perm filename IBMPC[MF,ALS] blob
sn#770533 filedate 1984-10-01 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00012 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00003 00002 ∂02-Sep-84 1939 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #90
C00059 00003 ∂07-Sep-84 1631 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #91
C00120 00004 ∂10-Sep-84 1106 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #92
C00142 00005 ∂12-Sep-84 1540 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #93
C00196 00006 ∂14-Sep-84 1342 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #94
C00267 00007 ∂14-Sep-84 1848 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest v3 #95
C00289 00008 ∂19-Sep-84 2010 INFO-IBMPC@USC-ISIB.ARPA INFO-IBMPC Digest V3 #96
C00333 00009 ∂20-Sep-84 0528 INFO-IBMPC@USC-ISIB.ARPA INFO-IBMPC Digest V3 #97
C00403 00010 ∂21-Sep-84 0209 INFO-IBMPC@USC-ISIB.ARPA INFO-IBMPC Digest V3 #98
C00437 00011 ∂28-Sep-84 0357 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #99
C00485 00012 ∂01-Oct-84 0136 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #100
C00527 ENDMK
C⊗;
∂02-Sep-84 1939 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #90
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 2 Sep 84 19:38:18 PDT
Date: 2 Sep 1984 18:45:00 PDT
Subject: Info-IBMPC Digest V3 #90
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Sunday, 2 September 1984 Volume 3 : Issue 90
This Week's Editor: Eliot Moore
Today's Topics:
Xenix on PC/AT
PC/IX and Xenix on AT (2 messages)
AT Floating Point Performance (2 messages)
A Secondhand AT review
AT&T 6300 vs. AT
AT Flight Simulator
Some ditties from the AT Tech Ref
AT disk performance
Nestar PLAN
Timestamping
DOS Calls from Interrupt Routine
dBase II/III
MUMPS
8087 option slot
Deleting non-DOS partitions
Floppy interchange with Morrow MD3
Floppy parity problem
Ramdisk driver needed
PC/IX C Compiler
LISP and C compilers
INDOS flag
Public Domain Assembler
Expansion box wait state problem
Graphics enhancements
MIT Terminal Emulator
Linker problems (2 messages)
----------------------------------------------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 29 Aug 84 23:29-PDT
Date: 29 Aug 84 0:41:33-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!intelca!kds @ Ucb-Vax.arpa
Subject: Xenix on PC/AT
Just coming from the Bay Area Uniops meeting where Unix on PC-types
was discussed, I was distressed to hear that Xenix for the PC/AT will
not run in protected mode on the 286. Now whatever you think about
the protection mechanisms on the 286, they are there, and judging
from comments made here and elsewhere, protection is very important
for any "valid" port of Unix. In addition, it is difficult (read
d**m near impossible) to get at memory >1Mbyte unless you are
in protected mode on the 286, so you have all that memory space that
is practically useless. On top of all that, after hearing the speakers
moan about the problems with swapping medium/large model programs
for the 86, I would think they would welcome the 286-style segment
descriptor thingys (which aren't really physical addresses). On top
of all that, Xenix is available for the 286 in protected mode
(like, I am typing on it, right now!, although I am not really
up on its availability, I believe Intel has been shipping it for
at least a year now...) Anyway, can anyone out there confirm or
dismiss this vile rumor, or at least explain what is going on here?
--
I've got one, two, three, four, five senses working overtime,
trying to take this all in!
Ken Shoemaker, Intel, Santa Clara, Ca.
{pur-ee,hplabs,and,scgvaxd,dual,idi,omsvax}!intelca!kds
---the above views are personal. They may not represent those of Intel.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Aug 84 0:12-PDT
Date: 29 Aug 84 6:25:22-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!rochester!ur-laser!tomk @ Ucb-Vax.arpa
Subject: PC/IX and Xenix on AT
IBM announced that they will support both xenix and pc/ix on the AT.
They might do this to encourage people to use the system best suited for
their needs. (i.e.) Xenix if you want to do development pc/ix for running
applications.(C compiler etc. optional). This also gives customers the
illusion that their un*x is suited for them. Also I believe only one
of them is available right away so maybe they just wanted to get the
product out the door.
--------------------------
Tom Kessler {allegra |seismo }!rochester!ur-laser!tomk
Laboratory for Laser Energetics Phone: (716)- 275 - 3786
250 East River Road 275 - 3194
Rochester, New York 14623
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 8:01-PDT
Date: 30 Aug 84 8:44:40-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!sdcrdcf!randvax!jim @ Ucb-Vax.arpa
Subject: PC/IX and Xenix on AT
A little clarification on PC/IX vs Xenix on the AT:
Yes, only one is available now on the AT - Xenix. Yes, PC/IX will be made
available on it. IBM said they expected it to happen 3Q84 or 4Q84.
However, PC/IX will be running in "8088" mode, while Xenix will be using
80286 instructions. (Again, according to the IBM guys - Jim Balter at
Interactive could set this straight if he's on... You on, Jim?)
I assume they'll do something with all that extra potential memory (??).
Note that DOS 3.0 is also an 8088-type system, since it has to run on the
whole line.
So PC/IX should be the UNIX of choice if you're planning applications across
the PC line, while you might prefer Xenix if you need the extra increment
of power.
Jim Gillogly jim@rand-unix {decvax, vortex}!randvax!jim
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 1 Sep 84 23:45-PDT
Date: 30 Aug 84 15:42:52-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!duke!phs!jmk @ Ucb-Vax.arpa
Subject: AT Floating Point Performance
I recently had a chance to try out some programs written in C
(CI-86) on an AT with the 80287 floating point chip. One program did
number crunching operations primarily and we tested versions with and
without the floating point chip. Both versions had been compiled
on an XT. The non-80287 version ran about 3x faster on the AT, compared
to the XT -- about the same speed improvement as other programs IBM
demonstrated. The 80287 version ran only about 10% faster on the
AT, compared with the XT! Inquiries to IBM brought the comment that
the internal speed of the 80287 is essentially the same as the 8087.
Pity... -- we had hoped the AT would be faster in number crunching
operations!
Mailen Kootsey
Duke University
phs!jmk
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 1 Sep 84 23:48-PDT
Date: 30 Aug 84 17:59:07-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!ecsvax!dgary @ Ucb-Vax.arpa
Subject: AT Floating Point Performance
<>
A surprising fact surfaced at an IBM PC AT demo the other day:
While programs generally run 3x as fast on the PC AT as on the
PC, programs making heavy use of the floating point chip (8087 on
the PC, 80287 on the PC AT) seem to run at the same speed. For
example, inverting a 30x30 matrix in IBM's APL took about 10
seconds on both machines. This was confirmed by tests of at least
two programs (the APL interpreter and an program written in (I think
it was) assembler. Very curious. My suspicion is that the 8087 and
the 80287 run at about the same speed, and that most of the
time being eaten up in these tests was floating point crunching.
I'm still surprised the faster and wider memory access didn't
speed things up!
D Gary Grady
Duke University Computation Center, Durham, NC 27706
(919) 684-4146
USENET: {decvax,ihnp4,akgua,etc.}!mcnc!ecsvax!dgary
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 1 Sep 84 23:30-PDT
Date: 31 Aug 84 21:56:05-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!watmath!jmsellens @ Ucb-Vax.arpa
Subject: Secondhand AT review
Well, my buddy in Toronto has had an AT for a few days. I had a chance to
fool around on it for a while. Here are my (our) impressions:
Fast! It screams! We tried a couple of video games on it - they ran
fast - pretty well too fast to play (NIght Vision Pinball for example).
Dir's go screaming by. Compiles seem to be about 5 (count 'em) times
faster.
Quieter - he used to use an XT with expansion box and two 10 meg IBM disks.
The AT whispers in comparison (in comparison to my lowly two floppy machine
too).
Keyboard - takes a tiny bit of getting used to. I kept hitting the ~` key
for escape (which is now at the upper left corner of the numeric pad).
Overall - nice, the big enter key does help a bit (but I've gotten used to my
little one).
Compatibility - seems to run anything.
Boots a lot faster - the machine waits for the disk to come up to speed
after it does the memory check (which goes a lot quicker).
Documentation - smaller paperback 'user guides' - more like "what to do
after you plug it in". More useful reference cards (booklets actually).
The DOS binder is a different colour (my, how pretty!). The BASIC 3.0
manual makes no reference to the "sys req" key - they probably forgot
to put the new extended code in the list at the back of the book (he
doesn't have the Tech Ref book yet).
Gripes: Still no "power-on" reset button (but that would be admitting
defeat, wouldn't it?). The high density floppy *seems* a little slow
(but we could be hallucinating).
Decision: When (if?) I get a job, I will refuse to work on anything
less than an AT. It has enough speed that you can actually edit and
compile at the same time, without going crazy waiting for it.
John.
------------------------------
Date: Fri, 31 Aug 84 15:06:41 est
From: sequoia!brownell@harvard.ARPA (Dave Brownell)
To: harvard!Info-IBMPC@USC-ISIB.ARPA
Subject: AT&T 6300 vs. AT
Does anybody have any information on the performance of IBM's new
beastie vs. AT&T's version?
I saw some timings that said that an 8086 was faster than the
80286 running at the same clock speed. I could believe that, but
I'd like some confirmation. Even if it the 80286 is marginally
faster, it seems like the 6 MHz (IBM) vs. 8 MHz. (AT&T) difference
could be quite significant. But I'd still like to see some nice
tasty facts.
Any takers??
Dave Brownell
sequoia!brownell@Harvard.ARPA
------------------------------
Date: Thu 30 Aug 84 20:44:13-PDT
From: Hellmut Golde <GOLDE@WASHINGTON.ARPA>
Subject: AT Flight Simulator
To: info-ibmpc@USC-ISIB.ARPA
InfoWorld magazine stated that the new PC AT does not run
Flight Simulator. Whether this is true or not and what in means in
regard to other programs I couldn't say.
--Peter Golde
-------
------------------------------
Date: Thursday, 30 Aug 1984 08:11:29-PDT
From: waters%viking.DEC@decwrl.ARPA (Lester Waters)
To: info-ibmpc@usc-isib.ARPA
Subject: Some ditties from the AT Tech Ref
The BIOS for the PC/AT appears to have been totally rewritten
(although functionality is maintained with the old BIOS). In some
spots, the code has gotten worse, in others, an improvement is shown.
Not all of the BIOS programmers took advantage of the 286's expanded
instruction set (such as ISHL <reg>,<count> = shift with immediate
count). This is just a quick summary of new features (which are geared
especially towards multitasking):
Through INT 15 (Cassette I/O) call:
* Device open, Device Close, and Program Termination
These apparently just exit and do nothing at the moment.
* Event wait - wait on some event. I can't really tell
what event, but it is tied with I/O ports 70h and 71h.
* Joystick support - read position and triggers
* System Request key - apparently just exits.
* Block move for virtual memory (above 1MB).
* Get extended memory (?)
* Swap to virtual (protected) mode
* Memory size determination
Through INT 1Ah (Time of Day) call:
* Set/Read clock (with BCD support and daylight savings)
Through INT ?? (periodic timer/event)
* >>>> This is documented as INT 50 in the BIOS manual <<<<
* Set up a scheduled event (to be called through INT 4Ah).
* Operates at 1024 times per second
* Can be polled
Through INT 13h (Diskette/fixed disk I/O):
* Additional options, especially for fixed disk support
Through INT 10h (video I/O) call:
* 3 'reserved' functions - they just exit.
* A Write String function
Here is a set of notes that I generated while perusing the PC/AT Tech.
Ref. Man:
Interrupts:
- IRQ8-15 are slaved to IRQ2 on master 8259 PIC
- IRQ9 is re-directed by the ROM BIOS to INT 0Ah (old IRQ2)
- IRQ8-15 are offset 70h (Interrupts 70 - 7F)
System shut-down (Programmable processor RESET):
- Store function code in NVM
- Reset processor
- Dispatch according to fcn code in NVM during power-up
(0 means normal power on/restart)
Hard disk:
- Uses REP INSW/OUTSW (186/286 instructions) to port 1F0h to
read/write sector buffer.
- WD1010 disk controller chip
Keyboard:
- System interface by 8042 on mother board
- Bi-directional (simple interface due to 8042). Apparently has
mode for supporting PC/XT keyboard.
- SYStem REQuest key (See INT 15h extionsions)
- Keyboard has identical electrical interface. New half-duplex
error checking protocol used to communicate between the keyboard
and mother board micros.
- [Editorial] 2nd keyboard processor has been added to system in
order to unravel all the mistakes in the initial PC implementation.
Witness the problem with trying to emulate the PC/XT keyboard on
the PCjr.
INT 15h (XBIOS!):
- SYStem REQuest make/break notification.
- Extended memory size determine (Above 640K)
- Joystick support; current sws; current resistive inputs
- Block move. Uses Protected mode and system shutdown function
to exit back to Real mode.
- Switch to protected mode.
- Real time delays using RTC (CMOS clock). If interval timer already
in use then this function fails.
- Real time interval timer interrupt (single). User supplies an
address of flag to set or hooks INT 4Ah.
NOTE: Both timer functions accept input in
microseconds. Clock runs at 1024Hz, therefore
resolutions is 976us.
- Multitasking hooks for process synchronization. Application
specified, only stubs are in ROM.
Video:
- New function to output string (ala "Write TTY")
Specify: String
Length
Cursor position
Video page
Fixed/Variable attribute
Optional reposition cursor
Timer:
- Read/Set date/time
- Set/Reset TOD alarm (INT 4Ah)
0) 80286 vs. 8088/86. There are some real differences here that I will
cover later if requested.
1) Max memory of 15Mb (less 128Kb ROM space). IBM says that the AT only
supports 3MB. This is due to the fact that they only have 512Kb option
cards and 3Mb fills 6-slots. Watch for 3rd party 1-2Mb options.
2) 16 levels of interrupt support. Accomplished by adding a 2nd 8259.
IRQ9 is redirected to IRQ2 to compensate for the fact that IRQ2 is
no longer on the bus, but IRQ9 is. (I don't know why!)
3) 4 8-bit DMA channels (chl 0 is no longer used for refresh).
3 16-bit (word boundary only) DMA channels (2nd 8237 cascaded)
4) Real Time Clock/Calendar/Alarm/NVM (Motorola 146818) Large Lithium
battery attached to case with Velcro and motherboard with connector
5) Fixed Disk (not even closely compatible except at the ROM BIOS
interface and even there they have changed it slightly):
No DMA! 512 byte sector buffer that is directly accessible
memory. ST412 Interface, 5Mb transfer rate, WD1010 controller
Max support: 16 heads, 1024 cyl.
3:1 interleave instead of 7:1
6) 14.31818 Mhz clock independent of processor clock. Used for driving
8253 Timer and Video sub-system.
7) High density 1.2Mb diskettes (made by YE Data). 360 RPM, 15 Sectors,
96 tpi, 2-sided, can read 48tpi by double stepping. Supports status
change line.
8) 84 key keyboard (extra SYS REQ key). Identical software interface, but
electrically incompatible. Bi-directional + LED support in ROM.
9) Keylock freezes keyboard and cover. You can still power it off.
10) 64Kb ROM:
- Move arbitrary 32Kb of memory to/from addresses above 1Mb in
Real mode.
- Enter Protected mode (cannot use BIOS any longer).
- Multi-tasking hooks. (more later)
- TOD support with alarm function
- Video write string function added
- Status line change from high density diskette
- Read/Set drive type for diskettes and fixed disks.
- New ROM BIOS fixed disk support.
11) Minor changes to the 8-bit bus. Hardware depending on a specific
system clock frequency will no longer work. The 14Mhz clock is
still provided but is not synchronous with the system clock.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Aug 84 7:02-PDT
Date: 29 Aug 84 17:46:25-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!ut-sally!cyb-eng!topher @ Ucb-Vax.arpa
Subject: AT disk performance
Specifically, what is the average access time? Mail your responses to me,
and I will post their geometric mean. (-:)
Cheers,
Topher Eliot
Cyb Systems, Austin, TX
{seismo, allegra, ihnp4}!ut-sally!cyb-eng!topher
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Aug 84 3:41-PDT
Date: 29 Aug 84 18:29:05-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!nsc!proper!dsmith @ Ucb-Vax.arpa
Subject: Nestar PLAN
[If this line is missing, call BUGBUSTERS at 1-800-654-1984]
Does anyone else out there in Netland use a Nestar PLAN series
Local Area Network? We at Galileo HS (not Proper) currently have three,
if you can call the old Cluster 1 (PLAN 1000) one of the series. I'd
like to find people who are also using the things and trade info on them.
Heaven knows that the only way to find anything out the way Nestar
supports us!
David Smith @ Proper Unix
USnail: Galileo Computer Center
1150 Francisco St.
San Francisco, CA 94109
(415) 474-0319 or (415)771-3150 x46
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Aug 84 0:28-PDT
Date: 27 Aug 84 6:10:29-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From:
hplabs!hao!seismo!harvard!wjh12!genrad!grkermit!masscomp!bonnie!clyde!watmath!wateng!broehl @ Ucb-Vax.arpa
Subject: Timestamping
Easy -- just use INT 21H, function code 57 (get/set a file's date and
time). It's documented on page D-51 of the DOS 2.00 manual.
--
-Bernie Roehl (University of Waterloo)
...decvax!watmath!wateng!broehl
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 31 Aug 84 2:02-PDT
Date: 25 Aug 84 8:57:00-PDT (Sat)
To: info-ibmpc @ Usc-Isib.arpa
From: pur-ee!uiucdcs!seaburg @ Ucb-Vax.arpa
Subject: DOS Calls from Interrupt Routine
The DOS calling within DOS screws up because of stack overflows. If I
remember correctly, you can do disk I/O type stuff while in DOS
character I/O space and vice-versa. You can't do char I/O while in
DOS char I/O and same for disk I/O.
Hope this helps.
Gunnar Seaburg
(supplemental IBMer)
------------------------------
Date: Fri 31 Aug 84 14:29:30-PDT
From: ALFIERI@ECLD.#ECLnet
Subject: dBase II/III
To: Info-IBMPC@USC-ISIB.ARPA
Here are a few random comments about dBASE III:
In general, I think it is an excellent reworking of
dBASE II. I say "reworking," because for all intents
and purposes if you already know dBASE II, you will
have no problems with III. Virtually all the
"interactive" (i.e., dot prompt) commands are the
same. There are some changes (not difficult to
master) in the programming area, such as the ! function
has now been converted to "upper", and there is now
even a "lower" for lowercase conversion.
I am very happy with the new functions for data and
time, which makes life a lot easier for programming.
The tutorial part of the manual is good, but it has
some noticeable omissions. If you know dBASE II and
walk through the tutorial, you'll quickly understand
what I mean. But I pity the novice who is first
starting out.
The reference section is not terribly great, but it
does give a basic run-down on all the commands and
functions.
Another great feature is the "memo" field, which
allows you have a free-form "comment" for each record.
There is also a new "date" field. Creating new data-
bases is much easier, since the program does not
allow you to enter field type, etc. incorrectly -- or,
rather, it at least prompts you once you've entered
a field type about the rest of the info for the field.
You can now modify the structure without going through
the rigamarole once necessary.
What I do not at all like, however, is "Prolock"! This
is an inconvenience beyond belief. You must have the
program diskette in the A drive for the program to work.
So much for hard-disk users. But that is the way
Lotus works, and it doesn't seem to have hurt its
sales much. I still wish one were allowed to copy the
program at least once or twice.
Please don't hesitate to call on me for more specific
answers.
--vince alfieri
computing information services
university of southern california
alfieri@usc-eclb
(213) 743-5474
------------------------------
Return-Path: <Gaffney@LLL-MFE.ARPA>
Date: Sat, 1 Sep 84 08:15 PDT
From: Gaffney@LLL-MFE.ARPA
Subject: MUMPS
To: INFO-IBMPC@USC-ISIB.ARPA
Professor Richard F. Walters from the University of California, Davis,
supplies a mumps interpreter for about $50 and a mumps file manager
for another $50.
Our experience indicates that both work pretty well, the documentation is
adequate but it is essential to have a mumps background before reading
it, and professor Walters is very responsive to bug reports.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 1 Sep 84 23:47-PDT
Date: 30 Aug 84 16:11:02-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!ecsvax!dgary @ Ucb-Vax.arpa
Subject: 8087 option slot
<>
For the record, EVERY version of the PC, including the original ones, has
an 8087 slot. Where did this rumor that the 8087 slot was a recent
addition start??
-Gary
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 9:00-PDT
Date: 30 Aug 84 9:34:25-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: ihnp4!mgnetp!burl!hou3c!ka @ Ucb-Vax.arpa
Subject: Deleting non-DOS partitions
I have an IBM XT with one DOS partition and three non-DOS par-
titions. I want to dedicate the entire fixed disk to one DOS
partition. Unfortunately, the fdisk program does not appear to
be able to delete non-DOS partitions. Does anyone know what I
should do?
Kenneth Almquist
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 8:42-PDT
Date: 30 Aug 84 18:12:13-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: ihnp4!houxm!hogpc!pegasus!mel @ Ucb-Vax.arpa
Subject: Floppy interchange with Morrow MD3
Help please. I use WordStar, SuperCalc 2, and dBase II on both a Morrow
MD3 and an AT&T 6300 (IBM PC alike), and would like to be able to interchange
data files between the two. The Morrow can write IBM CPM-86 disks (and
Osborne I and Xerox 820), but not PC-DOS/MS-DOS disks. I called several
of the advertisers of CP/M to PC-DOS disk copy programs, but none claim to
be able to go MD3 to IBM or visa-versa. Does anyone have first hand
experience with a method to do this? Please reply to me, I'll summarize
to these newsgroups. Thanks. Mel Haas ihnp4!houxe!mel
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 2:15-PDT
Date: 31 Aug 84 10:42:06-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: ihnp4!inuxc!fred @ Ucb-Vax.arpa
Subject: Floppy parity problem
My IBM-PC is having the following problem. After working
merrily away for an hour or so, the time comes to save the work
on a disk. The save command is given and the drive A ( this is
a one drive system) LED comes on and the computer goes out to
a PARITY CHECK 1 error, thus locking up the world and losing
a great deal of work.
Note the error is intermittent, about 50% of the time,
and it occurs on three different software packages.
I've run diagnostics over night and find no errors.
I've taken the computer to the local computer store selling
and servicing IBM equipment, they have cleaned and realigned the
head, they also have found no problem after 24 hours of testing
on the "ADVANCED" diagnostics.
Yet the problem remains, asking the machine to write to
the disk is like flipping coins, half the time I win.
Any ideas on where the problem is?
Fred Mendenhall
AT&T CP Indianapolis
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 1 Sep 84 23:55-PDT
Date: 29 Aug 84 12:58:59-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From:
decvax!genrad!grkermit!masscomp!bonnie!clyde!watmath!utzoo!utcsrgv!utai!indra @ Ucb-Vax.arpa
Subject: Ramdisk driver needed
Has any one ever written a ramdisk.dev for use in MS/PCDOS 2.1x ?
This is supposed to be called in config.sys .
=============================================================================
..!utzoo!utcsrgv!utai!indra
Indra Laksono @ U of T, Ont
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Aug 84 6:01-PDT
Date: 29 Aug 84 17:23:09-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!sdcrdcf!trwrb!trwspp!brahms @ Ucb-Vax.arpa
Subject: PC/IX C Compiler
[}{]
My supervisor came to me with a problem. When compiling a c program under
PC/IX, the compiler gave him the error "To many local variables declared"
or something similar. So, my question is, is there a way around this?
Is there a compiler switch that can expand the local symbol table?
I know of such a beast using Aztec C, but I have never used c on
an IBM-PC under PC/IX.
Please send all replies to me. Thank You.
-- Brad Brahms
usenet: {decvax,ucbvax}!trwrb!trwspp!brahms
arpa: Brahms@USC-ECLC
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 29 Aug 84 22:44-PDT
Date: 28 Aug 84 7:01:26-PDT (Tue)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!linus!philabs!mcvax!guido @ Ucb-Vax.arpa
Subject: LISP and C compilers
>Lattice is more expensive and I understand that it is less
>compatible with UNIX.
Lattice may be (somewhat) more expensive, but it is certainly not less
compatible with UNIX (the library was, but the new version (2.10) seems
to solve most problems).
In other respects, it is BETTER than CI86: supports all combinations
of small/large code/data space (so you can have 300 kb of code but 64 kb
of data, which allows 16-bit pointers, which in turn makes smaller and
faster code), better treatment of structure items (different struct's
may have items of the same name in different positions).
I find Lattice's documentation superior to CI86's.
Admittedly, CI86 includes source of the library.
Concluding, for really big programs, I would strongly recommend Lattice.
--
Guido van Rossum, "Stamp Out BASIC" Committee, CWI, Amsterdam
guido @ mcvax
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Aug 84 3:55-PDT
Date: 28 Aug 84 14:12:17-PDT (Tue)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!tektronix!uw-beaver!microsoft!gordonl @ Ucb-Vax.arpa
Subject: INDOS flag
I've seen several articles recently on the location and use of the
"INDOS" flag to allow system calls from within interrupt service routines.
Naturally, you own your machines and your software and can do whatever
you please, but those writting software for distribution and/or support
should be aware of (what I hope is) the obvious:
1) this is not supported under any version of the DOS
2) it usually works under DOS 2, but there may be circumstances
when it doesn't (general disclaimer, don't know of a specific
circumstance)
3) it will usually not work under DOS 3 and DOS 3.1; the DOS is
considerably restructured and this flag takes on additional
meanings and uses
4) it will fail catastrophically under DOS 4.0 and forward.
