCP code which works fine with both versions
of the 3Comm software. One can not access the remote file system while
running the MIT code, but as I have a large JRAM disk this isn't a
problem.
------------------------------
Date: Mon, 18 Aug 86 16:14 N
From: <INNO%HWALHW5.BITNET@WISCVM.WISC.EDU>
Subject: Small Agricultural Programs Wanted
To: info-ibmpc@usc-isib.arpa
I am interested in programs made for small projects in developing
countries. These programs might be applications written for a
standard package like Multiplan or Dbase, but also programs written in
Basic Fortran etc. are very welcome.
My main interest is small scale agricultural projects where a Personal
Computer is used for administration and calculations (statistical),
but other applications are also very welcome.
My adress is INNO@HWALHW5.EARN
My own name is Gert Lokhorst.
------------------------------
Date: Mon 18 Aug 86 21:38:35-EDT
From: Drew D. Perkins <DP4Q@TE.CC.CMU.EDU>
Subject: Zilog 8530 and X.25 Queries
To: tcp-ip@SRI-NIC.ARPA, info-ibmpc@B.ISI.EDU
I'm looking for a card for IBM PC's that supports syncronous/sdlc/hdlc
communications up to 64kb. What I have in mind should hopefully have
a Zilog SCC (8530) chip, interrupt capability, and fullduplex transmit
and receive DMA capability. An onboard processor would be nice but
not required. The intended application is running TCP/IP over the
line. I'm currently using the Tangent Technologies PC MacBridge card
which satisfies all requirements but the transmit dma capability. If
anyone knows of another card, PLEASE let me know, I'm getting
desperate!
Also, if anyone knows of any public domain X.25 LAPB code, I'd also
like to hear about it. Preferably written in C. I'd like to provide
flow control and error recovery on the 64kb link without reinventing
the wheel.
Thanks,
Drew
------------------------------
Date: 18 Aug 1986 22:18-EDT
Subject: Foreign Language Fonts Wanted
From: ABN.ISCAMS@D.ISI.EDU
To: INFO-IBMPC@B.ISI.EDU
Can anyone point me to public domain source for drivers, fonts, and/or
example code (Pascal, Assembler, C, in order of preference) for
keyboard, screen, and/or printer for the foreign languages Greek and
Arabic? I've got it in Hebrew, but don't have the heart to draw all
them new characters! Also, is there a "standard" ASCII table for
Greek and Arabic alphabets? I'm gonna crib the keyboard format from
foreign typewriters, but need to program in English.
Nope, contacting my friendly IBM service representative didn't help
here .. seems these are available in foreign countries, but NOT sold
or listed in this country (the US).
Thanks in advance,
David Kirschbaum
Toad Hall
------------------------------
To: info-ibmpc@b.isi.edu
Subject: Turbo Pascal Compiler Limit Problem
Date: Mon, 18 Aug 86 21:11:56 -0800
From: Alastair Milne <milne@ICSE.UCI.EDU>
I have just encountered a rather odd situation which leads me to ask
for the experiences of the net.
The compilation of a 5000-7000 line program was ceasing with an error
98, which, according to the manual, means there is no more variable
space available: presumably, as much has been allocated to the DS
segment as it can hold. The problem is that the error was occurring
in the middle of executable code. I thought it might be allocating a
temporary that was going over the limit, but that wasn't it; besides,
doesn't Turbo allocate temporaries in the stack segment, not the data?
The problem was relieved by making some of the larger procedures into
overlays. It was *not* relieved by cutting down the capacities of a
large number of string variables (therefore allocation of temporaries
could hardly be the problem).
Now the question: error 98, on the face of it, should refer only to
allocated data. Is it also in fact used for code segment overflows?
I don't mean a compiler overflow: this occurs when the symbol table
gets too large. I mean actually using up the entire 64K of space in
CS.
This would be a disappointing ambiguity of messages; but then, Turbo's
error reporting capacity has always been very weak.
Thanks in advance to anybody who can answer.
Alastair Milne
PS. The machine was an IBM AT, with loads of memory and hard disc space.
The Turbo version was 3. I don't know the precise subversion.
------------------------------
Date: 19 Aug 1986 22:16-EDT
Subject: Turbo Pascal BBS Code Wanted
From: ABN.ISCAMS@D.ISI.EDU
To: info-ibmpc@B.ISI.EDU
Has anyone converted one of the Turbo Pascal RBBS programs to run on a
PC compatible? I've downloaded the Turbo BBS from DEC-MARLBORO, but
don't have the strength to convert all the Rainbow-specific stuff!