This is just a specific example of the generic principle for all
software products: "if it isn't documented then it isn't supported and
will probably go away or change in future releases."
Gordon Letwin
Microsoft
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 29 Aug 84 2:48-PDT
Date: 27 Aug 84 14:14:35-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!tektronix!uw-beaver!ssc-vax!ssc-bee!bmac2 @ Ucb-Vax.arpa
Subject: Public Domain Assembler
I am looking for a public domain MS-DOS assembler
if such a thing exists. Please send responses to
me at
Roy Sherrill bmac2@ssc-vaxb
Thanks
------------------------------
Date: Tue 28 Aug 84 16:29:53-MDT
From: Carl Diegert <DIEGERT@SANDIA.ARPA>
Subject: expansion box wait state problem
To: info-ibmpc@USC-ISIB.ARPA
Two PC's here, both with fixed disks in official IBM expansion boxes have
problems with writes to the fixed disk under certain application programs.
The problem is that, after correctly writing the first couple of sectors, the
remaining sectors are filled with ones (FF's).
Has anyone else had this problem? (Please respond to DIEGERT@SANDIA or call at
505-844-0977.)
The problem may be caused by the wait states that are inserted by the IBM
extender card, used to interconnect the PC and the Expansion box. This card
has a 4 switch pack which "should be set to indicate the maximum contiguous
read/write memory housed in the system unit" (Tech Ref page 1-85). Appendix
G states that these switches should be set to hex "A" (off-on-off-on) with
640KB memory in the PC. With the exception of accesses to ROM at F0000 to
FFFFF, I expect that this causes wait states in all reads and writes to
locations above RAM, including the fixed disk buffer (including DMA access?)
at C8000. Anyone have a logic analyzer to clip to the 8088 and look?
Does anyone understand the advanced diagnostic error I get with the PC
configured as described above:
Error in Expansion Option - 1813 S
Data 0213 = AE / AC SW=A
It looks to me like the diagnostic was trying r/w's to segment A, yet the
switches are set to A indicating, according to Tech Ref Appendix G that
memory in the PC only extends through segment 9 (640KB).
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 28 Aug 84 1:15-PDT
Date: 27 Aug 84 6:54:46-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!tektronix!stever @ Ucb-Vax.arpa
Subject: Graphics enhancements
From hints in various PC columns over the last several months it
appears that IBM will offer several levels of graphics beyond what
they offer now up to and including 1024 x 1024 with the earliest
announcement in Sept. I would love to see this, however I am not
overly optimistic given their performance with the PCjr and
3270-PC graphics *improvements*!
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 29 Aug 84 21:56-PDT
Date: 28 Aug 84 13:53:09-PDT (Tue)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!uwvax!hagens @ Ucb-Vax.arpa
Subject: MIT's Terminal Emulator
I am using mit's terminal emulator. When it writes to screen memory,
I get colored static on my screen (I am using a IBM color display).
Specifically, in vi when typing a line, I get static as each key is
pressed.
Has anyone seen this problem, or know of special tricks to use
on a color monitor, when writing directly to screen memory?
Thanks,
Rob Hagens
--
Rob Hagens @ wisconsin
...!{allegra,heurikon,ihnp4,seismo,sfwin,ucbvax,uwm-evax}!uwvax!hagens
hagens@wisc-rsch.arpa
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 11:57-PDT
Date: 30 Aug 84 16:38:21-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!tektronix!uw-beaver!microsoft!markz @ Ucb-Vax.arpa
Subject: MIT Terminal Emulator
When it writes to screen memory, I get colored static on my screen (I am
using a IBM color display).
The IBM PC color card does not have dual-ported memory. As a result, whenever
you are poking the memory at the same time the display controller is reading
it, it will read garbage due to the extra stuff on the bus.
To work around this, examine the ROM code in your IBM Tech Ref manual to
see how it waits for vertical/horizontal retrace before poking with things.
You will have 12 Microseconds for a horizontal retrace and 12 Milliseconds
for a vertical retrace. If you need more time than these, you may elect
to shut off the video beam altogether (but you'll get an annoying flicker).
Remember, in you instruction timing calculations, to add the 1 wait state
for every memory fetch on the PC.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 7:27-PDT
Date: 31 Aug 84 7:07:29-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!mo @ Ucb-Vax.arpa
Subject: Linker problems
We are desperately searching for a linker better than the stock PC-DOS
program. We would like a linker which supports groups of arbitrary
size so our Large-code-model object files can be linked without
undue screwing around, and we would also like a linker which honors
case distinctions. In these days of ASCII-96, linkers supporting
monocase names is pretty poor.
Any pointers deeply appreciated.
-Mike
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 10:11-PDT
Date: 31 Aug 84 12:14:39-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!mo @ Ucb-Vax.arpa
Subject: Linker problems
Let me correct an impression which may have been left by my previous
note on linker problems. Microsoft is not to blame for our problems;
they are implementing Intel standard designs in a conscientious way,
which is probably correct from many viewpoints. However, it isn't
adequate for what I need to do. The fact is that the Intel memory
layout models are rather at odds with the way Unix programs work.
(That is, text, data, bss, and stack, all only limited by the available
memory on the machine.) I tried to give up tinkering with explicit
memory layout when I graduated from assembler!
That, coupled with the incomplete support of the Large addressing
model by a compiler which claims to support it is the base of my
troubles.
I did not wish to impune Microsoft, but I do wish someone creating
such standards would realize there is an upper half to the ASCII table,
and that ASR33's are largely museum pieces.
-Mike O'Dell
------------------------------
End of Info-IBMPC Digest
************************
-------
∂07-Sep-84 1631 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #91
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 7 Sep 84 16:29:44 PDT
Date: 7 Sep 1984 15:42:48 PDT
Subject: Info-IBMPC Digest V3 #91
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Friday, 7 September 1984 Volume 3 : Issue 91
This Week's Editor: Richard Nelson
Today's Topics:
SIMTERM
Object-Oriented LISP: XLISP
uuencode/uudecode
Professional Pascal (2 msgs)
Word Perfect on Z-150
In-Depth MS DOS, CP/M Books
INDOS Flag and Undocumented Features
Some Answers to PC/IX Problems
50Hz frame, IBM mono. monitor
IBM MS-DOS 3.0 Linker
MS-DOS Linker
AT Compatibility
AT Flight Simulator
Flight Simulator Update
AT&T 6300 & hungover IBM cards
AT&T's PC Query
TeleVideo Block Mode Emulation Query
Modem7 for IBM-PC Query
DOS Call Hang Problem
----------------------------------------------------------------------
Date: 4 Sep 1984 12:48:28 PDT
Subject: SIMTERM
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
Ted Shapin has sent us Jim Holtman's public domain HP Terminal emulator.
We have combined some 40 source files into one large file SIMTERM.SRC.
This is a very extensive HP2621 terminal emulator. It has scrolling, ventel
modem support, script files, modem protocol, and menu driven help facilities.
It is written in Pascal, with portions in C and assembly language. It includes
yet another async communications package. Even if one isn't interested in using
this as a terminal emulator there is a lot of code here worth borrowing for
other applications.
I am sorry we had to combine this to one large file, but we just couldn't
maintain another forty files for one application. The file SIMTERM.EXE
can be downloaded via Kermit but the documentation files are included in
SIMTERM.SRC so I recommend FTPing both files.
------------------------------
Date: 7 Sep 1984 09:42:51 PDT
Subject: Object-Oriented LISP: XLISP
From: Richard Nelson <SWG.NELSON@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
Jay Weber has sent us David Betz' public domain XLISP, "an experimental
object oriented language" which combines features of LISP with an object
oriented extension.
Written in C, XLISP is in several chunks: XLISP.DOC, XLISP.H, XLISP1.C,
XLISP2.C, and XLISP3.C
------------------------------
Date: 7 Sep 1984 10:14:57 PDT
Subject: uuencode/uudecode
From: Richard Nelson <SWG.NELSON@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
We now have the unix uuencode/uudecode in files UUENCODE.C and UUDECODE.C:
from: ihnp4!cbosgd!mark @ ucb-vax.arpa
subject: re: wanted: uudecode
Here are the sources to uuencode and uudecode. They come with 4BSD
and are indeed in the public domain. (I wrote them while at Berkeley.)
------------------------------
Date: 7 Sep 1984 10:19:52 PDT
Subject: Serial Port Device Driver: IO5150
From: Richard Nelson <SWG.NELSON@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
Eric Roskos has sent us his public domain device driver for the IBM PC
(5150) serial port (AUX). Written in TURBO Pascal, the program shows how
to get the address or a procedure in order to set up interrupt vectors in
TURBO. The program is in IO5150.PAS.
------------------------------
Date: 4 Sep 1984 2206-PDT
From: KOTLER@USC-ECLB.ARPA
Subject: Professional Pascal
To: Info-IBMpc@USC-ISIB
cc: Kotler@USC-ECLB
I have been a Beta Test Site for a new Pascal Compiler for about 9
months now and recently the product was announced for sale. I have been
so happy with this product that I thought others might be interested in
hearing about it.
The product is called Professional Pascal* and the word professional is
truly the correct adjective for the compiler.
First of all, the compiler has no silly restrictions for the maximum
size of sets, the maximum number of enumeration literals, the maximum
number of fields in a record, etc., etc. The restrictions in general
correspond to very reasonable 8086 limitations, i.e. there can be no
more than 65535 elements in a set,etc. When using other compilers on
the PC and even on mini or mainframes, I generally run into all kinds of
silly limitations imposed by the compiler and not by the language. With
Professional Pascal this has not been a problem at all.
The code quality produced by the compiler is very good. Optimizations
performed include common subexpression elimination, constant folding,
jump optimization, case statement optimization and strength
reduction(i.e. substituting shifts for more expensive multiplies and
divides when possible, etc.). In addition 5 different memory models are
supported which include all the usual combinations of small code, large
code/small data, large code/medium data, large code/large data etc.
The compiler has full 8087 support and comes with an alternate
library(one for each of the memory models) which emulates the necessary
8087 instructions. I must confess that I haven't used this feature of
the compiler at all since I'm not a numerical programming type.
As far as the actual language, the compiler may be run in ANSI mode or
extended mode.
In ANSI mode(I have not actually used this mode), the compiler will only
accept programs which conform to ANSI Pascal rules.
In the extended mode, the Pascal Language is extended in many truly
amazing ways, making it considerably more powerful than ANSI Pascal.
The list of extensions is numerous(in fact there are probably 200 pages
devoted to the extensions and their associated rationale in the manual
that comes with the compiler). I'll try to give a reasonable partial
list so that the reader can get an idea:
1) Packages are provided with similar properties to Ada packages.
2) C operators are supported for bit manipulation.
3) Many facilities for the implementation of abstract data types are
provided. These include a form of abstract type itself (similar to Ada
private types), iterators(an extremely powerful and addictive feature
implemented along the lines of CLU iterators), abilities to open up
packages in the same way as records are opened via 'with' statements and
the ability to selectively import/export items from packages.
4) Built-in string data type.
5) Facilities for breaking the strong typing of Pascal when necessary.
In Addition numerous low level facilities for manipulating addresses and
doing address arithmetic are provided.
6) Many new useful intrinsics.
7) User definable operators are allowed. In addition there are
numerous syntactic nicities allowed. For example if I have a function
In←List which takes two arguments and returns true or false depending on
whether the first argument is contained in the list denoted by the
second argument, I can write(as is usual) 'if In←List(E, L) ...' or
alternatively I can write 'if E %In←List L then .. '. Similarly for
unary functions I can write expressions like 'j := X!Factorial' instead
of the usual 'j := Factorial(i)' if I like.
8) Numerous features which support separate compilation are provided.
Also some additional retrofit modes for converting code written with
other schemes is provided(I have not actually used this feature).
9) An ability to tell the compiler to use other than its usual calling
conventions for external procedures/functions is provided for
interfacing to other languages. This is a definable feature with such
options as the order the parameters are pushed, whether the callee pops
parameters or the caller pops parameters, whether parameter passing by
reference is required etc. The scheme is very flexible. (I have not
actually used this feature).
10) Nested declare blocks as in Ada, and a limited form of static
variable initialization is allowed.
The list goes on and on. In general there has been a tremendous amount
of thought that has gone into all the extensions. Also great care has
been applied to avoid adding many new keywords to the language, thus
providing most extensions using non keywords so as to reduce the
likelihood that legal ANSI programs would not compile in the extended
mode. In fact there are only six new keywords(declare, iterator,
otherwise, package, pragma, value).
The compiler has been almost completely bug-free. In the entire time
I've used it, I've never had the compiler generate bad code.In a few of
cases(less than 5) I did discover some disallowed construct at
compile-time which should have been allowed. Usually I am a very good
compiler breaker since I tend to stretch languages to their limits, but
in the case of Professional Pascal I have been unable to break the
compiler except in a few cases. I have about 20,000 lines of code in
Professional Pascal so I have given it a pretty good workout. The guys
that wrote the compiler have over 100,000 lines of code written in it
plus a lot of code has been written by other Beta sites.
In addition to the compiler a number of useful utility programs are
provided, including a cross reference generator capable of handling
multiple compilation units. Also a library of useful packages is
provided.
The documentation for the compiler is excellent and includes a users
manual and a language extensions manual. Additionally tutorials are
provided to demonstrate the use of many of the extended Pascal features.
I should also say that the compiler was written by Frank DeRemer, Tom
Penello, and David Pickens of Metaware* Inc. Those of you who are
familiar with compiler-compiler research will most likely have seen
articles by Deremer and Penello. The original definitions of LALR(1)
and SLR(1) are from Frank's Thesis from MIT in 1969. Since then Frank
and Tom have been noted contributors to the field of LR parsing and
error recovery and both have taught at the University of California at
Santa Cruz. David Pickens was one of the authors of the IBM Pascal/VS
which runs on OS/370 systems.
I have known Frank and Tom professionally for a number of years and have
always been quite impressed with their work and I have used software
they have developed in the past.
The compiler is currently available on the IBM-PC, Victor, HP-150 and
should run on any MS*-DOS 2.0 or greater machine.
Coming soon will be a Vax resident compiler and 8086 cross compiler.
Also, other micro, mini and mainframe compilers for the same language
will be available in the future.
The Compiler is being distributed by Microtec* Research. Microtec also
has a lengthy(10 page) description of the compiler which includes sample
code produced by the compiler which they will send you upon request.
The following is the ordering information:
Professional Pascal
Microtec Research Inc.
Phone: (408)733-2919
cost: $895
MetaWare is a trademark of MetaWare Inc.
Microtec is a trademark of Microtec Research Inc.
MS is a trademark of Microsoft Corp.
Professional Pascal is a trademark of Metaware Inc.
Sincerely,
Reed Kotler
General Transformation Corp.
------------------------------
Date: 5 Sep 1984 12:20:03 PDT
Subject: Professional Pascal
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: KOTLER@USC-ECLB.ARPA
cc: info-ibmpc@USC-ISIB.ARPA
Does this compiler use standard DOS (or compatible) linkers? Are
calling conventions consistent with any other languages?
------------------------------
Date: Mon, 27 Aug 84 10:42:28 edt
From: ANDERSEN <sigurd%vax1%udel-cc-relay.delaware@udel-relay.ARPA>
To: elefante@radc-multics.ARPA
Subject: Word Perfect on Z-150
Cc: info-ibmpc@usc-isib.ARPA
The "flash" version of Word Perfect works fine on the Zenith Z-150.
Satellite Software Int'l (SSI) no longer refers to it as the "flash"
version, though. One of the more recent distributions of the program
has an "IBM monochrome/color" version equivalent to the previous
"flash" version, and an "IBM color/compatibles" equivalent to the
non-"flash" version. The only thing I've found that doesn't work on
the Z-150 is Hercules' "HBASIC," because it assumes the existence of
a "BASICA.COM" and Zenith supplies a "BASICA.EXE," which HBASIC does
not find.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 10:00-PDT
Date: 31 Aug 84 10:53:04-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!cmcl2!lanl-a!rgt @ Ucb-Vax.arpa
Subject: In-Depth MS DOS, CP/M Books
<>
I sent out a request for in-depth books on CP/M and MS-DOS.
Here is an edited list of the responses. Thanks to all!
Richard Thomsen
[1]
From: cmcl2!seismo!ut-sally!utmbvax!mmr (Mike Rubenstein)
Subject: Re: Request for CP/M and MS-DOS in-depth books
Only book I've found that explains much is Inside CP/M by David E Cortesi
(Holt, Rinehart and Winston). It's primarily about CP/M-80, but the
file structure for CP/M-86 is the same.
You also should get hold of one of the public domain programs for
looking at the disk by sector. Best known is DU in both CP/M-80 and
CP/M-86 versions. Just about any RCPM (CP/M bulletin board) will have
the CP/M-80 version. Larger ones will have the CP/M-86 version also.
Good luck. CP/M's pretty simple, but the poor documentation makes it
pretty hard to figure out what's going on.
[2]
From: Bob Van Cleef
<cmcl2!seismo!decvax!sdcsvax!noscvax!revc@marlin.Nosc.ARPA>
Subject: Re: Request for CP/M and MS-DOS in-depth books
"Inside CP/M" - Jerry Coretese (sp). The best I've
seen for CP/M 80.
[3]
From: Eric Stork <cmcl2!seismo!hao!hplabs!ucbvax!STORK@MIT-MC.ARPA>
Subject: Request for CP/M and MS-DOS in-depth books
For CP/M 80 (2.2), you cannot beat Andy Johnson-Laird's
The Programmer's CP/M Handbook, pub by Osborne/McGraw Hill.
I have read just about every book I could find on CP/M
and that's the best by far.
Good luck,
Eric
[4]
From: cmcl2!ihnp4!alberta!jeff
Subject: Re: Request for CP/M and MS-DOS in-depth books
The Programmer's CP/M Handbook, by Andy Johnson-Laird (Osborne/Mcgraw-Hill)
was reviewed in this month's BYTE, and they say that it was quite good.
If you buy an uninstalled version of CP/M-80 or -86 you should also get lots
of info. As for MS-DOS, forget it. They don't believe in giving out info
about it, everything you will get is in the manuals they give you. If you
have almost no technical information, you could try buying the IBM DOS 2.1
Technical Reference Manual, but that's about it.
[5]
From: cmcl2!seismo!decvax!linus!utzoo!aesat!bmw
Subject: Re: Request for CP/M and MS-DOS in-depth books
Take a look at "The Programmer's CP/M Handbook" by Andy Johnson-Laird,
published by Osborne/McGraw-Hill. I read the review of it in the August
'84 Byte (pg 90) and purchased it last week. I am a fairly experienced CP/M
programmer (I have done a complete port of CP/M to my home-brew Z80 with hard
disk and non-standard
floppies), but this book has helped me a lot by revealing some useful
tricks/gotchas and other info which is missing or poorly documented in the DR
manuals. It also contains a LOT of source, either in ASM or in BDS/C, for a
complete BIOS as well as a bunch of enhanced utilities. Almost all meat and no
fat!
Bruce Walker ..decvax!utzoo!aesat!bmw
[6]
From: lbl-csam!noao!jacoby
Subject: MS-DOS books
I have also been looking for more in-depth books than "type dir
for a disk directory" sort of thing. Not much out there.
I have found one book that is intermediate and well written:
The IBM PC-DOS Handbook, Richard Allen King, Sybex publication, $17.
Please let me know if you find anything better.
George Jacoby
[7]
From: cmcl2!rna!rocky2!datagen!boykin
Subject: MS-DOS books
SIG/86 is an International MS-DOS Users Group dedicated
to the sharing of knowledge and software among its members.
Our group is composed of owners and users of any
manufacturers' hardware which supports MS-DOS. To our
members we offer a quarterly newsletter, public domain
software, discounts on commercially available software, a
bulletin board, and much much more.
[8]
Original-From: George W. Sherouse <sherouse@unc>
Subject: Re: Request for CP/M and MS-DOS in-depth books
DEC does have technical references for both OSs, though
their sales people don't seem to know it. I have ordered
both and have had considerable trouble getting them. The
latest word is that I will have them real soon. Check with
me in a while for a review.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 18:42-PDT
Date: 31 Aug 84 7:57:10-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!sdcrdcf!sdcsvax!akgua!gatech!gitpyr!roy @ Ucb-Vax.arpa
Subject: INDOS Flag and Undocumented Features
In response to the comment from Microsoft about the INDOS flag, and
the use of undocumented features, about all I can say is that if the
people writing operating systems would create and document all the
facilities needed by programmers, then there would be no need to use
undocumented features.
Perhaps the people at Microsoft could suggest some method for writing
an interrupt handler that uses DOS without using the INDOS pointer?
Or maybe we're supposed to go straight to the BIOS? :-)
And again, using the SWITCHAR=- configuration statement is undocumented
because lots of programs don't check it and/or produce incorrect results.
But since function 37H isn't documented, we can't even write "correct"
programs NOW!
It really seems that the only thing that can be done is to check the
DOS version number at the start of the program, and bomb out if it
isn't OK. But it sure would be nice if all the useful functions of
the operating system were documented.
What is really needed is an open architecture operating system to go
along with the open architecture IBM PC. I for one would love to be
able to buy a MS-DOS Tech Ref Manual, complete with listing on MS-DOS.
Roy J. Mongiovi. Office of Computing Services.
Georgia Institute of Technology, Atlanta GA 30332
Phone: (404) 894-6163, (404) 894-4660 [messages]
...!{akgua,allegra,and,hplabs,ihnp4,masscomp,ut-ngp}!gatech!gitpyr!roy
...!{rlgvax,sb1,uf-cgrl,unmvax,ut-sally}!gatech!gitpyr!roy
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 19:15-PDT
Date: 28 Aug 84 21:04:33-PDT (Tue)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!cca!ima!ism780!isc.net @ Ucb-Vax.arpa
Subject: Some Answers to PC/IX Problems
The following is a response from your good friends at Interactive to some
recent gripes/problems voiced on the net about PC/IX. The original notes
are prefixed with ">"; our comments are indented.
Note that the "next release of PC/IX" should be available quite soon.
>There definitely seem to be some port related problems with PC/IX. Not
>only the run-PC/IX-after-MSDOS problem Lee Merrill spoke of, but other
>somewhat mysterious conditions where tty0 or tty1 will get into a wedged
>state and refuse further commands. An attempt to shutdown at that point
>will cause shutdown to sit quietly for close to a minute, print the
>message "init: something won't die%s", and then MAYBE finish the shutdown.
This behavior is the result of two bugs, both of which have been
fixed for the next release of PC/IX. The first is a hardware bug
in many of the async cards used in PCs; I believe this has been
mentioned before on the net. We have altered the kernel to
circumvent the problem. The second is a bug in the kernel shutdown
procedure which we have fixed.
>One problem that is still plaguing me is a behavior problem with certain
>utility programs *only* when used from tty0 or tty1. They're fine from
>the local console. Seems that any extra characters entered after
>the final c/r of the password prompt and before the system types to you
>when using login, su or passwd will kill the port. You can still send
>to it and the machine will respond (you'd only know that if it were
>right next to you), but no output of any kind comes from the port. Once
>you have disconnected, the port is in its famous wedged state and nothing
>at all seems to clear it. Only solution at that point is a one minute
>shutdown (described above). IBM says its because PC/IX isn't running on
>an IBM! What help.
>
>For example: login: yourid<c/r> password: yourpass<c/r> <c/r>
>will kill it every time. Anything entered at the point of the second c/r
>after your password, but before you receive more output from PC/IX on
>any of the three utilities I mentioned will do it.
>
>I'm using the MicroLine BabyBlue II board for both ports. It works
>flawlessly in MS-DOS.
>
>Any suggestions? (please mail me directly, and I'll summarize if necessary)
>
>--Bill Blue {sdcsvax, sdchema, ihnp4}!bang!crash!bblue
We cannot duplicate this behavior. It is quite possible that it is
also due to the bad async card. Sorry, but that's the best we can
tell you!
>I have been "experimenting" with PC/IX since May 1, 1984. My problems
>are/were:
>
>1. I was only able to install PC/IX after IBM replaced EVERY board
>in my PC XT. restore -x didn't like my computer!
Your hardware was a lemon. Sorry!
>2. I could only get connect to work after spending several days
>with local "gurus". When I connect to a DEC VAX 11/780, I
>can't use vi, but with SIMTERM, etc. I can. Termcap problems?
The configuration files associated with connect (/etc/sites and
/usr/lib/INnet/connect.con) are really not that hard to figure
out, although I suspect somewhat better documentation would be
helpful.
There was a bug in the connect "talker" program which
prevented programs such as vi and e from operating properly
when connected. It is fixed for the next release of PC/IX.
>3. I had to rm /dev/lp and ln /dev/lp1 /dev/lp to get the printer
>to work.
We distribute the system with /dev/lp linked to /dev/lp0, which
seems the most logical choice. You must have put your printer card
in the second slot instead of the first, which obviously means
reconfiguring the devices (or /etc/qconfig) accordingly.
>4. I can't run more than one or two processed without crashing the
>system.
> mm filename > temp &
> ps -ef
>( the above will kill my system, ungracefully)
There was a highly unfortunate bug in the shell which went out
with PC/IX. It will manifest itself in relatively random ways
whenever the shell is running on a heavily loaded system. I
suspect that if just the two processes above killed your system,
that you must be configured with a very small amount of memory
(perhaps the minimal 256K?).
The bug is fixed for the next release of PC/IX. However, you
can (and probably should) circumvent the problem, at the cost
of some efficiency, by executing the following commands:
mv /bin/sh /bin/sh.old
cp /bin/sh.old /bin/sh
chmem +60000 /bin/sh
>5. I can't print on standard output the graphics character set while
>in connect mode ( e.g., echo "\0333" won't generate graphics )
Same bug as item 2., I believe.
>6. usr/games/bj doesn't work!
Shocking! Actually, this is pretty surprising, as the level of
testing that went into the PC/IX release should have caught this
easily. We will look into this bug.
>7. You can't type in the password after the login UNTIL the password
>prompt appears
This is standard UNIX System III. An ioctl with a TCSETAF command
is issued just prior to printing the password prompt, in order to
discourage people from typing in their password while echoing is
still enabled.
>8. INed ( e editor ) is the SLOWEST editor I have ever used. It can't
>keep up with my typing speed which is 10 words/minute on a good day.
Again, you must be configured with VERY little memory. On our
640K systems, e keeps up with anything we can give it; indeed,
it performs about as well as it does on a lightly-loaded VAX.
Also, keep in mind that while some of the function keys may seem
slow, they often perform complex tasks with a single keystroke,
so the illusion of sluggishness can sometimes manifest itself.
>The bottom line is that I am going to shelve the product and try
>Santa Cruze's XENIX.
No comment. Have fun.
>David S. Green Bell Labs mhuxi!dsg phone 201-564-4468
--isc.net (ima!ism780!isc.net) INTERACTIVE Systems Corp.
------------------------------
Date: 3 September 1984 12:39-EDT
From: Boyd D Blackwell <BDB @ MIT-MC>
To: INFO-PC @ MIT-MC
Subject: 50Hz frame, IBM mono. monitor
I put together a kludge for a friend who needed to drive a 50Hz
frame rate monitor with the IBM color graphics adapter. I was able to
increase the total number of character rows and adjust the vertical
sync position by creating a new video parameter table in ram, just as
"MODE ,r" does, but only for the 40 and 80 column alpha modes. The
6845 chip is limited to 127 character rows plus 31 scan lines, so in
graphics mode, can only generate displays of up to 127*2 + 31 line
(total lines), making the minimum frame rate about 55 Hz, not 50 Hz.
Fortunately, the monitor in question was just able to lock in to 55
Hz, but this left me wondering how the export versions of the color
adaptor work in countries where the power line frequency (and
therefore the TV frame rate) is 50Hz? I can imagine that the adaptor
could be modified to generate PAL signals instead of NTSC, but the
apparent limitation to >55Hz is built into the 6845 and IBM's choice
of a two page (odd/even line) graphics memory map.
The same problem would be encountered in trying to set up a video
parameter table to drive the IBM monochrome monitor from the graphics
adaptor. While the higher horizontal frequency could probably be
obtained by reducing the total (displayed plus nondisplayed)
horizontal characters, this would put the 50Hz frame rate even
further out of reach in graphics mode. If there is a trick (or
a special 6845??) to give the Europeans 50Hz, then maybe this would
allow a video parameter table to be set up for the IBM Monochrome
monitor??
------------------------------
Date: 3 Sep 84 19:26:16 PDT (Monday)
From: Kadifa.PA@XEROX.ARPA
Subject: IBM MS-DOS 3.0 Linker
To: info-ibmpc@USC-ISIB.ARPA
cc: goldberg%viking.DEC@DECWRL.ARPA, Kadifa.PA@XEROX.ARPA
In Info-IBMPC Digest V3 #84, Marshall R. Goldberg
(goldberg%viking.DEC@decwrl.ARPA (Marshall R. Goldberg LJ02/E4 DTN
282-2325)) mentioned a linker that Microsoft ships to its OEM customers,
called BIGLINK, which handles large numbers of publics. He used it for
linking together GW-BASIC and was very fast ("PLINK took over 45 minutes
to link a moderate sized GW-BASIC. BIGLINK does the same job in less
then 10 minutes !!")
I received from him a new message saying that the new LINK that comes
with IBM MS-DOS 3.0 will easily link GW-BASIC and seems to be a version
of BIGLINK. It is also available in local computer shops for $65.
This message is sent to Info-IBM Digest to disseminate this info on LINK
3.0
Abdo.
------------------------------
Date: Tuesday, 4 Sep 1984 06:40:40-PDT
From: waters%viking.DEC@decwrl.ARPA (Lester Waters)
To: info-ibmpc@usc-isib.ARPA,
amd!fortune!hpda!hplabs!hao!seismo!mo@decwrl.ARPA,
decvax!harpo!seismo!mo@decwrl.ARPA
Subject: MS-DOS Linker
There have been several messages posted inquiring about a
linker which could support more than 1024 PUBLIC symbol declarations
without bombing out. The LINK program which is being distributed
with PC-DOS 3.0 supports more than 1024 PUBLICs.
Finding a linker which distinguishes between upper and lower
cases will be hard to find. In many environments where a project is
a group effort, there would have to be MUCH more communications about
symbol names. Mistakes are easier to make. For example,
Programmer #1:
routine←name PROC FAR PARA PUBLIC
Programmer #2:
EXTRN Routine←name:FAR
Now note that the simple capitalization in Programmer #2's code
will cause an UNDECLARED EXTERNAL error. This can become very tedious.
- Lester -
------------------------------
Date: 3 Sep 1984 0205 PDT
From: Jeff Skaletsky <JEFF@JPL-VLSI.ARPA>
Subject: AT Compatibility
To: info-ibmpc@isib
Hold on a minute! About this compatibility question (which the media
has been strangely silent about):
Four of us trooped down to the local IBM Product Center (Pasadena) with
a mess of software under a few arms. The salespeople seemed as interested
as we did in the results: the main surprise was that Framework DOES NOT
RUN! After the first sign-on message, it dies! If any Ashton-Tate/Forefront
People are reading, any ideas on why and when you'll fix? I also understand
from Burbank Computerland that Microsoft Word does a similar belly-flop.
Yes, Info-World was right. Flight Simulator dies after half writing the
view screen after answering the questions (it actually stalls: low putt-putt
continues). We threw a few more things in, like Lotus, Symphony, Leading
Edge Word Processor, etc., and they all ran incredibly fast. I must comment
here that the Compaq DeskPro loaned us this week had NO problems with any
of the software. Now for a few quickie benchmarks of the processors:
doing 1000 cosines:
10 print time$
20 for i = 1 to 1000
30 x = cos(i)
40 next i
50 print time$
(Yes, I know, but it is an indication)
The AT ran it in 9 seconds, the XT in 27, and the DeskPro in 12.
One final note: I've ordered an AT (being the official Guinea Pig for
my organization) and in the process of so doing got some wrong information
from the Product Center staff (the ones who supposedly "knew"): they
told me to order the standard Asynchronous adapter for a second serial
port. The announcement clearly states that this will not work... I have
to order a second Serial/Parallel adapter, it seems. Don't trust any
one answer! These machines are so new that a lot of invalid assumptions
seem to be taking place. This also points up an interesting question:
If the standard Async adapter doesn't work, will a Smartmodem 1200B?
Anyone try this yet? The bus-bound hardware compatibility is something
we've heard very little about, beyond the "skirt" problem.
By the way, the full-up DeskPro is a really nice machine... my only fear
is the reliability; we've bought at least 30 Compaq Plus's and the infant
mortality rate seems to be about 75%!! Anyone else experience anything
like this? Repairs of these machines take forever, too- apparently a parts
problem. Problems include video display death, disappearance of the hard
disk, power supply (running boards way below the recommended load).
Thing is, I really like the machines!
Jeff Skaletsky
Jet Propulsion Lab
------------------------------
Date: Tuesday, 4 Sep 1984 06:31:29-PDT
From: waters%viking.DEC@decwrl.ARPA (Lester Waters)
To: info-ibmpc@usc-isib.ARPA, golde@washington.ARPA
Subject: AT Flight Simulator
No, Microsoft's Flight Simulator does NOT run on an IBM PC/AT
machine. This is because of the differences between the 8086 and the
80286. In the 8086 (or 8088), when a division by zero exception occurs,
a pointer points to the instruction AFTER the instruction which caused
the exception trap. It should point to the instruction which caused it,
not after. This was fixed in the 80286, and thus, confuses Flight
Simulator when this division by zero exception trap occurs.
I will try to find a suitable patch for this (after spending
many evenings with DEBUG).
- Lester Waters -
------------------------------
From: Jim Gillogly <jim@rand-unix>
Date: 04 Sep 84 19:55:39 PDT (Tue)
To: info-ibmpc@isib
Subject: Flight Simulator Update
If you're a legal owner of the original Flight Simulator for the IBM PC,
you can get an update from Microsoft as follows:
Call Customer Service, (206) 828-8088 to get a Return Authorization
Number, which is necessary to proceed further.
Send your original distribution disk, original documentation, and
$30.00 in an envelope with the RA # on the outside to:
Customer Service Dept.
Microsoft Corporation
13221 SE 26th St.
Bellevue, WA 98005
Don't bother to do this step without a valid RA #.
Receive UPS the new Flight Simulator, which includes an expanded
spiral bound manual and a disk for both the
PC and PCjr. It includes an overlay for the old-style PCjr
keyboard. Glad I didn't frisbee mine down the hill...
I sent my money on 16 Aug and received the update today, 4 Sep.
Jim Gillogly { My only connection with Microsoft is
jim@rand-unix { as a satisfied paying customer.
{vortex, decvax}!randvax!jim
[ Where we going? Planet 10! When we going? Real Soon!! ]
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 17:54-PDT
Date: 31 Aug 84 10:47:10-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!oliveb!olivej!bb @ Ucb-Vax.arpa
Subject: AT&T 6300 & hungover IBM cards
A couple of articles in this group recently said that IBM PC cards
which hang "below" the edge connector don't fit in an AT&T 6300, because
the AT&T's second edge connector gets in the way. Well, that's only partly
true:
As shipped from the factory, the AT&T machines have 7 expansion slots.
The second, 38-pin edge connector is installed for only 4 slots. (The
other three slots have a place on the circuit board for the second
connector, but there's no connector soldered in.) So, you can plug in
up to 3 IBM cards that extend below the edge connector, plus 4 that don't.
Or, of course, up to 4 16-bit cards.
At least, that's the way the European version of the AT&T machine is
sold (the Olivetti M24), and that's the way the early AT&T machines were
shipped. Anyone out there seen one different?
Bart Berger @ Olivetti ATC
{hplabs|fortune|ios|idi|ihnp4|tolerant|allegra|tymix}!oliveb!olivej!bb
------------------------------
Date: 4 Sep 1984 0945-PDT
From: HAL.DOVE at Ames-VMSB
Subject: AT&T's PC Query
To: info-ibmpc at isib
Reply-To: HAL.DOVE@Ames-VMSB
Seems many people thought I was talking about the PC/AT when actually
my questions were referring to the merits of AT&T PC. Does anyone
out there have one? Could you possibly give a mini review including
some semi-benchmark.
Mike
------------------------------
Date: Wed, 29 Aug 84 10:28 PDT
From: NEWMAN%SAV@LLL-MFE.ARPA
Subject: TeleVideo Block Mode Emulation Query
To: info-ibmpc@usc-isib.arpa
I have a need for a terminal emulator program for the IBM PC
that does emulation of any of the TeleVideo terminals in
block mode.
Does such a beast exist ?
Thanks,
Gerard Newman
Newman%SAV@LLL-MFE.Arpa
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 5 Sep 84 3:19-PDT
Date: 9 Aug 84 13:17:02-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: ucsfcgl!harrison @ Ucb-Vax.arpa
Subject: Modem7 for IBM-PC Query
There was an old version around that did not do any dialing or
have terminal emulation.
Is there any improved version?
Also, does anyone know of a program implementing the XMODEM protocol
on a PDP 11/70 running iAS operating system.
Please reply by mail.
harrison@ucsfcgl.UCSF
ucbvax!ucsfmis!harrison
ucsfmis!harrison@berkeley
One of those ought to make it to me.
Peter Harrison
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 2 Sep 84 16:26-PDT
Date: 31 Aug 84 17:26:53-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: ihnp4!houxm!hogpc!pegasus!hocsl!dmt @ Ucb-Vax.arpa
Subject: DOS Call Hang Problem
I need a little help with a problem I'm having with PC/MS-DOS.
(At least I think the problem is DOS.)
I've built a relatively simple program, in assembler, for
my IBM PC almost-alike. It ends, as I THINK it should,
with an INT 20H (but I get the same effect with DOS call 0).
The program executes right up to the INT 20 exactly as it should.
But then it hangs, and I have to reboot. Mysteriously enough,
when run under DEBUG, it works fine and gives the "program
terminated normally" message.
To try to localize the problem, I created a program that does
NOTHING but call INT 20. Once again, it ran fine under DEBUG,
but hung when called from the command interpreter as a .BIN
file (I get the same problem with a .EXE).
What am I doing wrong?
Thanks in advance for any suggestions.
Dave Tutelman
------------------------------
End of Info-IBMPC Digest
************************
-------
∂10-Sep-84 1106 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #92
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 10 Sep 84 10:59:07 PDT
Date: 10 Sep 1984 10:32:17 PDT
Subject: Info-IBMPC Digest V3 #92
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Monday, 10 September 1984 Volume 3 : Issue 92
This Week's Editor: Dwight Baker
Today's Topics:
Professional Pascal
AT XENIX runs in protected mode
.OBJ/.COM Files (2 msgs)
RAM disk driver
Used PCs available
AT arithmetic problem (FORTRAN).
Z80 add-on boards Query
----------------------------------------------------------------------
Date: 5 Sep 1984 1542-PDT
From: KOTLER@USC-ECLB.ARPA
Subject: Professional Pascal
To: BRACKENRIDGE@USC-ISIB
cc: KOTLER@USC-ECLB
In response to your message sent 5 Sep 1984 12:20:03 PDT
Professional Pascal can use normal Dos LINK, PLIN86 and I assume
it would be compatible with other such DOS linkers/library managers etc.
The calling conventions that the compiler uses by default are most like
C(i.e. the parameters are pushed in reverse order) though they
differ with respect to who pops the parameters after the call. However,
since the compiler allows the user to override the default calling convention on either a per procedure basis or per compilation basis, the
compiler can be made to be consistent with other
compilers.
The way that one overrides the default calling conventions is via a
pragma called 'calling←convention'. This pragma takes a number of
predefined attributes(which are specified in Pascal set notation).
The attributes are:
By←ref, Callee←pops←stack, Propagate(for languages which allow
procedural parameters), Save←Regs, Stack←Result, Reverse←Parms.
There is a package provided called other languages which one uses, as in the following example from
the manual to define some alternate calling conventions:
with Other←languages;
const PLM = [Callee←pops←stack, Propagate]
const C = [Reverse←parms, Propagate];
const Pascal = C + [Pop←Stack];
pragma Calling←Convention(PLM);
procedure Q(I: Integer);
....
pragma Calling←Convention(); -- restore the default
All the procedures/functions declared between the two pragmas would then
be called using PLM conventions.
There is also a file which comes with the compiler(and also a lot more
documentation on this in the manual) which contains the definitions for
some of the more popular calling conventions.
In addition, the user can override compiler defaults for the external
name that is associated with a particular procedure/function or
variable(although variables by default are not given their own external
name the compiler will if requested).
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 5 Sep 84 20:04-PDT
Date: 3 Sep 84 15:23:54-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: microsoft!gordonl @ Uw-Beaver
Subject: AT XENIX runs in protected mode
Contrary to Ken Shoemaker's note, XENIX on the AT does run in
the protected 286 mode. This provides a several megabyte address
space as well as memory management and protection, key items in
a multitasking environment.
SCO offers a XENIX which runs on an unprotected XT; I presume this
can be made to run on an AT, though I don't know if its been done
or why anyone would want to do it. Perhaps this was the start of
the mis-information.
Gordon Letwin
Microsoft
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 6 Sep 84 4:13-PDT
Date: 13 Aug 84 15:28:33-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: microsoft!markz @ Uw-Beaver
Subject: .OBJ/.COM Files
"... I am not including this statement because with it exe2bin would not
convert the file..."
EXE2BIN will convert an EXE file to a COM-format file under the following
circumstances:
o Entry point is 0:100h
o No Stack segment is present
o Image size is < 65536-100h-2
o There are NO relocatable long references.
I strongly suspect that there is something else at work here.
"... I suspect that it is something to do with having data segments
separate from the code segment..."
You have probably just hit the nail on the head. If you are using the macro
assembler from IBM, you'll find that it will emit segment definitions in
alphabetic order rather than in declaration order. This *may* be causing
your data segment to precede your code segment which might violate the first
condition above.
I really question your need of COM files in the first place. The *only*
place that they might be useful is in device drivers. Outside of these,
MSDOS is quite capable of performing all relocation for you. I strongly
urge anyone who is thinking of using COM files to stick with the EXE format.
Reasons:
o No need for EXE2BIN.
o Full use of instruction set (you can't effectively use the long call
instruction, for example) by being relocatable.
o Has extensible header should any changes to the format be needed.
o EXE files don't have 64K-100h-2 size restriction.
o Full control of stack placement. COM files have no guarantee on minimal
size of stack. I have seen a great numbers of COM files that fail on
small-memory machines or on reading large data over where their stack
was.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 6 Sep 84 6:43-PDT
Date: 26 Aug 84 21:16:00-PDT (Sun)
To: info-ibmpc @ Usc-Isib.arpa
From: pur-ee!uiucdcs!bradley!ali @ Ucb-Vax.arpa
Subject: .OBJ/.COM Files
I'm not sure how familiar you are with the structure of COM files but
basicaly a COM file is a memory image of an executable code with or without
data fields. The loader portion of the DOS performs the following before
giving control to the module:
[] a Program Segment Prefix is built and the appropriate parts of that
prefix is initialized(such as FCB1, FCB2, the INT 20H instruction at
offset 0, this is covered in detail starting on page E-3 in the DOS
manual)
[] All the general purpose registers are set to zero, all the segment
registers are set to the segment address of the PSP(Program Segment
Prefix)
[] The Instruction pointer(IP register) is set to 100H.
[] The COM file is read in at location 100H(this is a straight load
without any modification, editing or relocation)
[] a word of zero is pushed on the stack(since SP is zero from step 2,
the zero word is actually now at the very end of the segment. The
zero word serves as a return address to location zero which contains
the instruction INT 20H)
[] finally control is transferred through a long jump to PSP:100H
(this is how CS and IP are set)
It is very common to make the first three bytes in a COM file a jump
to the program start address(this is done by the programmer and not by the
loader or any other part of the system, usually to make it possible to
put data items before they are referenced in the code, this way the assembler
can generate the correct code knowing the size of the data items)
Creating COM files is restricted by the following:
1) The program may not contain any DATA segments ( you can change the DS
later in your code to point to anything you want, but the header of
EXE file produced by the linker must not contain a DATA segment)
2) there must not be a stack segment (this is trivial since a stack will
be setup by the loader, and there is really no need for the program
to have a separate stack unless the program is very large, in which
case you can relocate the stack during the initialization part of your
program)
3) The entry point must be at 100H. If you did not specify an entry point
using
END entry-point
statement in your program, in which case the linker assumes entry point
at CS:0 and sets the entry point address in the EXE file header to that,
so exe2bin will perform a simple conversion to produce a COM file. If on
the other hand you did specify an entry point then you better use the
ORG 100H
statement to move the program up, since exe2bin will delete the first
100H locations while converting this EXE file to a COM one and it
better not be part of your program.
In any case try something close to the following format:
CODE SEGMENT
ASSUME CS:CODE
ORG 100H
BEGIN PROC NEAR
JMP INIT
BEGIN ENDP
MSG DB 'Hello world.',0DH,0AH,24H
INIT PROC NEAR
MOV DX,OFFSET MSG
MOV AH,9
INT 21H
RET
INIT ENDP
CODE ENDS
END BEGIN
of course you can omit the PROC ENDP stuff and use labels followed by
colons. Also you can name the segment so it will be combined with other
segments of the same name in a separately assembled program but I suggest
you make it relocatable on paragraph boundary.
CODE SEGMENT PARA PUBLIC 'name'
I hope this will help you write COM programs, and I'm sorry for length
of this response, reading it back I noticed that I have included a lot of
information that is already available in the DOS manual.
good luck.
Ali Ezzet {ihnp4,uiucdcs}!bradley!ali
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 6 Sep 84 6:47-PDT
Date: 26 Aug 84 21:58:00-PDT (Sun)
To: info-ibmpc @ Usc-Isib.arpa
From: pur-ee!uiucdcs!bradley!ali @ Ucb-Vax.arpa
Subject: RAM disk driver
Seems like the driver is not recognizing the size of the disk and
allocating memory only for 180K. However, when writing to the disk it
checks the PBP and finds that it is a 360K drive and continues writing
beyond the 180K that was allocated for it, thus overwriting any program
that happens to be above it(usually the command processor).
Try poking around with debug, find the INIT routine and try to
modify the requested amount of memory returned by the routine to DOS to
reflect the correct amount of memory needed(this will at least save your
command processor from being overwritten, and possibly crashing the
system)
You can also change this(if you have the source code typed) change
line 248 (in the listing of DOS manual), since the 180k drive requires
2D000H bytes of data space, change it to 5A000H bytes. remember they
are setting a segment address so it should be 5A00H.
By the way if you do have the driver source typed, could you possibly
send it to me. You will save me a lot of typing.
good luck.
Ali Ezzet {ihnp4,uiucdcs}!bradley!ali
------------------------------
Date: Thursday, 6 Sep 1984 07:49:43-PDT
From: waters←1%viking.DEC@decwrl.ARPA (Lester Waters)
Subject: Used PCs available
INFO-PC Digest is not a place for ads, but I think this is
of interest to all. There is a company in Boston, MA which sells
used PCs (of all types). All you do is call them up and describe
the system you wish to sell. You don't have to send them your system
If it is sold through them, they take a flat 10% commission. Meanwhile,
if you sell it yourself, just call them and tell them. There is no
obligation.
In addition to complete systems, you can buy accessories such
as modems, monitors, and software (all used). At the moment, there are
a number of IBM PCs for sale and an expansion unit, among other things.
I am not connected with this service - I'm just posting it as a place
for all to save some $$$.
Boston Computer Exchange
P.O. Box 1177
Boston, MA 02103
Phone: (617) 367-5773 <-- Call to list equipment
and to find out whats
available.
- Lester -
------------------------------
Date: 5 Sep 1984 1145-PDT
From: STERNLIGHT@USC-ECL.ARPA
Subject: AT arithmetic problem (FORTRAN).
To: info-ibmpc@USC-ISIB
I have a FORTRAN program which is supposed to test to see if there is
an 8087 present, and if so, use it. If not, it is supposed to use the
8088 for arithmetic. It is compiled in Microsoft (IBM/Microsoft)
FORTRAN and works fine on the PC, PC Portable, and PC-XT. In running
it on the AT, if an 80287 is installed it works fine. If not, all
numbers are displayed as 1.0. Has anyone encountered this? What is
the cause, the 80286 or the software? I understand Lifeboat C has a
similar problem. Does IBM know about it? Is it to be fixed?
--david--
------------------------------
Date: 6 Sep 1984 7:55:08 EDT
From: Robert M. Simmons <simmons@EDN-UNIX>
Subject: Z80 add-on boards Query
To: info-ibmpc@usc-isib.arpa
-p
Can anyone point me to some manufacturer(s) of add-on cards
for the PC which support a Z80, RAM, and serial IO ports?
Replies addressed directly to me would be greatly appreciated.
Bob Simmons
DCA/DCEC
1860 Wiehle Avenue
Reston, VA 22090
simmons@edn-unix
------------------------------
End of Info-IBMPC Digest
************************
-------
∂12-Sep-84 1540 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #93
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 12 Sep 84 15:39:28 PDT
Date: 12 Sep 1984 14:27:26 PDT
Subject: Info-IBMPC Digest V3 #93
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Wednesday, 12 September 1984 Volume 3 : Issue 93
This Week's Editor: Richard Nelson
Today's Topics:
IBM Price List Available
New Quarter Archive Update
PC/AT Floating Point Performance (3 msgs)
PC/AT Performance (4 msgs)
Use of Undocumented OS Features
Voice Message Service
Simulation Language for XT: SIMSCRIPT
IMSL Math and Stat Libraries in MS-FORTRAN
TeleVideo Block Emulation and more: SOFTERM PC
Microsoft FORTRAN 3.20 Bugs
SIMTERM Bug
SIMTERM Volunteer Maintainer Needed
ANSI.SYS and MS-DOS 2.0 (2 msgs)
Open CON: Hang w/ Lattice C Problem
Lattice C Exit Call Problem
PC/AT Alternatives Query
Foreign Language Character Sets Query
AT&T PC Technical Info Query
High Resolution Graphics Boards Query
Diskette Volume Label C Program Query
MS-FORTRAN Clock and ISA Extensions Routines Query
Graphics Terminal Emulation Query
----------------------------------------------------------------------
Date: 7 Sep 1984 02:55:23 PDT
Subject: IBM Price List Available
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
If anyone is interested there is a file called PRICE.LIST in the INFO-IBMPC
directory. It is about 80K bytes long and inappropriate to send on the
INFO-IBMPC mailing list. This file contains the IBM part number and price
of every item offered by IBM that has anything to do with the IBM-PC.
I will only keep it around for a few weeks, so FTP it now if you are
interested.