No, I'm not interested in BASIC or .EXE code ... Turbo Pascal source
only.
Thanks in advance,
David Kirschbaum
Toad Hall
------------------------------
Date: Tue, 19 Aug 86 19:12 EDT
From: Ya'akov_Miles%UBC.MAILNET@MIT-MULTICS.ARPA
Subject: Multi-Tasking Query
To: info-ibmpc@B.ISI.EDU
We at TRIUMF are conducting a search for a suitable multi-tasking
real-time operating system to run on our IBM-PCs and IBM-PC/ATs
I have heard rumours that MS-DOS version 4 has already been released
in Europe, and that it supports multi-tasking. I have also heard
rumours that MS-DOS version 5 will support the virtual (protected)
addressing on the IBM-PC/AT.
Could anyone provide suitable details as to future releases of DOS:
o Real-time response to an interrupt?
o Context-switch time?
o Virtual (protected) memory mode used on IBM-PC/AT?
o Network support?
ID=TRYN%UBC.MAILNET@MITS-MULTICS
[Any issue of PC Week will provide you with a generous supply of
multi-tasking DOS rumors. -rag]
------------------------------
Date: Tue, 19 Aug 86 12:35:18 PDT
From: Bill_Krane%SFU.Mailnet@MIT-MULTICS.ARPA
To: INFO-IBMPC@B.ISI.EDU
Subject: Intermittent 1701 error
I have a Sysdyne 10 meg hard disk in an old PC-1 with an XT clone
power supply (150 watts). The disk has worked flawlessly for
almost a year now but lately I've been getting a 1701 error
message intermittently from a cold boot. I've had the controller
replaced (under warranty) and the power supply checked out. Some
further investigation showed that occasionally the disk does not
spin on power-up (hence the 1701). This behavior seems to occur
at random and otherwise it functions normally. Has anyone had a
similar experience with this type of drive? Any suggestions about
what the problem might be? I'm starting to get a little anxious
since my warranty runs out in a week or so.
Please reply directly to me as follows and thanks in advance:
Krane%SFU.MAILNET@MIT-MULTICS.ARPA
------------------------------
To: info-ibmpc@usc-isib.ARPA
Subject: Word Perfect 4.1 Problem
Date: Wed, 20 Aug 86 20:53:17 -0500
From: Mark H. Granoff <mhg@mitre-bedford.ARPA>
I think I have found a bug in Word Perfect 4.1, but I would like to
know if any one else has had a similiar problem:
My document is 20 pages in length. After doing a word count (and
having started at the top), I tried to go back to the top. Typing
[Home][Home][Up Arrow] caused a "Repositioning" message to be
displyed and then a line of garbage characters to be displayed at the
bottom of my screen, effectively hanging my system. (I also tried to
page up one page at a time and after about 4 or 5 pages the same
wonderful garbge appeared.) The only resident program in memory is
Sidekick. Interestingly, although I can't get Word Perfect to respond
to my keystrokes, they are buffered; if I then invoke SideKick, the
keystrokes that I typed are picked up by SideKick. For example, typing
'C' and then calling SideKick, I get the calculator.
I experienced this same problem when I had SideKick, CED (a memory
resident command line editor), and SCRNBLNK (a screen blanking
program). Having heard that SCRNBLNK sometimes causes problems, I
immediately ejected it from my AUTOEXEC.BAT (leaving SK and CED). I
find it hard to believe that CED or SK could cause contention; they
don't with anthing else I run and I have not had this problem until
this week.
For completeness, I have an AT&T 6300 with a 20mb hard drive running
MS-DOS 2.11.
Any suggestions/ideas/fixes/prayers/etc?? Thanks in advance.
Mark H. Granoff
ARPA: mhg@mitre-bedford
DDD : (617) 271-8438
------------------------------
Date: 19 AUG 86 09:46-PST
From: BEN%YMIR.BITNET@WISCVM.WISC.EDU
To: Info-IBMPC@B.ISI.EDU
Subject: EGA Screen Dump Wanted
I'm calling the Graphics Development Toolkit from IBM Pascal V2.0 to
produce graphics on my EGA card. My problem has been creating a
printer dump of what I see on the screen. I've tried using the
printer drivers included with the GDT and had no luck. GRAPHICS.COM can
only work with a CGA. Am I going to be forced to muck around in
assembler, or is there a simple way to get the print out I would like?