------------------------------
Date: 8 Sep 1984 14:55:51 PDT
Subject: New Quarter Archive Update
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
Please note that with the new quarter we have updated all the INFO-IBMPC
archive files and the KWIC-INDEX file is included in the quarterly updated
files. Of course our Table of contents files are updated monthly.
Thanks to all who contribute and particularly to the volunteers who prepare
the table of contents and Kwic-index files.
------------------------------
Date: Sat, 8 Sep 84 16:50:05 EDT
From: GMM Labs <eed←wgmm%jhu.csnet@csnet-relay.arpa>
To: info-ibmpc@usc-isib.arpa
Subject: PC/AT Floating Point Performance
I heard mentioned that the reason IBM chose the slower speed of 6MHz
over a potential of at least 8MHz is solely because the 80287 is only
available in a 6MHz config. Can anyone confirm/deny this?
If this is the case, it seems that an 8MHz 286 would outperform a
6MHz version w/ a 287. A rather cursory comparison between the AT and
the Tandy 2K showed the Tandy to be around 15% faster. Hmmm...
I would like to have the benefits of a Unix system running in protected
mode on a 286 machine. But, I don't think the AT is a good choice.
For the $$$$'s, I'd rather go with an Intel 286/310 or other similar
machine.
-Richard Wilkes
RICK@MIT-MC
eed←wgmm.jhu at csnet-relay
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 8 Sep 84 10:46-PDT
Date: 6 Sep 84 15:09:50-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: sun!gnu @ Ucb-Vax.arpa
Subject: PC/AT Floating Point Performance
Indeed, the 8087 and 80287 are the same inside. There's just a
different bus interface.
Somehow changing the bus interface took them about as long as it took
to produce 80286's. They had working samples in April '83 but at the
time various people were figuring out how to hook them up to 68000's.
For some mysterious reason they decided to withdraw the part until the
80286 was ready.
Here's to open systems.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 9 Sep 84 9:26-PDT
Date: 7 Sep 84 23:35:42-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!ut-sally!ut-ngp!shell!starr @ Ucb-Vax.arpa
Subject: PC/AT Floating Point Performance
Check the specs of the chips... it takes the same number of clock
cycles for the FP operations on both chips. The AT is clocked faster
than the PC, so the 80x87 should run a tad faster on the AT.
Yes, it is a little disappointing.... but real application speeds
(in C) of 10-12K flops isn't shabby for a micro!
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 8 Sep 84 9:41-PDT
Date: 7 Sep 84 0:25:56-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!intelca!kds @ Ucb-Vax.arpa
Subject: PC/AT Performance
I've seen it! A friend picked one up a few days ago, and ran a benchmark
he was particularly interested in: Spice. The approximate times
to spice the same things (using the microsoft fortran, I believe) were
VAX 11/750 w FPA ~30 seconds
IBM PC/AT w 287 FP ~2 minutes
IBM PC/XT w 87 FP ~6.5 minutes
I have no idea what he was spicing, your mileage may vary.
I've got one, two, three, four, five senses working overtime,
trying to take this all in!
Ken Shoemaker, Intel, Santa Clara, Ca.
{pur-ee,hplabs,and,scgvaxd,dual,idi,omsvax}!intelca!kds
---the above views are personal. They may not represent those of Intel.
------------------------------
Date: 9 Sep 1984 12:49:55 PDT
Subject: PC/AT Performance
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: hplabs!intelca!kds@UCB-VAX.ARPA
cc: info-ibmpc@USC-ISIB.ARPA
I don't think your Spice benchmark is an accurate indication of floating point
performance. Unless the floating point calculations are simulated in software,
Spice spends more of its time in integer index calculations than floating point
calculations.
Assuming that the times for the floating point calculations are only slightly
faster for a 6MHz 287, the bulk of the speedup is comming from array index
calculations which should be faster on the Intel 286. Here the 16 bit data
paths and faster effective address calculation of the 286 is giving you the
bulk of the 3 to 1 speedup. The integer multiply was significantly speeded up
on the 286. As FORTRAN uses this instruction for array index calculations, this
should account for an increase in speed. I would expect to see even more
speedup for a large Lotus spreadsheet update which uses no floating point
but does extensive address calculations.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 11 Sep 84 5:26-PDT
Date: 10 Sep 84 6:12:43-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!rochester!ur-laser!tomk @ Ucb-Vax.arpa
Subject: PC/AT Performance
Four times as long as a 750. Now what happens if you were to put 2 users
on it. I bet it would gag, I have seen a 750 running spice and was not
overly impressed by its speed. As Peter Honeyman said, "Its a little
light on cycles".
Tom Kessler {allegra |seismo }!rochester!ur-laser!tomk
Laboratory for Laser Energetics Phone: (716)- 275 - 3786
250 East River Road 275 - 3194
Rochester, New York 14623
------------------------------
Date: 11 Sep 1984 10:16:35 EDT (Tuesday)
From: Jeffrey Edelheit <edelheit@mitre>
Subject: PC/AT Performance
To: hal.dove@ames-vmsb
Cc: info-IBMpc@usc-isib, edelheit@mitre
Mike - I ran a benchmark on the PC/AT, the AT&T, an "old" IBM PC and a Compaq
Deskpro. The Benchmark consisted of a Lotus macro that deleted a large
number of lines, inserted 2 columns, copied one column into another and put
data into another column and finally printed the spreadsheet to a file.
The results were:
PC/AT 1 min. 10 seconds
AT&T 1 min. 15 seconds
Compaq 1 min. 28 seconds
"Old" PC 2 min. 40 seconds
My only observation about the AT&T was that on the 2 systems we used in
the benchmark, the keyboard refused to stay plugged into the computer, and
the keyboard at the computer store was not holding up very well to the
use (abuse?) it was getting.
Hope this helps.
Jeff Edelheit
(edelheit at mitre)
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 8 Sep 84 9:58-PDT
Date: 2 Sep 84 11:11:27-PDT (Sun)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!linus!philabs!cmcl2!seismo!ut-sally!utastro!nather @ Ucb-Vax.arpa
Subject: Use of Undocumented OS Features
>What is really needed is an open architecture operating system to go
>along with the open architecture IBM PC. I for one would love to be
>able to buy a MS-DOS Tech Ref Manual, complete with listing on MS-DOS.
>--
>Roy J. Mongiovi. Office of Computing Services.
>Georgia Institute of Technology, Atlanta GA 30332
I second the motion. IBM's success with the PC is certainly due, in large
part, to the open architecture ... and what have they lost? When they
published their copyrighted BIOS source, they were able to protect themselves
very well against the people who copied it, substituting "a few NOPs" here
and there so the ROMs didn't compare. What has Microsoft to lose by doing
the same with MS-DOS? What they would gain -- in loyal users who can help
with awkward problems, suggest useful extensions, and can recommend MS-DOS
widely because its operation is known and predictable, even in new situations,
-- seems to me well worth while.
Maybe someone else will do it first, for the PC XT or AT -- guess how fast
I'd switch!
Ed Nather
{allegra,ihnp4}!{ut-sally,noao}!utastro!nather
Astronomy Dept., U. of Texas, Austin
------------------------------
Date: 9 Sep 1984 13:01:54 PDT
Subject: Voice Message Service
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
Dick Gillmann has a voice DYM running on an IBM-PC. DYM stands for "Dial
Your Match" and is a popular form of computer bulletin board where people
make a modem connection, match to strangers, and exchange messages.
This system differs from the usual computer bulletin board in that no modem
is required to talk to the computer. This IBM-PC is capable of digitally
recording and playing back speech and interpreting touch tone commands.
The PC has 320K bytes of memory and a 33M Byte external hard disk and a
Compufone telephone interface card. This card interprets the touch tones
and uses enhanced CVSD to record and play back speech at 19.2 bits/sec.
The telephone # is (213)822-9390 and the system is available from 7:00 PM
to 9:00 AM PDT.
------------------------------
Date: Mon, 10 Sep 1984 13:43:14 EDT
From: Macintosh Devaluation Manager <AXLER%upenn-1100.csnet@csnet-relay.arpa>
Subject: Simulation Language for XT: SIMSCRIPT
To: info-ibmpc@usc-isib.arpa
A new version of Simscript II.5 is now available for the IBM PC-XT and
compatible computers. Rather than just move Simscript as it now is
available on mainframe computers, a personal simulation environment
called SimLab(tm) has been developed. With this environment, you
conveniently develop and validate your model. Productivity is greatly
increased through the interactive dialog and extensive error checking
that SimLab supports.
A charter user group with a reduced price and other special benefits
is being formed. Contact Hal Duncan, CACI, 3344 North Torrey Pines
Court, La Jolla CA 92037 (619-457-9681).
The above is taken from simsnips, a simulation newsletter published
by CACI (the makers of both Simscript and SimLab). We use Simscript a
great deal here for various simulation problems, at both the classroom
and research levels, and can recommend it whole-heartedly. If SimLab is
anywhere as good, it'll be an excellent product. I'd suspect, though,
that you'll need a fair amount of memory (and, perhaps, a floating-point
co-processor) to get the best use of the package.
------------------------------
Date: Mon, 10 Sep 1984 13:50:11 EDT
From: Macintosh Devaluation Manager <AXLER%upenn-1100.csnet@csnet-relay.arpa>
Subject: IMSL Math and Stat Libraries in MS-FORTRAN
To: info-ibmpc@usc-isib.arpa
IMSL Inc., makers of the well-known IMSL Library (a collection of 500+
Fortran subroutines for math. and stat. problem solving) have just announced
that two subsets of the library will be made available for the IBM PC and XT,
using the Microsoft Fortran 3.20 compiler under MS-DOS or PC-DOS. The sub-
sets are as follows:
MATH/LIBRARY: differential equations; integration; eigensystem analysis;
error and gamma functions; interpolation, approximation, and smoothing;
linear algebraic equations; vector/matrix arithmetic and sorting; zeros and
extrema.
STAT/LIBRARY: basic statistics; regression analysis; analysis of variance;
nonparametric statistics and tests for goodness-of-fit; time series analysis
and forecasting; random number generation; probability distribution functions
and their inverses.
IMSL has not announced prices on these packages as yet. For additional
information, contact them at IMSL, NBC Building, 7500 Bellaire Blvd., Houston
TX 77036, or call (toll-free) at 800-222-IMSL.
------------------------------
Date: Mon, 10 Sep 84 15:47:13 EDT
From: Dave Swindell <dswindel@bbn-labs-b>
Subject: TeleVideo Block Emulation and more: SOFTERM PC
To: info-ibmpc@usc-isib.arpa
Cc: Newman%sav@lll-mfe.arpa
Softronics in Memphis, TN markets SOFTERM PC, a general terminal
emulation package which emulates many non graphics terminals such as
the TeleVideo 910, 925, and 950. Block mode transmission is supported
under TeleVideo emulation as are function and editing key support.
Although I have had no first hand experience with SOFTERM PC, I have
been using SOFTERM II for the Apple II/II+/IIe for the past year and
have been quite pleased with the package. In addition to the
TeleVideo series, SOFTERM II (and presumably SOFTERM PC) supports the
DEC VT100, VT52, HP 2622(?), Honeywell (various models), as well as
several other non graphics terminals.
In addition to terminal emulation, the SOFTERM package supports computer
to computer file transfer using a non error checking/correcting capture
protocol, the XMODEM error checking/correcting protocol, and an 'in
house' developed error checking/correcting Softerm protocol.
SOFTERM PC is a new product (released August, 1984). I am not sure
how much of SOFTERM II has been included, however, from what I have
been able to glean from information on Softronics' on-line bbd,
SOFTERM PC appears to be a superset of SOFTERM II.
I do not and have not ever worked for Softronics.
Dave Swindell,
BBN Labs
------------------------------
Date: Mon 10 Sep 84 21:04:34-EDT
From: Eric.Crane@CMU-CS-C.ARPA
Subject: Microsoft FORTRAN 3.20 Bugs
To: info-ibmpc@USC-ISIB.ARPA
Date: 10 Sep 84 20:36:25 EDT
From: D. M. Rosenblum <DR01@CMU-CC-TE>
Subject: ALLEGED MICROSOFT FORTRAN 3.20 BUG
To: EC0N@CMU-CC-TE
In the Info-IBMPC digest, Vol. 3, number 7x (x a digit between 1 and 4
inclusive; I think 2), there was a message from Craig Douglas of Yale,
dated 12-Jul-1984 11:00:26, documenting certain bugs in Microsoft Fortran
version 3.20. One of them involved what looked like a pretty serious
error in array indexing, so I wrote a test program to see if I could
reproduce it. The test program, closely following Mr. Douglas's code,
was:
INTEGER H, M(9,5)
H = 24
DO 1 I = 1,5
M(1,I) = H
M(5,I) = H - 1
M(6,I) = H - 1
M(9,I) = H - 2
1 CONTINUE
WRITE (0,*) (M(1,I),M(5,I),M(6,I),M(9,I),I=1,5)
STOP
END
Mr. Douglas claimed (unless I misunderstood him) that this would
generate some strange values. When I compiled it and ran it using
Microsoft Fortran, both versions 3.13 and 3.20 gave the correct
answers, viz. 24,23,23,22 repeated five times. Did I misunderstand
Mr. Douglas or is this bug not replicable?
(Note: I don't have ARPANET access, but the Info-IBMPC digest is
forwarded to our system from systems that do, which is how I get to
read it. Thanks to Eric Crane for forwarding this on to the editors
of the digest.)
-- Daniel M. Rosenblum, Ph.D. candidate,
School of Urban and Public Affairs, Carnegie-Mellon University
<DR01%CMU-CC-TE@CMU-CS-anything>
[The bugs were mentioned in V3 #73. -ed]
------------------------------
Date: Mon 10 Sep 84 08:06:56-PDT
From: Doug Brutlag <brutlag@SUMEX-AIM.ARPA>
Subject: SIMTERM Bug
To: info-ibmpc@USC-ISIB.ARPA
cc: billw@SRI-KL.ARPA
While trying to use SIMTERM from the INFO-IBMPC library with the MODEM
program at SUMEX I ran into the bug described in the following set of
messages. Bill Westfields analysis of the bug proved to be correct and
there is a problem with the SIMTERM protocol as described below. Bill's
suggestion of giving the command MODEM RQ FILENAME.EXT and then waiting
for MODEM to send a C, allows one to use SIMTERM in spite of the bug
but one should not have to use this hack.
Could you please forward this following correspondence to the
Holtzman's. Thank you very much.
Doug Brutlag
PS. This also solves the same problem with PCCOMM from MIT.
Date: Sun 9 Sep 84 17:53:25-PDT
From: Doug Brutlag <brutlag@SUMEX-AIM.ARPA>
Subject: TOPS-20 MODEM PROBLEM AT SUMEX
To: pattermann@SUMEX-AIM.ARPA
cc: billw@SRI-KL.ARPA
Ed,
I was trying out the SIMTERM terminal emulator from Jim and Eric
Holtzman (it is on [ISIB]<INFO-IBMPC>SIMTERM.EXE) on my IBMPC. It
includes support for MODEM file transfer protocol. I found that I could
easily download from SUMEX to my IBMPC with the SUMEX MODEM program but
when I tried to send a file to SUMEX, SIMTERM complained that it never
received an initial NAK and that it received 67 instead. I had a
similar problem with another MODEM program from MIT called PCCOMM.
However, I tried using both of these programs with the MODEM on
ECLD and both of them (SIMTERM and PCCOMM) transfered files in both
directions very well. I got a copy of the ECLD MODEM program on my
directory for my own use. It works with SIMTERM, PCCOMM, IBMFTP and
MacTerminal in both directions whereas the SUMEX version of MODEM would
only download with SIMTERM and PCCOMM.
Do you know what this incompatibility is with the SUMEX MODEM
program? Would you like a copy of SIMTERM to try on your Compaq?
Doug
Date: 9 Sep 1984 22:48-PDT
Subject: Re: TOPS-20 MODEM PROBLEM AT SUMEX
From: BILLW@SRI-KL
To: brutlag@SUMEX-AIM
Cc: pattermann@SUMEX-AIM
Cc: billw@SRI-KL
the 67 is a capital "C", which is used by versions of the modem program
that will use a CRC instead of a simple checksum (the checksum used by
MODEM was found to be especially unreliable on phone connections).
Theoretically, if a version of MODEM supports CRCs, it responds to the
"C" with the first block of the file. If it doesn't support the CRC
option, it ignores the "C" and continues to wait for a NAK (which in
the tops20 version of modem will occur about 4 seconds later). It is
a bug in SIMTERM that the "C" is not properly ignored. However, there
is a simple way around the problem: after typing MODEM RQ FILE.EXT,
wait for the C to be output (about 15 seconds) before starting the
FTP process at the PC side... The tops20 modem will keep sending
"initial NAKs" for quite a while, so you don't even have to hurry
to start it up in the available 4 seconds :-)
BillW
PS: I can't get to ecld, and MODEM.EXE in your directory is protected
so that I can't get at it - I suspect that they either have a version
from before I added CRCs, or perhaps they have disabled.
Date: Mon 10 Sep 84 07:43:57-PDT
From: Doug Brutlag <brutlag@SUMEX-AIM.ARPA>
Subject: Re: TOPS-20 MODEM PROBLEM AT SUMEX
To: BILLW@SRI-KL.ARPA
cc: pattermann@SUMEX-AIM.ARPA
Bill,
Thanks very much for your explanation of the problem. I will send
my problem and your comments on the INFO-IBMPC people at ISIB. I have
unprotected my copy of the ECLD MODEM.EXE if you want to look at it, but
with your explanation I may not have to use it anymore.
Doug Brutlag
------------------------------
Date: 10 Sep 1984 13:45:09 PDT
Subject: SIMTERM Volunteer Maintainer Needed
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
Would anyone like to volunteer to maintain SIMTERM? This is the HP terminal
emulator combined with a MODEM protocol file transfer module.
It looks like a fine program, but we aren't using it here at ISI as we already
have a commercial HP terminal emulator and use Kermit for file transfers.
We don't want to get in the business of maintaining programs for the INFO-IBMPC
library.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 8 Sep 84 11:14-PDT
Date: 1 Sep 84 13:42:17-PDT (Sat)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!noao!hsi!stevens @ Ucb-Vax.arpa
Subject: ANSI.SYS and MS-DOS 2.0 Query
We have an application that uses cursor addressing and screen
control (reverse video, etc.) on the PC. It was developed under
PC-DOS using the ANSI.SYS driver, since that seemed the easiest
way to do it. However, moving to the Burroughs ET-2000 and the
HP 150 I find that their implementations of MS-DOS 2.0 don't have
ANSI.SYS.
Is ANSI.SYS supposed to be standard MS-DOS 2.0, or is it another
IBM-only ?
Richard Stevens
Health Systems International, New Haven, CT
ihnp4 ! hsi ! stevens
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 11 Sep 84 6:01-PDT
Date: 9 Sep 84 11:00:11-PDT (Sun)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!genrad!john @ Ucb-Vax.arpa
Subject: ANSI.SYS and MS-DOS 2.0
I don't know about Burroughs ET-2000 or HP 150
I do know that the TANDY 2000 DOES have ANSI.SYS.
------------------------------
Date: Mon 10 Sep 84 10:54:31-EDT
From: Joseph M. Newcomer <NEWCOMER@TL-20B.ARPA>
Subject: Open CON: Hang w/ Lattice C Problem
To: info-ibmpc@USC-ISIB.ARPA
I'm using a Lattice 2.10 C compiler (small memory model this week). I
have to write a filter that can (at user option) interact with the
keyboard/screen. The first observation is that the 'direct console
I/O' features aren't; they really seem to deal with stdin and stdout.
Thus, asking a question and asking for a response (as the Blaise CTools
do) gets nuked by the redirection of input and output. The file support
with Lattice C talks about opening special devices like 'con:'. Well,
it isn't really like the documentation claims; if you open 'con:', you
get either stdin or stdout depending on the direction. BUT if you
open 'con' (note the absence of the colon) you really get to talk to the
keyboard and screen, even when output is redirected.
OK, great. Now, about 3/4 of the time I open 'con', the PC hangs in a
power-sequence-reboot-needed state. I put appropriate "fprintf(stderr,...)"
statements before and after the file open, and it is clear that it dies
while opening 'con'. Since Lattice doesn't distribute sources, I have
little hope of figuring this out, unless someone with more time than I
has actually taken the time to figure this out. Alternatively, any
suggestions about what I might do would be appreciated. I have NOT
done instruction-by-instruction tracing because the silly thing fails only
SOME of the time. Sigh.
I'm sending a bug report to Unipress, from whom I purchased it. I don't
feel optimistic. At least one '.h' file was missing from the distribution
and I'm still waiting for it after a month.
Any help/suggestions would be appreciated.
thanks
joe
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 11 Sep 84 19:31-PDT
Date: 10 Sep 84 13:23:24-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!noao!hsi!stevens @ Ucb-Vax.arpa
Subject: Lattice C Exit Call Problem
Anyone know what the error message
Memory allocation error
Cannot load COMMAND, system halted
means with PC-DOS 2.0 ? We have an application program compiled
with Lattice C, V2.12 that generates this error after the program
calls exit(). Tracing it with the debugger shows that exit() calls
←exit(), which calls xcexit() which finally does an "int 21" with
AH=0x4c (Terminate a process). Whats puzzling is that the program
always runs fine with one set of input data but always produces the
above error with another set of data. We are not malloc'ing
ourselves at all. Thanks for any leads ...
Richard Stevens
Health Systems International, New Haven, CT
ihnp4 ! hsi ! stevens
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 8 Sep 84 16:52-PDT
Date: 6 Sep 84 4:59:33-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!ecsvax!ctk @ Ucb-Vax.arpa
Subject: PC/AT Alternatives Query
Hello, a group in our department is looking for four fairly large micros.
The machines will be used for communication with supercomputers located out
of state (via telenet, we think) and for local editing and *testing*. Most
of our code is floating point intensive. We've talked to ibm about the AT
and it looks good for the money (in terms of size, lots of ram, big hard
disk ) but it's a bit slow on floats. However there are C and **FORTRAN**
compilers that actually work. The IBM reps said that the compilers, the
Personal Editor, and most programs that call BIOS will work on the AT.
My question is: are we overlooking a good machine somewhere else?
Is there a 68000 box with hardware floats that we can get for < $9000
complete with keyboard, screen, C, FORTRAN (working, please), a decent full
screen editor, and a programmable serial port? With educational discount the
AT looks good for us (we'd like a 20m hard disk at a minimum).
If you know of, or better yet sell, such a box please let me know
about it (via mail, please. This is not a stonehenge, star wars, or net.micro.
msdos article). If I get enough interesting responses I'll report back to the
net.
Thanks in advance --
C.T. Kelley
Dept. of Math.
N.C. State U.
Raleigh, N.C. 27650
decvax!mcnc!ecsvax!ctk
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 9 Sep 84 4:30-PDT
Date: 7 Sep 84 13:13:16-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!duke!phs!lisa @ Ucb-Vax.arpa
Subject: Foreign Language Character Sets Query
<This *is* my first attempt at the infamous net>
Anyone out there know anything about foreign languages on the pc,
mac, or other microcomputers? We in humanities at Duke want to do
word processing and data base work with documents in Greek, Latin,
Hebrew, Coptic, and several other esoteric languages.
Our inclination is to go with one of the "standard" machines and
write patches to handle the non-roman alphabets (I've already done
a lot of this for our Victor 9000s). If, however, someone already
has the software, and would make it available *reasonably*, it would
save a lot of time for me not to reinvent the wheel.
Thanks for the help!
In "real life" --
Jeff Gillette USENET: duke!phs!lisa
The Divinity School
Duke University
Durham, NC 27706
------------------------------
Date: Mon 10 Sep 84 11:08:14-EDT
From: Joseph M. Newcomer <NEWCOMER@TL-20B.ARPA>
Subject: AT&T PC Technical Info Query
To: info-ibmpc@USC-ISIB.ARPA
About once a year, believing that mortification is good for the soul, I
try to deal with a computer store. It takes me about a year to recover.
This past weekend I visited our local ComputerLand. I am trying to decide
among a Tandy 2000, PC/AT, or AT&T PCoid. The Tandy 2000 looks great, has
great high resolution color graphics, dense floppies. Limits are a 10MB
disk, 768K memory, and a "closed" hardware architecture. PC/AT has networking,
DOS 3.x, 20MB disk, but lousy color graphics (same old PC resolution).
AT&T has high resolution graphics (with 4 levels of gray scale on the
monochrome monitor), 10MB disk, dense floppies. Now, try to extract any
meaningful technical info from these guys: "Well, it runs Symphony...".
How fast? "The AT&T has an 8MHz clock..." What does that translate to in
terms of how fast it is relative to a PC/XT or PC/AT? No data. "Oh yes,
the AT&T has an 8MHz clock, but when it boots it is slowed down to 4.7MHz
so it looks like a PC!" Huh? What good is an 8MHz clock if it only runs
that speed when it isn't booted? "Gee, that's a tough question..." Can
I see a technical manual? "No". Why not? "We don't have one." Can you
get one? "No" Why not? "They aren't available". Somebody has to program
these things...how do they do it? "No, nobody has to program them; you
just buy the software..." Somebody wrote it... "Well, it is PC compatible".
I don't want PC compatibility at the clock level, I want it to run as fast
as it can. How do I make that happen? "Gee, that's a tough question" Is
there anybody here who can answer it? "No" Can you get the answer? "No".
Why should I buy it? "It runs Symphony..."
Does anyone out there have any meaningful technical information on the
AT&T PCoid, or know where I can find it (even a number at AT&T would be
a help...they're so confused here I couldn't even find someone who
knew that they made the PC clone)? What is the story about the clock
speed? Anyone know about how file sharing will be handled in 3.x?
Can I use my machine as a file server while running ordinary programs,
for example? or a print server? Is the AT&T backplane really IBM-compatible?
Also, IBM wants $1125 (list) for 512K ram. This is about twice the cost
per K that I'm used to or willing to pay. If anyone knows of a 3rd party
vendor supplying memory cards for the AT I sure would like to know!
Of course, they couldn't tell me if my old RAM cards would work...
thanks
joe
------------------------------
Date: Mon 10 Sep 84 14:05:35-PDT
From: ALFIERI@ECLD.#ECLnet
Subject: High Resolution Graphics Boards Query
To: info-pc@USC-ISIB.ARPA
One of our astronomy professors would like to plot graphs on his PC
(using BASIC) and would thus need a high-resolution graphics card to
work with the monochrome monitor.
Any insights and recommendations are greatly appreciated. Thanks in
advance.
--vincent alfieri, ph.d.
computing information services
university of southern california
alfieri@usc-eclb
------------------------------
Date: Mon 10 Sep 84 18:16:50-EDT
From: Joseph M. Newcomer <NEWCOMER@TL-20B.ARPA>
Subject: Diskette Volume Label C Program Query
To: info-ibmpc@USC-ISIB.ARPA
I have need of either a program or technique to label unlabelled diskettes
or change the label on labelled diskettes. Using CTOOLS2 directory
management, I tried doing a drsfirst operation (which I believe is
DOS call 4E, find first entry) using various matching criteria, including
attribute 8 (volume label), "*.*" as the pattern, etc. It immediately
returned error 18, no more files match pattern. In any case, I had no
idea how, after finding the label, I could change it. I don't want
to have to use DEBUG to patch it; whatever technique has to be usable
by a nontechnical person (e.g., 'label a: foobar' is reasonable, much
more complicated is not). Suggestions would be appreciated. I would
prefer to write the function in C if I have to write it.
thanks
joe
[In Info-IBMPC Digest V3 #7, pur-ee!iuvax!apratt@ucb-vax mentioned a
Lattice C program called label.c that works with disk volume labels.
The program is not in the Info-IBMPC library, but was "posted to
net.sources" according to the message. -ed]
------------------------------
From: Roy Stehle <stehle@sri-tsca>
Date: 10 Sep 1984 1728-PDT (Monday)
To: Info-ibmpc@usc-isib.ARPA
Cc: Stehle@sri-tsca
Subject: MS-FORTRAN Clock and ISA Extensions Routines Query
I need a subroutine, compatible with Microsoft Fortran version 3.2, that
will allow me to read the PC's clock. I suspect that I would be
"reinventing the wheel" if I take the time to write one. Can someone
provide me with a copy?
Does anyone offer (sell) ISA process control subroutines compatible
with Microsoft Fortran?
Thanks in advance,
Roy Stehle
SRI International
------------------------------
Date: Tue, 11 Sep 84 15:19:29 EDT
From: Dave Swindell <dswindel@bbn-labs-b>
Subject: Graphics Terminal Emulation Query
To: info-ibmpc@usc-isib.arpa
I am interested in determining what graphics terminal emulators
have been developed for the IBM PC. Do any emulate terminals
other than the Tektronix 4010 series? Do any support graphics
input devices (mouse, cross hairs, etc...)? I'll summarize the
results if there is sufficient interest.
Thanks in advance,
Dave Swindell
BBN Labs
------------------------------
End of Info-IBMPC Digest
************************
-------
∂14-Sep-84 1342 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #94
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 14 Sep 84 13:39:57 PDT
Date: 14 Sep 1984 11:43:35 PDT
Subject: Info-IBMPC Digest V3 #94
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Friday, 14 September 1984 Volume 3 : Issue 94
This Week's Editor: Billy Brackenridge
Special Issue: Topview
----------------------------------------------------------------------
The following is some propaganda on IBM's new window package "Topview".
I suppose this is as close to any technical information as I can secure
on any windowing package. I would like to run a technical description of
Microsoft Windows, but Microsoft only releases information on windows
if you sign up for the Independent Software Vendor program. As this requires
signing a non disclosure agreement, we can't very well have a discussion
of the merits of these two windowing packages on INFO-IBMPC until the
package is released to the public.
Some industry pundit unfairly described Microsoft Windows as "Vaporware".
I can understand the man's frustration. Here is yet another announcement
of yet another "not ready for release" window package. I assume the intent
of announcing these packages months or years before they are released is
to keep you from buying the competition's product.
Billy Brackenridge
----------------------------------------------------------------------
Questions and Answers to assist
Developers of IBM Personal Computer applications
in evaluating
TopView
and
TopView Programmer's ToolKit
Questions and Answers
General
Multitasking
Windowing
Data Exchange
TopView Programming Conventions
Language Interface Support
Program Information File
Mouse Support
Device Driver Support
Program Installation
Problem Determination
General Questions
Q1. What does TopView do?
A1. TopView provides an operating environment giving applications
multitasking and windowing facilities. For applications that
are specifically designed to run with TopView (i.e., are
programmed using the TopView API) all facilities are avail-
able for developers to include in their applications. In
addition, many existing applications can run in the TopView
environment. But, in general, the user can only take advan-
tage of a subset of TopView facilities when running existing
applications. At the very least, the facility for switching
between co-resident applications is available with existing
applications. For the latest list of applications that are
compatible with TopView, ask your dealer or marketing repre-
sentative. TopView provides a multitasking and windowing
environment for applications that run with DOS 2.0, DOS 2.1
or DOS 3.0; however, only DOS 2.0 and DOS 2.1 functions are
available when using DOS 3.0. TopView features:
- Concurrent execution of multiple tasks and programs.
- Switching between coresident programs.
- Move, size and scroll of an application's windows.
- Display of multiple windows at one time.
- Mouse support.
- Data exchange.
- Several ease-of-use features such as pop-up menus, Help,
on-line tutorial, access to commonly used DOS facilities, etc.
- Support for text applications on both monochrome and
color displays, support for graphics applications on a
color display.
Q2. When will TopView be available?
A2. TopView is scheduled to be available in the first quarter of
1985.
Q3. How much does TopView cost?
A3. The IBM Product Center price is $149.00 (estimated price).
Prices are subject to change without notice.
Q4. What applications will run with TopView now?
A4. Many programs have been designated as "compatible with
TopView" by IBM or the vendor that wrote the program. These
programs have been tested in the TopView environment and can
coexist with other programs running with TopView. However,
depending on the way the program was written, all of the
TopView facilities may not be available when using that
program. For example, a program that writes directly to the
video buffer cannot be windowed or execute while in the
background. Following, is a partial list of IBM applications
that have been tested or are in the process of being tested
with TopView. The fact that an application does not appear on
this list does not mean that the application will not run in
the TopView environment. It may simply mean that the applica-
tion has not been tested in the TopView environment when this
document was written. The easiest and quickest way to find
out if an application will run with TopView is to add the
application to TopView and try it. See questions and answers
on Program Installation.
- BASIC
- Advanced BASIC
- BASIC Compiler
- COBOL Compiler
- DisplayWrite 1
- DisplayWrite 2 Version 1.1
- Easywriter* 1.1 and 1.15
- FORTRAN Compiler
- IBM Filing Assistant
- IBM Graphing Assistant
- IBM Reporting Assistant
- IBM Writing Assistant
- IBM 3101 Emulation
- Macro Assembler
- Multiplan** 1.1
- Pascal Compiler
- PC Cluster
- PCWriter
- Personal Communications Manager
- Personal Editor
- Professional Editor
- Script/PC
- VisiCalc*** 1.2
- Word Proof
* Easywriter is a trademark of Information Unlimited
Software, Inc.
** Multiplan is a US trademark of Microsoft Corporation.
*** Visicalc is a trademark of VisiCorp.
Q5. How does TopView compare to other products that do multi-
tasking and windowing?
A5. TopView provides an operating environment that features:
- Concurrent execution of multiple tasks and programs.
- Switching between coresident programs.
- Move, size and scroll of an application's windows.
- Display of multiple windows at one time.
- Data exchange within and between applications.
- Several ease-of-use features such as pop-up menus, Help,
on-line tutorial, access to commonly used DOS facili-
ties, etc.
- Support for text applications on both monochrome and
color displays, support for graphics applications on a
color display.
The user can run applications specifically designed to run
with TopView and many existing applications.
The TopView Programmer's ToolKit provides:
- Documentation of the TopView Applications Program Inter-
face (API).
- Macros and routines for interfacing applications written
in Assembler or Pascal to TopView.
- The Window Design Aid for creating windows.
- Several utilities for handling windows built with the
Window Design Aid and for building a Program Information
File.
After the product is available, ask your dealer or marketing
representative for a demo of TopView so that you can do your
own comparison of TopView with other products.
Q6. Who developed the TopView products?
A6. TopView was designed and developed by the IBM Corporation,
Entry Systems Division in Boca Raton, Florida.
Q7. Does IBM plan to enhance TopView?
A7. TopView is an important product for the IBM Personal Computer
family of products. Over time, it is IBM's intention to have
TopView capabilities available in the productivity environ-
ments, including 3270 and the IBM PC Network.
Q8. What kind of support and/or warranty is available for TopView
end users? For application developers using TopView and the
TopView Programmer's ToolKit?
A8. TopView is warranted "as is".
Q9. Does TopView support graphics applications?
A9. Yes. TopView supports graphics applications. Graphics
applications run in the foreground using the full screen.
However, TopView does not require graphics so our customers
who do not wish to run graphics applications are not required
to have a graphics adapter or software in order to run
TopView.
Q10. Does TopView provide an open API (Application Programming
Interface) so that any developer can write applications that
take advantage of the TopView function?
A10. Yes. The TopView Programmer's ToolKit has been developed to
document the TopView API, to provide the programming and
language interface tools needed by the developer and to
provide guidelines for the development of applications that
can run in the TopView environment. The TopView Programmer's
ToolKit is a separately priced product that costs $395.00
(estimated price). Prices are subject to change without
notice.
Q11. Can extra copy(s) of the TopView Programmer's ToolKit Refer-
ence Book be purchased?
A11. No. However, you can write to the following address for the
free technical information that may be available about
TopView:
IBM Corporation
Editor, IBM Personal Computer Seminar Proceedings
Post Office Box 1328
Boca Raton, FL 33432
Q12. Does TopView support communications with other IBM PCs or
with a host?
A12. No. The intent of the TopView product is not to provide
applications. TopView's primary function is to provide an
operating environment that has multitasking and windowing
facilities which can be used by software developers to build
programs that are more easily combined into integrated
packages. TopView does include several small applications,
but these are included as examples of programs that use
multiple windows.
Q13. What hardware is required to run TopView?
A13. Minimum hardware required to run TopView:
- IBM Personal Computer, IBM Personal Computer XT, IBM
Portable Personal Computer or IBM Personal Computer AT
- 256KB of RAM system memory
- Either two double-sided diskette drives or one double-
sided diskette drive and one fixed disk
- An 80 character monitor with appropriate adapter
Optional hardware
- Up to 640KB of RAM system memory (512 recommended)
- A color monitor with appropriate adapter
- A printer with appropriate adapter
- A mouse pointing device
Q14. What hardware is required to run the TopView Programmer's
ToolKit utilities?
A14. Mimimum hardware required to run the TopView Programmer's
ToolKit utilities:
- IBM Personal Computer, IBM Personal Computer XT, IBM
Portable Personal Computer or IBM Personal Computer AT
- 256KB of RAM system memory
- Either two double-sided diskette drives or one double-
sided diskette drive and one fixed disk
- An 80 character monitor with appropriate adapter
Optional hardware
- Up to 640KB of RAM system memory (512 recommended)
- A color monitor with appropriate adapter
- A printer with appropriate adapter
- A mouse pointing device (recommended but not required)
Q15. How does TopView affect the performance of compatible
applications?
A15. TopView should have very little effect on the performance of
most applications. However, the performance of an applica-
tion in the TopView environment may vary depending on several
factors which are not under TopView's control:
- Number of other applications running concurrently
- Coding techniques used for each application (e.g.,
constant polling of a device vs. waiting for an
interrupt)
- Type of resources being used by each application (e.g.,
compute bound, heavy I/O use, etc.)
Q16. Is TopView an operating system? If not, what operating sys-
tem(s) and versions are used?
A16. TopView is not an operating system. It works with DOS 2.0,
DOS 2.1 or DOS 3.0 to provide a multitasking and windowing
operating environment. Note that under DOS 3.0, TopView
supports only those functions that are available when running
with DOS 2.0 or DOS 2.1.
Q17. Does TopView run on the 3270 PC?
A17. Yes, TopView will run on the 3270 PC with Control Program
Version 1.2.
Q18 Is there a size limitation for applications running with
TopView?
A18. There is no theoretical limit to the size of an application
that runs with TopView. However, there is a practical limit
which is determined by the amount of system memory, the size
of DOS plus TopView and the total size of all other applica-
tions that the user wishes to run concurrently with this
application. TopView supports the use of up to 640 KB of
system memory. For a PC with 256 KB of system memory,
approximately 80 KB is available for the execution of appli-
cation programs.
Q19. What types of applications are best suited for the TopView
environment?
A19. The programs best suited for the TopView environment include
the following types:
- Programs written specifically for TopView that use the
TopView provided I/O facilities whenever possible.
- Programs that use DOS calls for video display rather
than writing directly to video memory. Alternatively,
programs that use the TopView Interrupt 10 convention to
allow TopView to provide the video buffer address and
redraw facilities.
- Programs that do not excessively poll the keyboard.
- Programs that do not control hardware devices directly.
- Programs that are small in size or that use overlays so
that more programs can be run under TopView at the same
time.
- Programs that close files when the files are not in use.
- Programs that do not require long periods of disk I/O.
Q20. Does TopView support batch files?
A20. No. TopView does not support batch files.
Q21. Do you plan to make TopView run with MS DOS? With a dif-
ferent operating system?
A21. IBM does not speculate on its future plans.
Q22. Does TopView run on "compatible" PCs?
A22. TopView was designed, developed and tested to run on an IBM
Personal Computer with DOS 2.0, DOS 2.1 or DOS 3.0. However,
IBM has made no conscious design decisions that would pre-
clude TopView from running on a "compatible" PC and TopView
has not been tested on this type of hardware.
Q23. How does TopView relate to the IBM 3270 PC?
A23. The two products are similar in some ways but different in
other ways. The user can select the product that best fits
their requirements.
Q24. Why did IBM develop its own multitasking and windowing
program?
A24. TopView was developed to satisfy the key requirements of the
PC users who need greater productivity than is currently
available in the current PC environment.
Q25. Does TopView run on the IBM PCjr?
A25. TopView is an important product for the IBM Personal Computer
family of products. Over time, it is IBM's intention to have
TopView capabilities available in the productivity environ-
ments, including 3270 and the IBM PC Network. However,
TopView function is currently not available on the PCjr.
Q26. Does TopView support IBM's position of an open architecture
for the IBM Personal Computer family?
A26. Yes. The TopView Programmer's ToolKit documents the TopView
application program interfaces (API) that a software develop-
er can use to access the TopView windowing and multitasking
facilities. Also, the ToolKit provides a number of sample
routines and programs that can be used or modified by the
developer as required. In addition, TopView has gone to
great lengths to allow many programs that were written before
TopView was developed to run in the TopView environment.
Multitasking Questions
Q1. How many application can be started at one time?
A1. Theoretically, 255 applications can be started at one time.
The practical limit is the number of applications that a user
can remember and control at one time. However, the real
determinant is the amount of system memory available. Only
23 applications will be displayed in the TopView switch list
menu at any one time but the options on the menu can be
scrolled when there are more than 23 entries.
Q2. What is an application? How is it defined to TopView?
A2. An application is a program that currently runs in the DOS
standalone mode using DOS 2.0, DOS 2.1, or DOS 3.0 or a
program which is specifically written to run with TopView.
It is defined to TopView by means of a "Program Information
File" which contains information describing the special
operating characteristics of the program.
Q3. Is there a difference between a task and an application?
A3. Yes. An application can have one or more tasks. A task is
defined by creating a window and supplying a pointer to a
piece of code (the task). If the pointer is 0, then there is
no task associated with the window. Each task is associated
with a window, but the window may be 0 by 0 in size. Any
window can be updated by any task that has a pointer to the
window. Multiple windows can be associated with the same
task. Each task is dispatched separately (i.e., each window
with an associated task is dispatched separately). First
level interrupt handlers are not interrupted by TopView.
Tasks are dispatched in a round robin fashion, the order of
which is not predictable or controllable by the application
or the user.
Q4. How many tasks can an application have?
A4. An application can have any number of tasks, subject to an
TopView limit of 255 windows (a task must have one or more
windows); however, some of these windows are used by TopView.
In addition, the number of tasks is constrained by the amount
of system memory available.
Q5. Can applications communicate with other applications?
A5. Yes. TopView provides each task with a mailbox and a way to
send and receive data. Applications can also use the shared
program facility to communicate with associated applications.
Q6. Can users specify nicknames for applications?
A6. Yes. When an application is installed, the user can specify
the name (up to 30 characters) by which the application is
known to TopView.
Q7. What data (application, TopView, DOS) is saved when TopView
switches from one task to another?
A7. Refer to the TopView Programmer's ToolKit Reference Book,
Chapter 8, Task State Information Saved By TopView, for a
list of the items that are saved and restored on a task
basis.
Q8. Is the application that is receiving input from the keyboard
or pointing device the only task that is executing?
A8. No. TopView allows all applications to execute concurrently,
whether the application is interactive (receiving keyboard
input) or not. However, TopView also allows the user to
control which non-interactive applications are executing (by
suspending the application). Also, there is a class of
existing applications that cannot execute except when they
are interactive (e.g., applications that write directly to
the video buffer).
Q9. If applications can run in the background (non-interactive)
as well as the foreground (interactive), how does TopView
determine when to stop executing one application and to start
executing another application?
A9. TopView uses a complex algorithm (time-slicing plus natural
program breaks) to determine how much time an application
receives and when switching is to occur. The foreground
application receives more time-slices (and therefore more
execution time) than a background application. Background
applications are divided into two catagories: those that are
compute bound and those that have break points (e.g., wait
for an I/O operation to complete). A program waiting on an
I/O event is moved to the top of the dispatch list when that
event occurs and so is dispatched more quickly than programs
that are compute bound.
Q10. Can the application developer or user determine or set the
dispatching priority ?
A10. No.
Q11. Can an application developer protect a critical section of an
application from being interrupted by TopView?
A11. Yes.
Q12. Can the user get a list (find out) what applications are
currently started?
A12. Yes. TopView provides a pop-up menu that contains a list of
all applications that are currently started. The menu can be
used to switch to another application.
Q13. Can an application determine what other applications are
started?
A13. Indirectly. TopView does not provide a direct facility for
determining which applications are currently started.
However, the application developer can determine if related
applications are running by coding this function in a shared
program.
Q14. Can an application set any global flags that will affect the
operating environment for other applications?
A14. Yes. Refer to the TopView Programmer's ToolKit Reference
Book, Chapter 5, Window Manager Command Bytes, for command
bytes that apply to the system as a whole and not just to the
application issuing the data stream.
Q15. Does TopView allow multiple opens of the same object by dif-
ferent applications?
A15. Each task can open separate objects from the same object
class, but one task cannot open another task's specific
object unless it can find out the object's pointer. These
addresses are not provided automatically. They would have to
be communicated from one application to another via task
mailboxes or a shared program data area.
Q16. Is each application really independent of all other
applications?
A16. No. Applications running in the TopView environment are
affected by each application's use and access of resources
that must be shared. If TopView programming facilities are
used to access these resources, TopView protects the applica-
tion from other applications (which also use the TopView
facilities). However, all applications are vulnerable to
other applications that directly access a resource.
Q17. Can TopView be set up to "remember" across IPL all of the
program variables that have been set?
A17. The basic TopView configuration is remembered (e.g., the
items specified in Setup). The applications that were being
used before the IPL are not.
Q18. After a TopView application is started, what objects exist
and what objects must the application create?
A18. After an application is started, the OBJECTQ, MAILBOX, WINDOW
(main window) and KEYBOARD objects are created, and the
MAILBOX and KEYBOARD objects are opened. The other objects
are not opened.
Q19. Does TopView support a wait/post mechanism? If yes, what can
an application wait on?
A19. Yes. OBJECTQ, MAILBOX, KEYBOARD, POINTER, TIMER.
Q20. Can an application wait on more than one event at a time?
A20. Yes. Via the OBJECTQ.
Q21. Is there any way for an application running in the TopView
environment to determine whether the printer or communication
ports are being used by another application?
A21. No.
Q22. Are interrupt handlers interrupted by TopView?
A22. No. First level interrupt handlers are not interrupted by
TopView.
Windowing Questions
Q1. Can an application have more than one window?
A1. An application can have as many windows as it needs. There
is a limit of 255 windows for TopView. However, some of
these are already used by TopView. The major constraint on
the number of windows in a system is the amount of system
memory available.
Q2. Does the application determine the initial size of a window?
Can it be a different size than full screen (25 by 80)? What
are the minimum and maximum sizes of a window?
A2. Yes, the application determines the initial (and maximum
size) of each window when the window object is created. The
maximum size of a window is 127 by 127. The minimum size is
0 by 0. However, the maximum size window that can be created
by the Window Design Aid is 25 by 80.
Q3. Can the same window be used for both input and output?
A3. Yes. TopView windows can be read from as well as written to.
Q4. Can an application change the size of a window or move the
window on the screen?
A4. Yes. Refer to the TopView Programmer's ToolKit Reference
Book, Chapter 4, Defining Window Data Streams.
Q5. Can applications share a single window?
A5. Yes. It is possible but not recommended because of the
difficulty of coordinating updates to the window.
Q6. Is an application required to have a window?
A6. Yes. However, the window may have a size of 0 by 0 and be
hidden from view.
Q7. Can an application have a window that can't be viewed by the
user?
A7. Yes. An application can hide a window when it is undesirable
to let the user view it.
Q8. Does TopView support multiple (2) displays at the same time?
A8. Yes. TopView will allow users with both a monochrome and a
color adapter to run applications on either display. How-
ever, only one of the displays is used at any given time.
The current foreground (interactive) application determines
which display is to be used.
Q9. Can an application update windows on different displays at
the same time?
A9. No. An application can only use one display at a time but it
can change video modes.
Q10. Does TopView support color for windows? What colors does
TopView use? Can the user change the colors?
A10. Yes. TopView supports both the color text and color graphics
modes on the color graphics adapter. TopView has defined a
set of conventions for color usage and defaults that are used
for both TopView usage and application usage. However,
application developers can use either the pre-defined TopView
colors or colors of their own choosing. Colors used for
TopView functions can be changed to black and white, if this
is preferred or if the user has a black and white monitor
connected to the color graphics card. Colors used by the
applications can be changed by the user if the application
allows it.
Q11. How does an application use a panel object?
A11. When an application sends a panel to TopView, the datastream
codes are used to create a window.
Q12. Can an application cause the screen to be redrawn or is that
under TopView's control?
A12. Yes and yes. The scheduling of the physical redrawing of the
screen is under TopView's control. However, an application
can request that any of its windows be redrawn at any time.
The requesting application is suspended until the redraw has
occurred.
Q13. What input modes does TopView support (full screen, fielded
data, etc.)?
A13. TopView supports a "full screen" input mode which allows a
complete set of fields to be returned to the application at
one time. It also provides character editing and cursor
positioning functions. "Keystroke" mode allows the applica-
tion to interpret and process each keystroke as it is entered
by the user.
Q14. Does TopView support field validation for keyboard input
data?
A14. No. TopView provides several editing functions such as
uppercase conversion and automatic clearing of default data.
However, no checking is performed on data entered in a field.
Q15. What is the Window Design Aid?
A15. The Window Design Aid is an interactive tool that can be used
to design windows for a TopView application. The windows
created by the Window Design Aid are stored in a disk file as
a sequence of datastream codes called a panel.
Q16. How does an application access panels created by the Window
Design Aid?
A16. TopView provides facilities to access panels that reside on
disk or in storage. To access panels from storage, a utility
is provided in the TopView Programmer's ToolKit which will
convert a panel to an object module which can be linked with
an application. Another utility merges groups of panels into
a library of panels so that they can be accessed more readily
from your program.
Data Exchange Questions
Q1. What data exchange functions does TopView support?
A1. TopView supports the CUT, COPY and PASTE of ASCII data taken
from or inserted into a window. However, TopView only
provides the COPY function directly. CUT and PASTE must be
implemented by each application as appropriate, in order for
the user to have this function available. Existing
applications can implement these functions by providing
appropriate filter table entries that translate the selection
of a scissors option into the appropriate keyboard commands
for the application (e.g., a CUT might be translated into a
BLOCK DELETE application command). When the CUT option is
selected by a user, the marked data would be copied to a
TopView buffer and the BLOCK DELETE command would be passed
to the application just as if that command had been entered
from the keyboard. New applications written using the TopView
API can implement the CUT and PASTE functions without requir-
ing filter tables.
Q2. How does TopView know how to manipulate the data in the
application windows in order to perform the data exchange
functions?