Ben Staat
Math Dept., Harvey Mudd College
bitnet==> ben @ ymir
arpa====> engvax!ymir!ben@csvax.caltech.edu
non-binary communication==> (714) 621-8006
------------------------------
Date: 22 Aug 86 09:46 GMT
From: ghicks @ KOREA-EMH
Subject: Disk Problems
To: Info-IBMPC @ USC-ISIB
A friend of mine purchased a 20mbyte hard disk from LOGIC
ARRAY in Costa Mesa recently. We (he ave the
following problem:
The DOS FORMAT command reports "Track 0 not fo
unusable."
The JOS REFORMAT command (from the Info-IBMPC Lib) reports
"Incomp Boot Sector. Returning to DOS. FAT par-
tially or wholly unusable." (Or something to that effect.)
The disk drive in question is an NEC-D5126 20 meg drive with
4 heads and 2 disks. The contrroll is a DTC 5150CX
controller.
No Software was provided. Other than the following
instructions, no documentation was provided:
"to physically format the hard disk e following
DEBUG:
g=C800:5 (We type)
Drive=1 (It says "Drive=", We say "1")
Set interleave factor to 3 (Contained in the structions.)
FDISK
Select the DOS partition
FORMAT c:/s/V
Can't continue for the reason cited above.
We formahe disk, ran FDISK (this showed 611 cylinders
not the expected 612...), SYS'd the disk for the system, and
then booted from the hard disk.
Our questions are:
1) do we (he, not I. I have a 'good' hard disk) have a
problem?
2) Should walone and hope for the
best? Or should we try and fix the problem?
3) What does "track 0 not found. Disk unusable." mean?
4) If track 0 is unusable, why can we boot from the hard
disk? How can the system appear to function with a mi?
5) Is there perhaps something we can do to fix the
problem? What software should we use? Where can we get it?
Your assistance in this matter will be appreciated.
Gregory Hicks
------------------------------
Date: 1986 Aug 19 18:43 EST
From: Bob Babcock <PEPMNT%HARVARDA.BITNET@WISCVM.WISC.EDU>
To: <Info-IBMPC@B.ISI.EDU>
Subject: Problem with Hercules in a Compaq
We recently tried to replace the video board in a Compaq portable with
a Hercules monochrome graphics card (and external monitor). Unfortunately,
the Compaq disk controller board and the Hercules boards both have
parallel printer ports LPT1, and neither works with both boards in the
system. Anyone out there know how to
1. Modify either the Hercules or Compaq boards to turn the port
into LPT2 (Hercules will tell us how to disable the port).
or 2. Have any recommendations for an alternate monochrome graphics
board with either a serial or parallel port which can be strapped
to be LPT2 or COM2?
Thanks
------------------------------
Date: Thu 21 Aug 86 10:46:30-EDT
From: Joseph M. Newcomer <Joe.Newcomer@c.cs.cmu.edu>
Subject: MSKERMIT query
To: info-ibmpc@b.isi.edu
I can't seem to find a way to turn off the automatic-renaming of
files when shipped via Kermit. I seem to be able to turn off the
warning that they have been renamed, but not to turn off the renaming
itself. I find this objectionable in the extreme. When I send a
file to another PC or retrieve one, I expect to find it under the
name I specified, not one concocted by Kermit as a "favor" to obviously
stupid me who couldn't possibly have meant to supersede that file!
I lost several hours debugging because Kermit kept renaming instead of
replacing a file I was updating on one machine and sending to another.
Does anyone know how to accomplish this short of making source
changes and rebuilding Kermit? If not, does anyone have just a PC-DOS
version of the source; I can't decrypt the two tons of stuff that
comes from Columbia to find just the procedure for locating the PC-DOS
sources and rebuilding them.
[I tried to remail this to Info-Kermit@Columbia, where it belongs,
but I get back that Info-Kermit is not a valid address. Does anyone
know what became of the Info-Kermit mailing list? -rag]
------------------------------
Date: Fri 22 Aug 86 11:28:39-PDT
From: Ted Shapin <BEC.SHAPIN@USC-ECL.ARPA>
Subject: Why does IBM async hold up CD?
To: info-ibmpc@B.ISI.EDU
Using the IBM Async board and an external modem, when the system comes
up, the carrier detect line is held on *from* the async board.
Carrier detect is supposed to be a signal *from* the modem. Is this a
defect in the design of the async board and how can I make CD depend
upon the modem signal?
I do not see this problem with async ports on other boards like the
AST.
Ted.
------------------------------
End of Info-IBMPC Digest
************************
-------