A2. TopView does not manipulate the data in an application's
window or in an application's file. These functions are the
responsibility of each application. TopView provides a
facility to initiate the function in a standard way and a
facility to store and transfer data between applications.
Q3. What is the syntax of the filter table entries for cut, copy
and paste functions?
A3. Refer to the TopView Programmer's ToolKit Reference Book,
Chapter 7, Creating a Filter Table.
TopView Programming Convention Questions
Q1. What interrupts does TopView use?
A1. 50H - 57H; 15H (AH=10, AH=12)
Q2. Can an application use the TopView interrupts too?
A2. No.
Q3. Does TopView reserve the use of any keys?
A3. Although TopView uses certain keys, it does not preclude an
application from using any key. For example, the 'Alt' key is
used by many applications in conjunction with another key and
is used by TopView to display the main menu. If the Alt key
is held down and another key is pressed, the keystroke is
passed to the application. If the Alt key is pressed and
released TopView displays its main menu.
Q4. What is the TopView call/return convention?
A4. Refer to the TopView Programmer's ToolKit Reference Manual,
Chapter 2, The TopView Subroutine Call and The TopView
Object/Message Call.
Q5. Does TopView support shared, re-entrant code?
A5. Yes, through the use of a shared program and data. See the
TopView Programmer's ToolKit Reference Book Chapter 8 for a
discussion about Shared Programs.
Q6. Does TopView support subroutine calls?
A6. Yes, see answer to Q4.
Q7. Is there a recommended way to use the printer so that my
application can share the printer resource?
A7. The recommended method for using the printer is to open the
printer (e.g., LPT1:), write to the printer, and then close
the printer. The printer remains in control of your applica-
tion. However, TopView provides a mechanism by which the
user can give control of the printer to another application
that requests it.
Language Interface Support Questions
Q1. What languages are currently supported?
A1. TopView currently supports an interface to the IBM Macro
Assembler. It also provides a sample interface to IBM
Pascal. However, programs that do not use the TopView API can
be written in any language as long as they do not violate the
TopView well behaved program guidelines.
Q2. Are there plans to support any other languages?
A2. Not at this time. However, using the existing interface as
an example, an application developer could write an interface
to another language. The source code for the TopView inter-
face to PASCAL is provided on the TOPVIEW Programmer's
ToolKit diskette to help facilitate this translation.
Q3. What functions do the language interface routines provide the
developer?
A3. The interface routines provide full access to the TopView
subroutine calls (INT 15, AH = 10) and object message calls
(INT 15, AH = 12). See the TopView Programmer's ToolKit
Reference Book for a complete description.
Program Information File Questions
Q1. What information is required by TopView so that an applica-
tion can run with TopView?
A1. TopView must have the following information in order to run a
program:
Program title
Full pathname of the program's start-up file (.COM,
.EXE, etc.)
Data files location (default drive and directory)
Memory requirements (minimum, maximum and system)
Screen type (80 by 25 color ...)
Number of screen pages
Range of software interrupt vectors used (modified)
Window size and offset
Shared program information
Program behavior characteristics
Refer to the TopView Programmer's ToolKit Reference Book,
Chapter 7, TopView's Program Information Files, for a detail-
ed description of each of these pieces of data.
Q2. Does each user have to develop the Program Information File
for each application they wish to run?
A2. Program Information Files have been developed for some
existing applications that are compatible with TopView.
These files are shipped on the TopView diskettes. However,
developers of new applications that run with TopView should
develop a Program Information File as part of their develop-
ment process and ship it on the diskette that contains their
application. It is also possible for a user to run an
existing application that does not have a Program Information
File. When the user adds (installs) the program, TopView
searches the location (disk and directory) for a file named
'program name.PIF'. If no file with that name is found, the
user is prompted for the necessary information. The user
only has to enter the Program Title, the location and name of
the start-up file and the memory requirement. The other
information is generated automatically as default values by
TopView. The user can later modify the program information if
needed.
Q3. Is the user expected to remember and specify all of the
information in the Program Information File each time the
program is used?
A3. No. When the user adds (installs) a program, TopView creates
a record of the information in a master file used by TopView.
There is one record in the file for each application that has
been added. This record must exist before a program can be
started, but it must only be added once.
Q4. How and where is the location of the application's Program
Information File specified to TopView?
A4. The Program Information File must be located on the same disk
and directory as the application.
Q5. When a Program Information File is supplied on the applica-
tion diskette and TopView prompts for this information, does
TopView supply any defaults?
A5. Yes. TopView supplies defaults which will in most cases
allow the application to run. However, these may need to be
modified in order to make the best use of TopView features.
Q6. What happens if the program information for an application is
incorrect?
A6. In most cases, an error message would be displayed when the
user tries to use (start) the program (e.g., "File not found"
or "Not enough memory"). In other cases, such as when the
software interrupt vectors are incorrect, the results can be
unpredictable.
Mouse Support Questions
Q1. Is a mouse required in order to use TopView?
A1. No. A mouse is optional for use with TopView. If necessary,
the cursor control keys and the keyboard can be used as the
TopView pointing device.
Q2. Must all applications support a mouse pointing device in
order to run with TopView?
A2. No. Applications do not have to provide support for a mouse.
However, TopView does provide a pointer device class and
commands that are documented in the TopView Programmer's
ToolKit Reference Book if you wish to provide mouse support.
TopView also provides support for developers of existing
applications that would like to support use of a mouse but do
not want to change their code. This support is provided
through the use of the TopView Filter program and applica-
tion filter tables which are also described in the TopView
Programmer's ToolKit Reference Book. A filter table defines
how TopView should convert mouse movements into keyboard
inputs for each application. The Filter program performs this
conversion.
Q3. What mouse devices are currently supported?
A3. TopView currently supports the following mouse devices:
- Microsoft* Mouse for IBM Personal Computers, Model
number 037-099 (parallel interface)
- Microsoft* Mouse for IBM Personal Computers, Model
number 039-099 (serial interface)
- PC Mouse** by MOuse Systems, Part Number 900120-214
(serial interface)
- Visi On*** Mouse by Visicorp, Part Number 69910-1011
(serial interface)
* Microsoft and the Microsoft Logo are registered
trademarks of Microsoft Corporation.
** PC Mouse is a Trademark of Metagraphics/Mouse Systems.
*** Visi On is a Trademark of Visicorp.
Q4. Can other mouse devices that are currently not supported by
TopView be used?
A4. Yes. Other types of pointing device that satisfy the follow-
ing conditions may be used:
- A standard TopView pointing device driver has been pro?-
vided for the particular device being used.
- The device is capable of presenting either absolute
position information or relative position information
that can be tracked and converted to absolute position
information.
- The device has three independent "buttons" or their
equivalent. For example, devices with two push-buttons
often map the third button to the simultaneous pressing
of both buttons.
To write your own mouse device driver, refer to the TopView
Programmer's ToolKit Reference Book, Chapter 9, Writing a
Pointing Device Driver.
Q5. Is there a special way to setup a mouse to work with TopView?
A5. All the devices listed above, with the exception of the
Microsoft Mouse (parallel interface), connect to an RS-232
serial communications port. These devices may be connected
to either COMM 1 or COMM 2. Care must be taken to insure
that there is not a conflict between the pointing device and
any communications programs over the use of a particular
communications port. When operating with two RS-232 attach-
ments, it is recommended that any serial mouse be attached to
Comm 2. The Microsoft Mouse (parallel interface) connects to
an interface card which must be installed in an expansion
slot. This card has a small jumper which selects the inter-
rupt level that will be generated by the interface card. The
setting of this jumper can be very important since the
interrupts generated by the interface card must not be at the
same level as any other interrupts generated in the system.
The interface card may be set to generate interrupts on level
2, 3, 4 or 5. The following interrupt usage may be helpful
in determining interrupt level settings that should be
avoided when installing the interface card in your system:
- 2 - reserved by IBM
- 3 - Asynchronous communications (secondary)
SDLC communications
BSC communications (secondary)
- 4 - Asynchronous communications (primary)
SDLC communications
BSC communications (primary)
- 5 - Fixed disk
Device Driver Questions
Q1. Can device drivers supplied with TopView be modified to
support other devices or adapter cards?
A1. No. The source code is not provided.
Q2. Can a device driver be written that will run with TopView?
How?
A2. Yes. See TopView Programmer's ToolKit Reference Book.
Q3. Can a device driver be written that will run both with and
without TopView?
A3. Any device driver, except a pointer device driver, written
according to DOS 2.0, 2.1 device driver format will run with
or without TopView. However, device drivers that are not
well behaved, use the CON: device or do keyboard I/O will not
run with TopView.
Program Installation Questions
Q1. Why must applications be installed before running the appli-
cation with TopView?
A1. TopView must know the special operating characteristics of
your program. This information is supplied in the applica-
tion's Program Information File. See the questions on the
Program Information File.
Q2. Since the application is not loaded into memory until I start
it, what actually is installed when the program is added to
TopView?
A2. TopView uses the information from the application's Program
Information File (or requests/defaults the information when
the PIF does not exist) to create a record in the master file
that contains the operating characteristics of all programs
that can be started. The name of the program is also added
to the "Start a Program" menu so that the user can start the
program. See the questions and answers on the Program
Information File.
Q3. How is an application installed to run with TopView?
A3. See Chapter 4 of the TopView User's Guide.
Q4. Must an application be re-installed after every IPL or power
on?
A4. No. Once you install an application, TopView remembers the
information it needs (unless the TopView's program informa-
tion is lost or damaged)
Problem Determination Questions
Q1. Does TopView provide any debug aids to allow the application
developer to determine what is happening at any point while
debugging an application or to determine what caused a
problem?
A1. No. TopView does not provide a debug aid. However, there is
a very good debug aid, the IBM Professional Debug Facility,
that can be used with TopView.
------------------------------
End of Info-IBMPC Digest
************************
-------
∂14-Sep-84 1848 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest v3 #95
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 14 Sep 84 18:47:48 PDT
Date: 14 Sep 1984 18:11:06 PDT
Subject: Info-IBMPC Digest v3 #95
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Friday, 14 September 1984 Volume 3 : Issue 95
This Week's Editor: Dwight Baker
Today's Topics:
Change Key Mapping of \ and shift key: SWAPKEY
Lattice C Update
Blaise CTools bug & fix
PC/AT Floating Point Performance
Foreign Language Character Sets
AT memory and relabeling diskettes
Need help reaching Charles Ditzel
PC TALK problem
EXEC Function Problem
dBASE II or III for Networks Query
----------------------------------------------------------------------
Date: 13 Sep 1984 14:44:18 PDT
Subject: Change Key Mapping of \ and shift key: SWAPKEY
From: Richard Nelson <SWG.NELSON@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
Roy Fan has sent us his program which will switch the key mapping of the
backslash key and left shift key on the IBM PC. Called SWAPKEY.ASM, this
assembly language program is located in <INFO-IBMPC>. The program cautions
that it may not work with certain PC versions or clones because it uses a
specific ROM location.
richard nelson
info-ibmpc editor
------------------------------
Date: Fri 14 Sep 84 00:52:06-EDT
From: Joseph M. Newcomer <NEWCOMER@TL-20B.ARPA>
Subject: Lattice C Update
To: info-ibmpc@USC-ISIB.ARPA
I just received a note that I will receive the Lattice C 2.12 compiler
distribution sometime within the next week. This will hopefully supply
the missing '.h' file. If I'm living right, it may even solve the
'con' bug, although I suspect the BIOS.
joe
------------------------------
Date: Fri 14 Sep 84 01:08:40-EDT
From: Joseph M. Newcomer <NEWCOMER@TL-20B.ARPA>
Subject: Blaise CTools bug & fix
To: info-ibmpc@USC-ISIB.ARPA
The 'stpcvt' routine in the Blaise CTools has a bug which clobbers various
assorted pieces of memory. If you use code 4, suppress trailing blanks
from string, NUL bytes will appear in random places in your world.
To fix it, take the 'stpcvt.c' source, and just before the for-loop
in the suppress-trailing-blanks code, put the assignment 'last=0'. Note
that otherwise 'last' is set only if the loop executes at least once,
and if you pass in a null string it won't and the uninitialized local
'last' then is used to store a NUL byte. I would be more explicit, but
the code IS copyrighted and it would be inappropriate (not to mention
illegal) to publish more. A big gold star to purveyors of utility
libraries who include the source! Although it took me many hours to
figure out where the damage was coming from (mostly convincing myself
that it wasn't my code), once I knew it was in 'stpcvt' it took me about
15 minutes to repair the problem, most of which was spent figuring out
how to use PLIB86 for the first time.
For what I've used, CTools and CTools-2 have saved me more time than this
bug cost me, and were cheap at the price.
joe
------------------------------
From: <pgc%CMU-ITC-LINUS@CMU-CS-PT.ARPA>
Date: 14 Sep 1984 15:35-EST
Subject: PC/AT Floating Point Performance
To: Info-IBMPC@USC-ISIB.ARPA
Hello,
I have noticed that a number of people are wondering why the
PC/AT is on slightly faster than a PC/XT when doing Floating Point
operations while it is significantly faster at most other CPU related
tasks.
The answer lies buried in the PC/AT Technical Reference (known
as TM) and the INTEL Microsystem Components Handbook Vol 1 (known as
MCH). (IBM Part 1502494 and Intel order number 230843-001
respectively.) Page 1-61 in the TR we see that pin 32 (CLK) and pin 37
(CLK286) of the 80287 are connected to pin 10 (CLK) of the 82284. A 12
MHz crystal is attached to the 82284 and pin 39 (CKM) of the 80287 is
tied to ground.
Looking in the MCH we see on page 4-77 the 82284 will drive CLK
at 12 MHz, the 80286 divides this by 2. Now turning to page 4-53 of the
MCH we see what the 80287 will do with this 12 MHz CLK signal attached
to pins CLK & CLK286. CLK286 "provides a sampling edge for 80287 inputs
.... It must be connected to the 80286 CLK input." Not too interesting
as far as speed is concerned. CLK is more interesting. From MCH "this
clock provides the basic timing for internal 80287 operations." This
looks pretty good, the chip is moving at 12 MHz but... CKM tells us the
full story. CKM, Clock Mode Signal is described as follows: "indicates
whether CLK input is to be divided by 3 or used directly." You guessed
it, CKM grounded means divide by 3. Now the 80287 is running at 4 MHz,
SLOWER than the 8087 in the PC & PC/XT running at 4.77 MHz. (You can
check page D-2 in the PC/XT Technical Reference PN 1502237 and data
sheets on 8284 in MCH.) This is about 20% slower Floating Point for the
PC/AT. If we compare the various timings needed for operations on the
80287 and 8087 on pages 4-71 and 3-525 of the MCH we see they are the
same except for some Effective address calculations that need to be
done for 8087. The time consuming operations such as multiply, divide,
transcendentals, require identical numbers of cycles to complete.
In summary, it appears that a program that executed only
Floating Point instructions will run slower on a PC/AT. I assume the
reason you are seeing any improvement at all is due to the great speed
enhancements of 80286 instructions for Moves, simple math, etc.
I hope this sheds some light on the issue of Floating Point speed on the
PC/AT vs PC/XT & PC.
Paul Crumley
------------------------------
Date: Fri, 14 Sep 84 14:30:58 edt
From: ANDERSEN <sigurd%vax1%udel-cc-relay.delaware@udel-relay.ARPA>
To: decvax!mcnc!duke!phs!lisa@ucb-vax.ARPA
Subject: Foreign Language Character Sets
Cc: Info-IBMPC@usc-isib.ARPA
You've got a couple of possibilities for foreign languages on the PC.
If you really want to define your own character sets for display on
the screen, you're going to have a rougher time since you'll either
have to exchange ROMs or use graphics. There's a product called
"Academic Font" that implements a Greek character set in ROM for the
monochrome display. One of the professors here at the U. of Del.
has been using that quite happily in conjunction with WordPerfect.
If you want to do something in graphics, the Hercules adapter is
the standard high-resolution adapter for monochrome displays. You'll
need the vertical resolution it offers if you have very much to do
with accents, etc. IBM just announced two new graphics boards, either
of which would handle your needs. They're expensive, about $550 and
$3,000. The cheaper works with current mono or color displays or a
new higher resolution color display. Of course, to use the graphics
you'll either have to write your own or use an existing program.
If you can afford it, you might want to look into Brit, which has
a simple way of defining new character sets. It's a word processing
program that works with foreign languages and math/scientific notation
with the screen showing what the final page will look like. Brit is
expensive, and requires the Hercules board and a good printer, such
as the Epson LQ1500.
If you don't HAVE to see on the screen the final product but can work
with the built-in ROM characters and learn some mapping of graphic
character to the character of your choice, then you could work with
a printer that had downloadable fonts to get your printed copy the
way you want. The word processing program that we're using for
Humanities faculty here is WordPerfect. We've been happy with it and
the support we get from Satellite Software Int'l.
------------------------------
Date: Fri, 14 Sep 84 14:36:14 edt
From: ANDERSEN <sigurd%vax1%udel-cc-relay.delaware@udel-relay.ARPA>
To: Newcomer@tl-20b.ARPA
Subject: AT memory and relabeling diskettes
Cc: Info-IBMPC@usc-isib.ARPA
Memory for the AT is expensive in part because the motherboard memory
uses pairs of chips that are piggybacked to get 128K bits per socket.
I assume they're soldered together. Your current memory boards will
NOT work in the AT. The 16-bit bus makes sure of that. It won't be
too long before third party boards start appearing for the AT; some
will probably use 256K chips to get more than 512K per board. The
prices on those should drop a lot over the next 6 mos. to a year.
To relabel diskettes, you have (at least) two easy choices. You can
buy the Norton Utilities, which will do the job for you, or get DOS
3.0, which now lets you relabel diskettes. I'm not certain whether
it will let you label an unlabeled diskette.
------------------------------
Date: 8 Sep 84 12:18:23-PDT (Sat)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!watmath!jmsellens @ Ucb-Vax.arpa
Subject: Need help reaching Charles Ditzel
Need help reaching Charles Ditzel at ssc-bee or ssc-vax ?
Apologies for posting to net.micro.pc too, but I don't believe that
net.net-people gets to ARPA types, and Mr. Ditzel mailed me in reponse
to a net.micro.pc article.
Mr. Ditzel: I can't get mail to you (I tried FOUR times!) - can you
please send me your address - the mail that arrived from you did not
have an obvious return address. Thanks!
John M Sellens
UUCP: {decvax|utzoo|ihnp4|allegra|clyde}!watmath!jmsellens
CSNET: jmsellens%watmath@waterloo.csnet
ARPA: jmsellens%watmath%waterloo.csnet@csnet-relay.arpa
[We don't have any name like this on the direct mailing list. If anyone
recognises this message please respond directly. -ed]
------------------------------
Date: 11 Sep 84 13:07:40-PDT (Tue)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!sdcrdcf!trwrba!escher!noao!jacoby @ Ucb-Vax.arpa
Subject: PC TALK problem
I'm having a problem getting XMODEM transfers to work when using the
compiled version of PC-TALK III. The BASIC version works okay.
Every time I initiate a "receive file" (ALT-R), I get the normal message
"Holding for start" and then the phone is hung up, terminating the
connection.
This occurs with DOS 1.1, 2.0, and 2.1. I'm running a Compaq (256K ram)
with a Hayes 1200B. The Hayes is the recent version with NO speaker
volume control.
Any suggestions welcome. Thanks.
ihnp4!noao!jacoby
------------------------------
Date: 7 Sep 84 9:31:20-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!mcvax!tnocsda!jaap @ Ucb-Vax.arpa
Subject: EXEC Function Problem
I cannot get the EXEC function working. In the example program I am trying
to execute the Personal Editor, editting the file 'test'. The EXEC function
results in executing PE, the filename is rubbish (should be test).
Exit from PE results in a hang up of the PC. Does anyone know a solution
to this problem? The DOS manual tells not much about the EXEC function.
Thanks in advance for any help
Jaap van den Eersten
EXAMPLE PROGRAM :
CMD DB 5,' test',0DH ; command line : edit file test
FILE DB 'C:PE.EXE',0 ; ASCIIZ string with filename
; save locations
SAVE←SS DW 0 ; save location for SS register
SAVE←SP DW 0 ; save location for SP register
; parameter block
PARBLK DW 0 ; WORD segment address of environment string
DW OFFSET CMD ; displacement of command line
DW DSEG ; segment containing command line
DW 4 DUP(0) ; no default fcb's
PUSH DS ; prepare return to DOS
XOR AX,AX
PUSH AX
MOV AX,DSEG ; init DATA segment register
MOV DS,AX
LEA DX,FILE ; (DS:DX) points to ASCIIZ string
MOV AL,0 ; free memory
MOV AH,4AH
MOV BX,4
INT 21H
LEA BX,PARBLK ; point (ES:BX) to parameter block
MOV AX,DS
MOV ES,AX
MOV AX,SS ; save SS register
MOV SAVE←SS,AX
MOV AX,SP ; save SP register
MOV SAVE←SP,AX
MOV AL,0 ; call the execute function
MOV AH,4BH
INT 21H
MOV AH,4DH ; call the wait function
INT 21H
MOV AX,SAVE←SS ; restore SS register
MOV SS,AX
MOV AX,SAVE←SP ; restore SP register
MOV SP,AX
RET
------------------------------
Date: Fri 14 Sep 84 16:08:52-PDT
From: ALFIERI@ECLD.#ECLnet
Subject: dBASE II or III for Networks Query
To: info-pc@USC-ISIB.ARPA
Does anyone know if Ashton-Tate did finally release a networking
version of dBASE II, and, if so, how it works?
Thanks in advance.
--vince alfieri
computing information services
usc
alfieri@usc-eclb
[ed. when sending in upgrade to dbase II you may specify 3com
network version..... anyone done that yet?]
------------------------------
End of Info-IBMPC Digest
************************
-------
∂19-Sep-84 2010 INFO-IBMPC@USC-ISIB.ARPA INFO-IBMPC Digest V3 #96
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 19 Sep 84 20:10:08 PDT
Date: 19 Sep 1984 18:54:32 PDT
Subject: INFO-IBMPC Digest V3 #96
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Wednesday, 19 September 1984 Volume 3 : Issue 96
Todays Editor: Billy Brackenridge
Today's Topics:
Professional FORTRAN
General Purpose Interface Bus Adapter & Programming Support
Data Acquisition and Control Adapter & Programming Support
----------------------------------------------------------------------
The following are a few announcements from IBM on new products I
thought would be of interest to the net community. I have edited much
of the IBM verbiage and excluded many of the announcements. I am not
keeping the complete text of the announcements online so contact your
friendly computer store for more information.
Special thanks to our source at IBM, Epsilon from Lugaru, and our
associate editors who are doing the normal digest while I was editing
these monsters.
The IBM Graphics announcements will follow in Issue #97
----------------------------------------------------------------------
IBM Personal Computer Professional FORTRAN
OVERVIEW HIGHLIGHTS
? Supports programs larger than 64K, although no single routine
can be greater than 64K
? Supports data areas and arrays larger than 64K
? Provides comprehensive, interactive symbolic debug
? Provides compile time options for debug, default integer size,
and warning-level error message suppression
? Supports the IBM 8087 and IBM 80287 Math Co-processors
? Includes a library manager
? Supports optional length specifications (INTEGER*2, INTEGER*4,
LOGICAL*1, LOGICAL*4, REAL*4, REAL*8, COMPLEX*8)
? Supports FORTRAN 66 level DO Loop (excluding extended DO Loops)
DESCRIPTION
IBM Professional FORTRAN is an extremely efficient compiler
that makes excellent use of storage space while also improving
execution time for a wide range of programs.
Code optimization is performed automatically at compile time
through a powerful set of techniques that include:
? Subexpression elimination
? Constant folding
? IF optimization
? DO Loop invariant expression removal
? Register allocation over loops
An interactive symbolic debug facility, selectable by the user
at compile time, is capable of a broad spectrum of functions
including:
? Statement breakpoint
? Resume execution with no debug
? Help facility
? Single-step execution
? Trace of each executed statement
CUSTINFO SCHEDULE
The planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
IBM Personal Computer Professional FORTRAN allows the explicit
specification of the following data types: LOGICAL*1, LOGICAL*4,
INTEGER*2, INTEGER*4, REAL*4, REAL*8, and COMPLEX*8. Through a
compile time switch, the user is allowed to set the default of the
INTEGER specification to INTEGER*2 or INTEGER*4.
IBM Personal Computer Professional FORTRAN provides an
interactive symbolic debug facility, which may be selected by the
user at compile time, that includes: statement breakpoint, help,
continue execution, resume execution with no debug, list
breakpoints, list variables, single statement execute, terminate
execution, trace, and log.
Programs greater than 64K are supported, although no single
routine can be greater than 64K. The user has the option at compile
time of specifying that arrays greater than 64K can or cannot be
passed to a subroutine. The default is the small array mode, which
achieves the greatest performance when addressing elements of
arrays.
Floating point arithmetic operations are performed using the
IEEE format and algorithms and the IBM 8087 or IBM 80287 Math
Co-processor.
The compiler is optimized for high-performance execution. Some
of the features included are: constant folding, IF optimization,
subexpression elimination, DO loop invariant expression removal,
optional suppression of warning-level messages, generation of
cross-reference listing, and FORTRAN 66 conventions for DO loop
(excluding extended DO loops).
SPECIFIED OPERATING ENVIRONMENT
HARDWARE REQUIREMENTS:
? An IBM Personal Computer with at least 192Kb of memory.
? A fixed disk and one dual-sided diskette drive, or two
dual-sided diskette drives.
? An IBM Color Display with the IBM Color Graphics Monitor
Adapter, or IBM Monochrome Display with the IBM Monochrome
Display and Printer Adapter. or equivalent.
? An IBM 8087 Math Co-processor or IBM 80287 Math Co-processor
SOFTWARE REQUIREMENTS: IBM Disk Operating System (DOS) 2.1 or later
version
PACKAGING: The compiler, the run-time and debug libraries, and a
sample program, along with the latest version of the linker and
library manager, are supplied on two dual-sided diskettes with the
FORTRAN manuals.
CHARGES
Part Feature One-Time
Description Number Number Charge
IBM Personal Computer
Professional FORTRAN 6024200 4200 $595
The IBM Personal Computer General Purpose Interface Bus Adapter
OVERVIEW HIGHLIGHTS
? Designed to the ANSI/IEEE-488 standard, including the 488A-1980
supplement
? Supports up to 14 devices or instruments
? Provides a direct memory access data rate of up to 300Kb/second
? Provides a programmed I/O data rate of up to 20Kb/second
? Allows user selection of direct memory access channels
? Allows user to select interrupt level
DESCRIPTION
The IBM Personal Computer General Purpose Interface Bus
Adapter and its programming support enable engineering and science
professionals to use personal computers from IBM to access and
control over 2,000 different instruments designed to the IEEE-488
standard. Each personal computer can accommodate up to four GPIB
Adapters and provide support for up to 48 devices. The IBM GPIB
Adapter can perform as a controller, a talker, or a listener with
these compatible devices. The IBM GPIB Adapter also provides
capabilities for data transfer between workstations, and the
connection of several computers for sharing of instruments or
peripheral I/O devices.
CUSTINFO SCHEDULE
The planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
CARD SIZE:
Length: 4.5 inches (11.43 centimeters)
Height: 4.2 inches (10.67 centimeters)
Weight: 4 ounces (114.3 grams)
FCC CERTIFICATION: Class B
POWER: 5 VDC .5% @ 1.0 A max.
OPERATING TEMPERATURE: 15.6 to 32.2 degrees centigrade (60 to 90
degrees Fahrenheit)
OPERATING RELATIVE HUMIDITY: 8 to 80% noncondensing
STORAGE TEMPERATURE: 10 to 43 degrees centigrade (50 to 110 degrees
Fahrenheit)
STORAGE RELATIVE HUMIDITY: 5 to 100% noncondensing
ALTITUDE: 2,187 meters (7,000 feet)
INSTALLABILITY: The General Purpose Interface Bus Adapter is a
half-size card that can be installed in the IBM Personal Computer,
IBM Personal Computer XT, IBM Personal Computer AT, IBM Portable
Personal Computer, or IBM Personal Computer Expansion Unit.
Multiple GPIB cards must all be installed in either the system
unit or the expansion unit, if they use the same interrupt level.
The IBM Personal Computer General Purpose Interface Bus
Adapter Programming Support is available to assist the user in the
installation of this option. It is separately priced and described
in another announcement letter.
A General Purpose Interface Bus Cable must be purchased
separately. It is available from the NDD Supply Catalog (as part
number 2720020).
CHARGES
Single
Unit
Feature Part Purchase
Description Number Number Price
IBM Personal Computer
General Purpose
Interface Bus 1503 6451503 $395
General Purpose Interface Bus Adapter Programming Support
OVERVIEW HIGHLIGHTS
? Allows for automated control and monitoring of IBM General
Purpose Interface Bus devices.
? Provides reconfiguration of address- and device-dependent
parameters via a menu-driven configurator program.
? Allows referencing by symbolic device.
? Interactive control program allows for immediate device control
and verification of the configuration.
? High-level functions provide direct device control, making GPIB
management complexities transparent to the user.
? Provides automatic serial polling, which will improve
programming productivity.
? Low-level functions provide direct application software control
of all GPIB Adapter interface commands.
? Supports application development in BASICA, compiled BASIC,
FORTRAN 2.0, Professional FORTRAN, and C.
? Includes sample application programs in each supported language
? Supports up to four GPIB Adapters and 48 devices.
? Provides programming technique examples for functions supported
by GPIB.
DESCRIPTION
The IBM Personal Computer General Purpose Interface Bus
Adapter Programming Support provides appropriate levels of
capability for both primitive functions that handle GPIB
activities and high-level functions that contribute to ease of use.
The result can be a highly efficient workstation that can help
improve accuracy and productivity in even complex test or
measurement systems.
IBM Personal Computer GPIB Adapter Programming Support offers
a flexible way to control large instrument and measurement systems
accurately and productively. The software includes a loadable
device driver and a high-level language subroutine library to
support the adapter.
The programming support includes an interactive configuration
program that enables the user to easily configure the device driver
software to the desired test configuration. The program permits the
user to define and save multiple software test configurations to
support a variety of applications and test environments. It may
also be used to view a current installation for verifying software
and hardware compatibility.
IBM GPIB Adapter software includes an Interface Bus
Interactive Control Program that allows realtime access to any
instrument on the GPIB. Users can then interactively control and
transfer data to and from GPIB devices. The program allows control
for any function, such as reading, writing, or polling, from the
keyboard. It can also be used for diagnostic activities such as
measuring system performance against program sequences, debugging
applications, or checking the performance of attached devices.
CUSTINFO SCHEDULE
The planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
INSTALLABILITY: The Guide to IBM Personal Computer General Purpose
Interface Bus Adapter Programming Support provides sections on
software requirements, configuration options and selections, and
installation requirements for selected hardware configurations. A
General Purpose Interface Bus interactive utility program is
provided to ensure proper software installation and configuration.
PREREQUISITES:
? IBM Personal Computer DOS 2.0 or a later version
? IBM Personal Computer General Purpose Interface Bus Adapter
PACKAGING: The required programming support will be packaged on one
dual-sided diskette. The Guide to IBM Personal Computer General
Purpose Interface Bus Adapter Programming Support is included.
CHARGES
Part Feature One-Time
Description Number Number Charge
IBM Personal Computer
General Purpose
Interface Bus Adapter
Programming Support 6024201 4201 $85
The IBM Personal Computer Data Acquisition and Control Adapter
OVERVIEW HIGHLIGHTS
Up to four IBM Data Acquisition and Control Adapters can be
attached to an IBM Personal Computer system.
? The Adapter has:
- Four analog input channels (12-bit resolution)
- Two analog output channels (12-bit resolution)
- 16-channel digital input port
- 16-channel digital output port
- Programmable sampling rates provided by a 32-bit timer
- Event counter, programmable rate generator, or programmable
time delay provided by a 16-bit user timer/counter
? The Distribution Panel has:
- Screw terminals provided to attach devices to the
distribution panel
- Multiple grounds for twisted pair terminations
- Shielded construction to minimize noise interference.
DESCRIPTION
The IBM Personal Computer Data Acquisition and Control Adapter
is an analog and digital I/O card that can be plugged directly into
the system unit of the IBM Personal Computer AT, IBM Personal
Computer XT, IBM Personal Computer or IBM Portable Personal
Computer. An optional IBM Personal Computer Data Acquisition and
Control Adapter Distribution Panel is provided to help interface
with external sources.
The IBM Personal Computer Data Acquisition and Control Adapter
and Programming Support enable engineers, scientists and
technicians to use personal computers from IBM for both process
control and custom instrument control. The capabilities of the
adapter and its software, coupled with the IBM Personal Computers,
offer new opportunities for workstations that can support data
acquisition, control, analysis and quality control testing
activities at the level of the laboratory, the pilot plant or the
full-scale production line.
In addition to the ease of obtaining data and automating the
laboratory through the IBM Personal Computer Data Acquisition and
Control Adapter, the IBM Personal Computer can be utilized for data
analysis and data base management.
Application areas that may be addressed with the IBM Personal
Computer Data Acquisition and Control Adapter include:
? Chromatography
? Electrochemistry
? Energy management
? Electronic testing
? Process control
? Data logging
? Robotics
Some parameters commonly monitored or controlled include:
? Pressure
? Flow
? Temperature
? Displacement
? Voltage
? Light intensity
? Rotational speed
Some examples of instruments or devices that may utilize the
Adapter are:
? Chromatographs
? Spectrophotometers
? Pressure gauges
? Relay controls
? Thermocouples
? Gas analyzers
? Humidity sensors
? Valve actuators
? Level gauges
? Load cells
? Conductivity cells
? pH meters
CUSTINFO SCHEDULE
Planned availability is October 1984.
TECHINFO TECHNICAL INFORMATION
TECHNICAL DATA FOR THE ADAPTER
ANALOG INPUT: The analog input functions of the adapter operate in
either programmed or interrupting mode. The analog input functions
provide 12-bit relative accuracy.
Resolution 12 bits
Input channels four differential
Input modes unipolar or bipolar,
user-selectable
Input ranges:
unipolar 0 to +10 volts,
user-selectable
bipolar ?5 and ?10 volts
user-selectable
Output mode:
unipolar straight binary
bipolar offset binary
Input impedance >100 megohms with
<100 picofarads
Input current limited to less than ?4 mA
Input voltage:
Normal mode ?30 volts maximum,
without damage, power on
or power off
Common mode ?11 volts maximum
Common mode
Rejection ratio 72 db
Integral linearity error ?1 LSB maximum
Differential linearity:
Error ?1/2 LSB maximum
Stability ?5 ppm/degrees C of
FSR maximum
Gain:
Error ?0.1% between ranges (maximum)
any range adjustable to 0
Stability ?32 ppm/degrees C of
FSR maximum
Offset:
Error adjustable to 0
Unipolar stability ?24 ppm/degrees C
of FSR maximum
Bipolar stability ?24 ppm/degrees C
of FSR maximum
Monotonicity 0 to 50 degrees C
Throughput to memory 15,000 conversions/second
minimum
The analog output functions of the adapter operate in
programmed I/O mode. The analog output functions provide 12-bit
relative accuracy.
Resolution 12 bits
Number of output
channels two
Output modes unipolar or bipolar,
user-selectable
Output ranges:
Unipolar 0 to +10 volts,
user-selectable
Bipolar ?5 and ?10
volts, user-selectable
Input code:
Unipolar straight binary
Bipolar offset binary
Output Current ?5 milliamps,
minimum, with normal
loading and
protection from
damage with the
output shorted to
common.
Output impedance 2 ohm, maximum
Capacitive loading 0.5 microfarads, maximum
Gain:
Error 0.1% between ranges (max.),
any range adjustable to 0
Stability ?35 ppm/degrees C of
FSR maximum
Offset:
Error unipolar ?3.25 millivolts maximum
Error bipolar adjustable to 0
Unipolar stability ?8 ppm/degrees C
of FSR maximum
Bipolar stability ?24 ppm/degrees C
of FSR maximum
Monotonicity 0 to 50 degrees C
Settling time 10 microseconds
max. to within 0.1% FSR
for a 10 volt step with
1000 pf load
Protection Protected for short
to common
Overshoot ?1% of FSR maximum
Throughput from memory 25,000
conversions/second
DIGITAL INPUT/OUTPUT: The digital input/output functions of the
Data Acquisition and Control Adapter operate in programmed I/O
mode. The adapter provides 16 digital input and 16 digital output
lines that are Transitor Transitor Logic (TTL) compatible. An input
line presents no more than two TTL loads and an output line is
capable of driving at least 10 standard TTL loads. The update rate
for these lines (throughput to/from memory) is at least 25,000
operations per second.
TIMER/COUNTER: Timer/counter functions of the Data Acquisition and
Control Adapter operate in either programmed or interrupting mode.
The adapter includes a 16-bit programmable binary counter.
DIAGNOSTICS: A diagnostic program to test hardware and software is
provided with each Data Acquisition and Control Adapter. The
diagnostic is loaded and controlled by the Diagnostic Control
Program (DCP) and is written in Assembler language. The diagnostic
does not require any device on the bus in order to run, but does
make use of a wrap connector for reliable fault isolation.
POWER REQUIREMENTS: Power requirements of the Data Acquisition and
Control Adapter's are: +5 volts, ?5% at 1.5 amps.
ENVIRONMENTAL REQUIREMENTS:
Operating
Temperature: 15.6? to 32.2?C (60? to 90?F)
Relative Humidity 8 to 80% non-condensing
Storage
Temperature: -40? to 60?C (-40? to 140?F)
Relative Humidity: 5 to 100% non-condensing
Altitude 2187m (7000 feet)
TECHNICAL DATA FOR THE DISTRIBUTION PANEL
The Data Acquisition and Control Adapter provides the optional
distribution panel with screw terminals for user interfacing. The
distribution panel is connected to the adapter by a shielded cable.
The distribution panel is a printed circuit board with four
barrier-type screw terminal strips for a total of 88 terminations.
A shielded flat cable with a 60-pin connector is permanently
attached to the board. The cable and terminal board assembly is
housed in a metal enclosure that is slotted to allow user cabling
to enter or exit. The cable is approximately 34 inches long. The
distribution panel meets FCC Class B requirements.
LIMITATIONS: All adapters must be installed as a group in either
the system unit or expansion unit.
CHARGES
Single Unit
Feature Part Purchase
Number Number Price
DAC Adapter #1502 6451502 $1275
DAC Adapter
Distribution Panel #1504 6451504 245
IBM Personal Computer Data Acquisition and Control Adapter Programming Support
OVERVIEW HIGHLIGHTS
? Automatic control and monitoring of industrial laboratory
instruments or equipment
? High-level language subroutine library support for the IBM
Personal Computer Data Acquisition and Control Adapter
? BASICA, Complied BASIC, FORTRAN 2.0, Professional FORTRAN, and
C support
? Sample application programs in each supported language
? Timer support of application programmable sampling rates and
delays
? Programmable event-counter support for applications
? Support of up to four IBM Personal Computer Data Acquisition
and Control Adapters in a system.
CUSTINFO SCHEDULE
The planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
PREREQUISITES:
? IBM Personal Computer DOS 2.0 or later
? IBM Personal Computer Data Acquisition and Control Adapter
PACKAGING: The required software will be provided on one dual-sided
diskette. The Guide to the IBM Personal Computer Data Acquisition
and Control Adapter Programming Support is included.
CHARGES
Part Feature One-Time
Description Number Number Charge
IBM Personal Computer
Data Acquisition and
Control Adapter
Programming Support 6024202 4202 $160
------------------------------
End of Info-IBMPC Digest
************************
-------
∂20-Sep-84 0528 INFO-IBMPC@USC-ISIB.ARPA INFO-IBMPC Digest V3 #97
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 20 Sep 84 05:24:52 PDT
Date: 19 Sep 1984 18:57:55 PDT
Subject: INFO-IBMPC Digest V3 #97
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Wednesday, 19 September 1984 Volume 3 : Issue 97
Todays Editor: Billy Brackenridge
Special Issue IBM Graphics Announcements:
Enhanced Graphics Adapter
Graphics Memory Expansion Card
Graphics Memory Module Kit
Professional Graphics Controller (#1501)
Enhanced Color Display
Graphics Development ToolKit
Graphical Kernel
Graphical File System:
Plotting System
Graphics Terminal Emulator
----------------------------------------------------------------------
The following are announcements of the new graphics related products
from IBM. These have been edited down to a size that shouldn't choke
anyone's mail system.
----------------------------------------------------------------------
Enhanced Graphics Adapter
OVERVIEW HIGHLIGHTS
? Supports the IBM 5151 Monochrome Display, the IBM 5153 Color
Display or the new IBM 5154 Enhanced Color Display
? Provides 640 x 350 support in up to 16 colors for graphics on
the IBM Enhanced Color Display
? Provides high quality (8 x 14 character box) text in color on
the IBM Enhanced Color Display
? Allows full 16-color graphics with the IBM Color Display in
either 320 x 200 medium definition graphics or 640 x 200 high
definition graphics modes
? Supports 640 x 350 graphics mode for the IBM Monochrome Display
with 8 x 14 character box for text in graphics mode
? Provides compatible 9 x 14 character box on IBM Monochrome
Display in text mode
? Allows users to select under program control from a palette of
64 colors when connected to the IBM Enhanced Color Display
? Provides 64Kb graphics memory on the IBM Enhanced Graphics
Adapter expandable to 128Kb with the Graphics Memory Expansion
Card.
The additional 64Kb of graphics memory increases the color
range for 640 x 350 graphics to 16 colors from the four colors
provided by the base of 64Kb on the IBM Enhanced Graphics
Adapter.
? Expands graphics memory to 256Kb with the Graphics Memory
Module Kit. This added memory may be used to support smooth
scrolling, panning (scanning sequentially through graphics
memory), and additional pages of graphics data.
? Allows compatibility modes to execute programs written for
either the IBM Monochrome Display and Printer Adapter with the
IBM Monochrome Display or the IBM Color/Graphics Monitor
Adapter with the IBM Color Display (see Software
Considerations). Composite video support for attaching analog
monitors or TV sets is not provided.
? Allows RAM-resident character generator to be loaded from user
programs allowing any set of 256 characters to be incorporated
into applications.
Character space expands to 512 with the IBM Graphics
Memory Expansion Card and to 1024 with the IBM Graphics Memory
Module Kit. Allows character box sizes up to 32 dots high and
8 dots wide.
? Supported by virtual device interface as described in today's
IBM Personal Computer Graphics Development ToolKit announcement
in both 640 x 350 graphics for the IBM Monochrome Display and
the IBM Enhanced Color Display and compatible graphics modes
for the IBM Color Display and the IBM Enhanced Color Display.
DESCRIPTION
The IBM Enhanced Graphics Adapter, IBM Graphics Memory
Expansion Card, and the IBM Graphics Memory Module Kit have been
designed to provide a higher level of color graphics and text for
the IBM Personal Computer, IBM Personal Computer XT, and IBM
Personal Computer AT.
The IBM Enhanced Graphics Adapter provides support for the new
IBM Enhanced Color Display (5154) and enhanced support for the IBM
Color Display (5153) and the IBM Monochrome Display (5151).Text and
graphics applications are both supported with a wide variety of
display modes. On the IBM Enhanced Color Display, the IBM Enhanced
Graphics Adapter provides up to 640 x 350 support in up to 16
colors for high definition graphics and an 8 x 14 high quality
character box for text. On the IBM Color Display, the IBM Enhanced
Graphics Adapter provides 16 graphic colors in either 320 x 200
medium definition graphics or 640 x 200 high definition graphics.
On the IBM Monochrome Display, the IBM Enhanced Graphics Adapter
supports 640 x 350 graphics with a 9 x 14 character box in text
mode. Under program control, the 16 colors may be selected from
a palette of 64 when using the IBM Enhanced Color Display.
The 64Kb built-in memory of the IBM Enhanced Graphics Color
Adapter expands to 128Kb with the Graphics Memory Expansion Card
and to 256Kb with the IBM Graphics Memory Module Kit added to the
IBM Graphics Memory Expansion Card. The IBM Graphics Memory
Expansion Card is installed in a socket on the IBM Enhanced
Graphics Adapter. The IBM Graphics Memory Module Kit is installed
in the sockets provided on the IBM Graphics Memory Expansion Card.
The IBM Graphics Memory Module Kit allows eight loadable
128-character sets, smoother panning and scrolling, and additional
pages of graphics.
CUSTINFO SCHEDULE
Planned availability is October 1984.
TECHINFO TECHNICAL INFORMATION
See Technical Data following "Charges" below.
SPECIFIED OPERATING ENVIRONMENT
HARDWARE REQUIREMENTS: The IBM Enhanced Graphics Adapter can be
installed in any full-size system expansion slot of an IBM Personal
Computer, IBM Personal Computer XT, and IBM Personal Computer AT
(not the 5161 Expansion Unit). See the table describing the
BIOS-supported operating modes at the end of this letter.
SOFTWARE CONSIDERATIONS: The IBM Enhanced Graphics Adapter may be
programmed to use the newly announced virtual device interface for
graphics output which allows device-independent programs to be
written that can be transported between various systems with
various graphics hardware.
IBM Personal Computer BASIC functions "COLOR," "PAINT," and
"SCREEN" support compatibility modes only.
A BIOS interface, compatible with the IBM Color/Graphics
Monitor Adapter and with the IBM Monochrome Display and Printer
Adapter, is provided. Programs using sub-BIOS programming may not
run with the IBM Enhanced Graphics Adapter.
Some existing application software uses attribute values not
specified in the IBM Technical Reference Manual to run without
change on either the 5153 or the 5151 displays. These programs may
have screens which appear on a 5151 driven by the Enhanced Graphics
Adapter with a different shade of background then when driven by a
Monochrome Display and Printer Adapter. Sometimes text becomes
unreadable. Examples of these programs are Multiplan* (logo screen
only) and Exploring the IBM Personal Computer.
* U.S. Trademark of Microsoft Corporation
Existing text programs using 80-column modes with the 8 x 8
character box will run without change and will automatically use
the higher quality 8 x 14 character box when using the IBM Enhanced
Color Display.
LIMITATIONS: This adapter may be co-resident with the Monochrome
Display and Printer Adapter or with the IBM Color/Graphics Monitor
Adapter. However, only one adapter may drive a color monitor at any
time, and only one may drive a monochrome monitor.
Note: If the IBM Enhanced Graphics Adapter is being used to
replace the IBM Monochrome Display and Printer Adapter (to provide
graphics capability for the Monochrome Display), an IBM Printer
Adapter must be added to replace the printer adapter function on
the IBM Monochrome Display and Printer Adapter. The IBM Monochrome
Display and Printer Adapter must be removed.
INSTALLATION/OPERATION: A four-switch module on the back of the
card (accessible without removing the card) specifies the display
attached, which monitor is primary when two displays are included
in the system, and which mode (80 or 40 character) the card will
assume when the system is powered on.
The IBM Graphics Memory Expansion Card can be installed in a
socket on the side of the IBM Enhanced Graphics Adapter. Modules
from the IBM Enhanced Graphics Memory Module Kit can be installed
in sockets provided on the IBM Graphics Memory Expansion Card.
Note: Only one system unit expansion slot is required
regardless of the amount of graphics memory installed.
PREREQUISITES: The IBM Personal Computer models made for domestic
use (U.S. and Canada) with serial #0300960 or lower require a new
BIOS module to install these new options. IBM Personal Computers
requiring a new BIOS module can be upgraded using the BIOS Update
Kit (#1005).
(The serial number is on the back of the system unit.) IBM Personal
Computer XTs or IBM Personal Computers with IBM Expansion Units or
IBM Cluster Adapters do not have to be modified.
The IBM Enhanced Graphics Adapter is a prerequisite for the
IBM Graphics Memory Expansion Card.
The IBM Graphics Memory Expansion Card is a prerequisite for
the IBM Graphics Memory Module Kit.
CHARGES
Single-Unit
Feature Part Purchase
Description Number Number Price
IBM Enhanced
Graphics Adapter 1200 1501200 $524
Mem. Expan. Card
IBM Graphics 1201 1501201 199
IBM Graphics
Memory Module 1203 1501203 259
TECHNICAL DATA
The following table describes the supported operating modes
that can be selected under program control.
Alpha/ Chars x Char Picture Pages Monitor
Graphics Colors Lines Box Elements (max.) Required***
Graphics 16 40x25 8x8 320x200 8 5153
Graphics 16 80x25 8x8 640x200 4 5153
Graphics 2* 80x25 8x14 640x350 1 5151
Graphics 2* 80x25 8x14 640x350 2 5151
Graphics 4/64 80x25 8x14 640x350 1 5154
Graphics** 16/64 80x25 8x14 640x350 2 5154
Alpha 16/64 80x25 8x14 640x350 8 5154
Alpha 16/64 80x43 8x8 640x350 4 5154
Alpha 2* 80x43 9x8 720x350 4 5151
Alpha 2 40x25 8x8 320x200 8 5153
Alpha 16 40x25 8x8 320x200 8 5153
Alpha 2 80x25 8x8 640x200 8 5153
Alpha 16 80x25 8x8 640x200 8 5153
Graphics 4 40x25 8x8 320x200 1 5153
Graphics 2 40x25 8x8 320x200 1 5153
Graphics 2 80x25 8x8 640x200 1 5153
Alpha 2* 80x25 9x14 720x350 8 5151
* Plus blink and high intensity attributes.
** This mode requires at least 128Kb of graphics memory.
*** IBM 5154 Enhanced Color Display can also support all modes
shown for the 5153.
IBM Personal Computer Professional Graphics Controller (#1501)
OVERVIEW HIGHLIGHTS
? IBM Personal Computer Professional Graphics Controller (#1501)
- Provides expanded graphics mode.
- Runs most of the existing IBM Personal Computer software in
IBM Color/Graphics Monitor Adapter emulation mode.
- Enhanced text character set in emulation mode.
- 640 x 480 pel definition.
- 256 colors from a palette of 4096.
- Hardware has built-in two-dimensional and three-dimensional
capability for:
. drawing
. rotating
. translating
. scaling
- Includes an Intel 8088 processor for high-performance
graphics.
- Move and draw with absolute or relative coordinates.
- User-redefinable color selection.
- Built-in or user-programmable character set.
- Variable character size.
- Supports all models of the IBM Personal Computer family
except IBM PCjr.
- Vector and polygon drawing, and polygon fill
- Power on diagnostics.
? IBM 5175 Personal Computer Professional Graphics Display
- Separate brightness, contrast and on/off controls
- Same physical dimensions as the IBM 5153 Color Display
- 640 horizontal x 480 vertical pel definition
- 67 pels per inch both horizontally and vertically
- Anti-reflective coating
- Darkened screen enhances contrast
- 0.31 mm, shadow mask display provides clarity and detail
DESCRIPTION
The IBM Personal Computer Professional Graphics Display and
Professional Graphics Controller enable engineers, scientists,
technicians and designers to use IBM Personal Computers as
integrated workstations which offer high-quality color graphics
capabilities for a wide range of specialized applications. The
display and controller offer a variety of graphic functions and
many ease-of-use and customizing features to help create advanced
graphics. These facilities can also help with computer-aided
design (CAD), computer-aided manufacturing (CAM), computer-aided
engineering (CAE), image processing, and business presentation
graphics.
The IBM Personal Computer Professional Graphics Controller
requires two adjacent slots in an IBM Personal Computer AT, IBM
Personal Computer XT, or IBM Personal Computer Expansion Unit. The
controller can emulate the current IBM Personal Computer
Color/Graphics Monitor Adapter.
Many graphics functions are provided on the controller cards,
reducing the need to load software subroutines for key graphic
activities. The following functions and commands are supported on
the Professional Graphics Controller:
Two-dimensional and Three-dimensional Graphics Primitives
Absolute and Relative Move
Absolute and Relative Line
Absolute and Relative Polygon
Absolute and Relative Point
High-Level Two-dimensional Primitives
Arc
Circle
Ellipse
Rectangle
Sector
Modeling Transformations
Modeling Matrix
Matrix Read/Write
Matrix Initialize (Identity)
Rotate X, Y, Z
Scale
Translate
Projection
Perspective Projection
Orthogonal Projection
Viewing Matrix
Matrix Read/Write
Matrix Initialize (Identity)
Rotate X, Y, Z
Distance
Translate
Window/Viewport/Clipping
Two-dimensional Viewport Clipping
Three-dimensional Hither/Yon Clipping
Two-dimensional Window - Viewport Mapping
Command List
List Begin
List Delete
List End
List Loop
List Read
List Run
Colors - Fills - Patterns
Area Fill
Seed Fill
Area Pattern
Line Pattern
Clear Screen
Set Color
Set Screen Color
Set Mask/Fillmask
Primitive Fill
Complementary Draw
Look-Up Table (LUT) Operations
Set LUT Values
Initialize LUT
Read LUT
Save LUT
Text
Text Angle
Define Own Text Characters
Text Font
Text Size
Text Justify
CUSTINFO SCHEDULE
Planned availability is October 1984.
TECHINFO TECHNICAL INFORMATION
? IBM Personal Computer Professional Graphics Controller
- 60 frames per second non-interlaced
- 16 x 8 character box in emulation mode
- Onboard 8MHz Intel 8088 microprocessor
- 25MHz video pel rate
- Screen clear/color flood feature
- 320Kb of display storage
. 20Kb display lists and internal variables
. 300Kb display data
- 64Kb of on-board graphics microcode
- 8 bit planes available for read/write
- Power requirements: 5 volts of direct current, plus or
minus 5%; 5.0 amps maximum
- Dimensions: The controller fits into two adjacent slots in
the IBM 5161 Personal Computer Expansion Unit, the IBM 5160
Personal Computer XT, or the IBM 5170 Personal Computer AT.
. Card set length: 333.25 mm (13.12 in.)
. Card set height (inch tabs): 106.7 mm (4.2 in.)
. Card set thickness: 1.52 mm (.06 in.)
. Maximum component height: 12.7 mm (.5 in.)
. Card spacing: 20.3 mm (.8 in.)
? IBM 5175 Personal Computer Professional Graphics Display
- Input requirements
. Video signals
RGB signals
0 to 0.6V in 40 MV increments
16 levels (4 bits)
- Sync signals
. Composite horizontal and vertical
. Non-interlaced mode
- Cables
. Signal - 1.5 meters
. Power - 1.8 meters with 3-wire grounded plug
- Interface frequencies
. Video Frequency: 25 MHz
. Horizontal frequency: 30.48 KHz
. Vertical Frame Rate: 60.0 Hz
- Display Tube
. 14 in. diagonal glass measurement, 13 in. diagonal
viewing area
. Contrast enhancement/anti-reflection screen
- Addressability
. 640 horizontal x 480 vertical
- Power Supply
. Input voltage: 90 volts AC min.,
137 volts AC max.,100-127 volts AC nom.
. Input Frequency: 60Hz ? 3Hz
? Graphics Controller and Graphics Display
- Electromagnetic
. FCC class B rating
- Operating Environment
. Temperature: 15.6 to 32.2 degrees C
(60 to 90 degrees F)
. Relative humidity: 8 to 80% non-condensing
- Shipping and Storage Environment
. Temperature: 10 to 43 degrees C (50 to 110 degrees F)
. Relative humidity: 5 to 100% non-condensing
- Altitude: 2187m (7000 ft.)
LIMITATION: The IBM Personal Computer system unit requires another
monitor to run diagnostics when the 5175 Professional Graphics
Display is attached via the 5161 Expansion Unit.
Model/ Single Unit
Feature Part Purchase
Number Number Price
IBM 5175 Personal Computer
Professional Graphics Display 001 -- $1295
The IBM Enhanced Color Display
OVERVIEW HIGHLIGHTS
? Operates with the IBM Enhanced Graphics Adapter or with the IBM
Color/Graphics Monitor Adapter
? Provides 640 x 350 pel definition in up to 64 colors for
graphics (16 colors at one time when operating with the IBM
Enhanced Graphics Adapter, lower limits when operating with
the IBM Color/Graphics Monitor Adapter)
? Displays high quality (8 x 14 character box) text in color
? Includes a dual frequency design that permits 22K Hz for new
modes (640 x 350) and 15.75K Hz for compatibility modes (640 x
200 and 320 x 200)
? Incorporates a dark, etched screen for high contrast and
reduced glare
? Provides brightness and contrast control located on front of the
display
DESCRIPTION
The IBM Enhanced Color Display is designed to provide an
improved level of color graphics when used with the IBM Enhanced
Graphics Adapter (#1200) for users of the IBM Personal Computer,
IBM Personal Computer XT, and IBM Personal Computer AT. The IBM
Enhanced Color Display provides 640 x 350 pel definition from a
palette of 64 colors, 16 at one time when operating with the IBM
Enhanced Graphics Adapter. The compatibility modes allow the IBM
Enhanced Color Display to operate with either the IBM Enhanced
Graphics Adapter or the IBM Color/Graphics Monitor Adapter.
The IBM Enhanced Color Display also features a dark, etched screen
for high contrast and reduced glare.
The IBM Enhanced Color Display is similar in size and shape to the
existing IBM 5153 Personal Computer Color Display.
CUSTINFO SCHEDULE
Planned availability is January 1985.
TECHINFO TECHNICAL INFORMATION
? Horizontal scan frequency 22K Hz or 15.75K Hz, depending on the
polarity of the vertical sync signal from the adapter
? Vertical scan frequency 50 to 60 Hz
? 0.31 mm dot pitch
? Non-interlaced operation
? Operating temperature: 15 to 30 degrees C
? Non-operating temperature: 10 to 43 degrees C
? Relative humidity: 8 to 80%
? Max. wet bulb dew point 27 degrees C
? Size: Width 15 1/2" (394 mm), Depth 17" (432 mm),
Height 11 1/2" (292 mm)
? Weight: Approximately 32 lbs
? Signal cable length: 3' 6"
? Normal persistence phosphors (P22 equivalent)
? Electrical: 90 - 137 volts AC, 50/60 Hz.
SPECIFIED OPERATING ENVIRONMENT
HARDWARE REQUIREMENTS: The IBM Enhanced Color Display requires that
the IBM Enhanced Graphics Adapter or IBM Color/Graphics Monitor
Adapter be installed in the IBM Personal Computer system unit it
will be attached to.
Single Unit
Description Type Model Purchase Price
IBM Enhanced 5154 001 $849
Color Display
The IBM Personal Computer Graphics Development ToolKit
OVERVIEW HIGHLIGHTS
? Provides device independence and allows applications to support
new graphic hardware devices without modification to the
application.
? Provides graphic and text functions via high performance
graphic subroutines including:
- Circle, arc, pie slice, bar charts
- Multiple colors and fill patterns ("Paint an area")
- Polyline, polymarker
- Graphic and cursor text
. Multiple fonts
. Rotatable
. Scalable
- Raster operations
- Input operations
? Supports BASIC, FORTRAN, and PASCAL compilers and IBM Macro
Assembler
DESCRIPTION
The IBM Personal Computer Graphics Development ToolKit is a
powerful and flexible tool for graphics programmers and application
developers. It can be used as a productivity tool for creating
graphics, and also provides an interface between device-independent
software and graphics devices.
Using the ToolKit enables the programmer or developer to
create graphics software that remains independent of the graphic
I/O devices supported. This gives the software a greater range of
portability and compatibility, and allows the end user a greater
range of choice in the selection of graphics hardware.
A number of basic and advanced functions are provided through
the ToolKit and the Virtual Device Interface at both the device and
program levels. The functions include:
? Workstation control that allows applications to open, close,
and update workstations
? Driver loading capability to dynamically swap device drivers
? Line drawing ability, including capability to draw lines on all
devices that do not support line drawing in the hardware
? Graphic text strings that are scalable, rotatable, and
available in multiple-fonts
? Graphic keyboard input that allows cursor movement in alpha
text
? Writing mode that allows 16 different boolean operations
performed on picture elements during graphic, graphic text, and
alpha text output to the display
? Function key reading from the keyboard by application programs
? Unity aspect ratio allowing applications to preserve aspect
ratios of graphics on devices that are not 1:1 (such as the IBM
Personal Computer Color Printer and IBM Color Graphics Adapter)
? Polygon and bar emulation, which guarantees the ability to draw
polygons and bars on all devices
? Line style emulation to provide the drawing of solid,
long-dashed, dotted, dash-dotted, medium-dashed, or dashed with
two dotted lines
? Marker emulation to provide marker symbols across all devices
? Text alignment allows text to be aligned using left, right,
center, top, and bottom justification
? Document text capability, including multiple fonts, variable
interline spacing, underlining capability, and other attributes
for generating formal documents
? Rubberbanding ability to stretch and contract lines dynamically
under application program control
? Inking ability to fill pre-existing polygons
? Background color allows selection of a background color during
graphics output
? Application programs have the ability to read cursor keys from
the keyboard
? Device-independent input manager, which provides input to the
application, independent of the type of input device (such as
mouse, joystick)
? Allows for simultaneous input and output
? Allows output to DASD for later display with Graphical File
System
The Graphics Development ToolKit implements the VDI controller
as a DOS device driver. The VDI controller allocates additional
space for the largest tool kit device driver only. The VDI
controller then dynamically loads the appropriate driver when
needed. Only one ToolKit device driver need be in memory at a time.
Enhanced performance is available by defining multiple swap areas
or by making multiple device drivers resident in main memory.
CUSTINFO SCHEDULE
The planned availability date is September 1984.
TECHINFO TECHNICAL INFORMATION
PACKAGING: Three dual-sided diskettes containing:
? A set of DOS device drivers for:
- VDI (Virtual Device Interface) Controller
- IBM Color/Graphics Monitor Adapter
- IBM Graphics Printer
- IBM Color Printer
- IBM Compact Printer
- IBM Game Adapter and Keyboard
- IBM PCjr Video Adapter
- IBM Enhanced Graphics Adapter
- IBM 7371 Plotter
- IBM 7372 Plotter
? A set of language bindings that provide access to functions
contained in device drivers. Language bindings and reference
booklets are provided for:
- IBM BASIC 1.0 Compiler
- IBM FORTRAN 2.0 Compiler
- IBM PASCAL 2.0 Compiler
- IBM Macro Assembler 1.0
? Sample programs
? Also included is a programming manual that provides a
description of the VDI functions and how the functions are
accessed from each supported language.
UPDATE INFORMATION SERVICE: The Graphics Development ToolKit
contains a card that allows the customer to receive updates.
This service is designed to ensure that all users are given the
opportunity to purchase products that reflect all announced
products or features and to maintain them on a current basis
at no charge. To participate in this program, the user must return
a completed Update Information Service registration card, which is
included with the Graphics Development ToolKit package.
IBM will provide you with any updates made available prior to
your purchase of the IBM Personal Computer Graphics ToolKit. In
addition, you will continue to receive updates for this product, as
they are made available, through February 28, 1986.
CHARGES
Part Feature One-Time
Description Number Number Charge
Graphics Development
ToolKit 6024196 4196 $350
IBM Personal Computer Graphical Kernel
OVERVIEW HIGHLIGHTS
? Provides portability of graphics applications between computer
systems that support the Graphical Kernel System (GKS) standard
? Provides device independence to graphics applications
? ANSI level implementation with two-dimensional segmentation
features of the GKS standard
? Supports segment rotation, scaling, and transformation
? Supports two-dimensional graphic primitives
? Provides language bindings of FORTRAN 2.0, Professional
FORTRAN, C, and BASIC Compilers
? Supports transformation between world, normalized device
coordinate, and device coordinate systems
DESCRIPTION
The IBM Personal Computer Graphical Kernel System is a
powerful tool for programming and applications developers. It helps
to increase both applications flexibility and the device base on
which these applications may be used. It is source code indepen-
dent, providing graphics program portability while using as much of
the device's inherent capabilities as are available.
CUSTINFO SCHEDULE
Planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
SPECIFIED OPERATING ENVIRONMENT
HARDWARE REQUIREMENTS: IBM Personal Computer, IBM Personal Computer
XT, IBM Portable Personal Computer, or IBM Personal Computer AT
with:
? Minimum configuration
- 256Kb memory
- Two dual-sided diskette drives
- Graphics display and appropriate adapter
? Recommended configuration enhancements
- 384Kb memory
- IBM 8087 or IBM 80287 Math Co-Processor, as appropriate
- IBM fixed disk and one dual-sided diskette drive
- Hardcopy device and input device
SOFTWARE REQUIREMENTS:
? DOS 2.1, or later version
? IBM Personal Computer FORTRAN Version 2.0, IBM Personal
Computer Professional FORTRAN, C, or BASIC Compiler
PACKAGING: The IBM Personal Computer Graphical Kernel System is
packaged in four dual-sided diskettes with the IBM Graphical Kernel
System Programmer's Guide and Language Bindings - Volumes 1, 2, and
3. Device drivers and associated modules are included in two
additional diskettes.
CHARGES
Part Feature One-Time
Description Number Number Charge
IBM Personal Computer
Graphical Kernel
System 6024203 4203 $295
The IBM Personal Computer Graphical File System:
OVERVIEW HIGHLIGHTS
? Pictures and computer-generated graphics stored independent of
devices
? Portability of graphics between computer systems that support
the Graphical Kernal System (GKS) standard, depending on each
computer system's functional capabilities and implementation
? Transfer of pictures or graphics images in standard format
between programs and systems
? Easy-to-use, icon-driven, interactive interface
? Program interface via language bindings to FORTRAN 2.0,
Professional FORTRAN, Compiled BASIC, and C
? Assembly of graphics images from totally different applications
CUSTINFO SCHEDULE
The planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
PREREQUISITES:
? IBM Personal Computer Disk Operating System (DOS) Version 2.1
or later
? FORTRAN 2, Professional FORTRAN, C, or BASIC Compilers
? IBM Personal Computer, IBM Personal Computer XT, IBM Portable
Personal Computer or IBM Personal Computer AT with at least
256Kb of memory
? IBM Graphics Display, IBM Professional Graphics Display, or
equivalent, with appropriate adapter
PACKAGING: The IBM Personal Computer Graphical File System is
packaged in three dual-sided diskettes. Device drivers and
associated modules are included in two additional diskettes.
The IBM Personal Computer Graphical File System Programmer's/User's
Guide is included.
qualifying contracts.
CHARGES, TERMS, AND CONDITIONS
Part Feature One-Time
Description Number Number Charge
IBM Personal Computer
Graphical File System 6024205 4205 $175
The IBM Personal Computer Plotting System
OVERVIEW HIGHLIGHTS
? Availability of area, bar, line, pie, scatter, schedule, step,
and text-only chart types
? Allows graphic and text annotation
? Supplies graphic input functions
? Permits linear and logarithmic axes
? Allows for a combination of chart types on a single chart
? Allows changing from one chart type to another chart type with
a single command
? Includes language bindings for FORTRAN 2.0,
Professional FORTRAN, C, and BASIC Compilers
? Allows device independence through the use of Virtual Device
Interface
DESCRIPTION
The IBM Personal Computer Plotting System facilitates custom
charting and chart applications, and provides the capability to use
a large variety of I/O devices. When the Virtual Device Interface
(VDI), which is described in today's IBM Personal Computer Graphics
Development ToolKit announcement, is used with the Plotting System,
programs will be device independent.
CUSTINFO SCHEDULE
Planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
SPECIFIED OPERATING ENVIRONMENT
HARDWARE REQUIREMENTS: An IBM Personal Computer, IBM Personal
Computer XT, IBM Portable Personal Computer, or
IBM Personal Computer AT with:
? Minimum configuration
- 256Kb memory
- Two dual-sided diskette drives
- Graphics display
? Recommended configuration enhancements
- 384Kb memory
- IBM 8087 or IBM 80287 Math Co-Processor, as appropriate
- IBM fixed disk and one dual-sided diskette drive
- Graphics hardcopy device
- Graphics input device
SOFTWARE REQUIREMENTS:
? DOS 2.1, or later version
? IBM Personal Computer FORTRAN Version 2.0, IBM Personal Computer
Professional FORTRAN, C, or BASIC Compiler
PACKAGING: The IBM Personal Computer Plotting System is packaged on
three dual-sided diskettes. Device drivers and associated modules
are included on two additional diskettes. Also included are the IBM
Personal Computer Plotting System Programmer's Guide and the IBM
Personal Computer Plotting System Language Bindings.
CHARGES
Part Feature One-Time
Description Number Number Charge
IBM Personal Computer
Plotting System 6024204 4204 $225
IBM Personal Computer Graphics Terminal Emulator
OVERVIEW HIGHLIGHTS
? Permits use of existing graphics applications supporting the
following terminals:
- Tektronix 4010 and 4100 protocols
- Lear Siegler ADM3A
? Provides an easy to use, icon-driven user interface
? Compatible modes
- ALPHA mode
- VECTOR mode
- POINT PLOT mode
- Graphics input mode
? Allows output to the following: display, printer, plotter,
Graphical File System
? Allows uploading and downloading of files to host systems
supporting ASCII protocol
? Permits user-definable asynchronous host communication protocol
file
? Executes local Personal Computer operating system commands
without leaving emulator mode
DESCRIPTION
The Graphics Terminal Emulator uses the Virtual Device
Interface (VDI), which is described in today's announcement of the
IBM Personal Computer Graphics Development ToolKit, providing the
user with access to a variety of graphics I/O devices. This makes
application output available on a wide range of output devices.
In addition to the capabilities derived from Graphics Terminal
Emulator software, the user is able to return to, and use, standard
IBM Personal Computer functions.
CUSTINFO SCHEDULE
The planned availability date is October 1984.
TECHINFO TECHNICAL INFORMATION
SPECIFIED OPERATING ENVIRONMENT
HARDWARE REQUIREMENTS: An IBM Personal Computer, IBM Personal
Computer XT, IBM Personal Computer AT, or IBM Portable Personal
Computer with at least 256Kb of memory.
? Minimum Configuration:
- Two dual-sided diskette drives
- Serial async port
- Graphics display with Virtual Device Interface support
? Recommended Configuration Enhancements:
- Fixed disk
- Locator input device (such as a mouse or joystick)
- Professional graphics display with Virtual Device Interface
support
SOFTWARE REQUIREMENTS: DOS 2.1 or later.
INSTALLABILITY: The customer, using the instructions in the IBM
Personal Computer Graphics Terminal Emulator User's Guide, will
install this product in conjunction with any necessary IBM Virtual
Device Interface (VDI) support and IBM Personal Computer DOS.
PACKAGING: This product is shipped on one dual-sided diskette with
the device drivers and associated modules included on the two
additional diskettes. The IBM Personal Computer Graphics Terminal
Emulator User's Guide is included.
CHARGES
Part Feature One-Time
Description Number Number Charge
IBM Personal Computer
Graphics Terminal
Emulator 6024206 4206 $295
-------
∂21-Sep-84 0209 INFO-IBMPC@USC-ISIB.ARPA INFO-IBMPC Digest V3 #98
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 21 Sep 84 02:07:00 PDT
Date: 21 Sep 1984 01:24:31 PDT
Subject: INFO-IBMPC Digest V3 #98
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Friday, 21 September 1984 Volume 3 : Issue 98
This Week's Editor: Billy Brackenridge
Today's Topics:
1978 Version of the 8088
The HP110
dBASE II or III for Networks
Case Mapping Patch for Microsoft Assembler
Control U
Looking for VI or NROFF clones
Displaywrite 2 Printing
Lattice C Exit Call Problem
Flight Simulator Update
Attribute Bits in PC-DOS 2.0
Sorting PC Directories
Databases for IBM-PC and Mini/Mainframes for Medical Research
Solution to PC TALK Problem
Public Domain Programs
Floating Point Performance on PC/AT
Laserjet and Graphics
Example of Using the EXEC Function from Assembly Language
IBM PC Expansion Chassis
IBM PC Pascal Advice
Reading Command Line in Turbo Pascal
Foreign Language Character Sets
Dbase II/III Dates
Everex Display Controller
----------------------------------------------------------------------
Date: 17 Sep 1984 15:54:55 PDT
Subject: 1978 Version of the 8088
From: Richard Gillmann <GILLMANN@USC-ISIB.ARPA>
There is an article on page 4 of the Sept. 4, 1984 issue of PC Week
about a problem with Framework. The article says that in programming
it they used a "shortcut" method of suspending system interrupts that
was written about in an article in Byte Magazine. Unfortunately, this
method does not work right on the 1978 version of the Intel 8088, only
on 8088s manufactured at a later date. The old chips have '78 printed
on them, the new ones have '78 '81 printed on them.
What is this "shortcut" method? How can anything be a "shortcut"
compared to using the CLI/STI instructions? I tried a little test
case on a '78 8088 and CLI/STI seems to work. Does anyone know which
Byte Magazine article they're referring to? Does this bug affect
other software? Does anybody have a test program that will reveal
if it's running on a '78 8088?
Dick Gillmann
-------
Date: 15 Sep 1984 22:37:56 PDT
Subject: The HP110
From: Randy Cole <COLE@USC-ISIB.ARPA>
The HP110 ("The Portable") has been available for a few months and I
recently purchased one. One nice thing about it is something that HP
hadn't said too much about until recently -- the way it works with
an IBM PC (or compatible). So I thought I'd relay my experiences
for Info-IBMPC readers.
For $150 (list) HP sells a board for the IBM PC which is in fact an
HP-IL (not to be confused with HP-IB) interface, along with a 5 1/4"
diskette of software for the PC. The software allows the 110 to
access the PC's disks and printer as if they were directly connected to the
110 itself. My PC's hard disk becomes Drive E on the 110, for example,
and I can copy to and from it and execute programs from it just as if
it were on the 110. Access times are somewhat longer, though, since the
HP-IL is a simple serial interface.
The printer on the PC acts just as if it were on the 110, with no problems
I know of so far. You have to tell the program on the PC whether you
are setting it up to access the printer or the disks (you can also print
to the PC's display), but you have to switch between the two -- you can't
spool to the printer and use the disks at the same time. You can easily
switch back and forth, though. The program does take over the PC -- you
can't do anything else on the PC while it is running.
Thus you can use the 110 strictly as a portable, and not have to buy
disks or a printer for it, and still be able to upload and download
text and programs quickly and conveniently. For someone with a PC in
the office who wants a machine at home, it would probably be a better
choice than a PCjr, for example. I think it would be especially nice for a
person who uses 123 a lot.
The interface board is a short board, and I believe it should fit in the
short slot of an XT.
Some miscellaneous comments about the 110:
1. HP must think their business is going to mushroom, because they keep
their customers and dealers in the dark and feed them BS. It took me 4
days to find out whether a Technical Reference manual was available (it
isn't yet, but something called a Programmer's Tool Kit is, but nobody
is sure what all it has in it, because HP dealers don't stock much
software they're not going to sell that day). For this I had to call
800-HPCOACH (cute, huh) and wait 4 days till they called me back... none
of the dealers knew or was told much except the name and stock #.
HP has an awful lot to learn about marketing computers and supporting
them after they're sold. It may not be too hard to get answers to
questions about business-type software, but HP apparently didn't intend
for anyone else to actually (yuck) PROGRAM one, at least not till they
get their manuals done.
This situation may be temporary.... I've seen the Tech Ref manual
for the HP150 and it is very good. It has hardware timing diagrams, complete
schematics, how to use DOS and the BIOS (called AGIOS) correctly, etc.
2. The built-in editor (called Memo-Maker) is aptly titled, since that's about
all it is good for. It is a Word!@#$ clone. I hacked up Mince and that's what
I am using.
3. Anyone who knows me knows that eyesight isn't my long suit, and I often
find the display hard to see.
4. It simply isn't a hacker's machine. The internal RAM disk is too small
for several passes of a compiler and a linker. Maybe if they took out 123
and their editor and replaced them with RAM.....
On the positive side:
1. Well-behaved IBM PC software (that which doesn't call the BIOS
directly) almost always works fine. The linker doesn't seem to work,
though. DEBUG works, and it DOES use BIOS calls.
2. The terminal emulator is well done, and the built-in modem has autodial
and autoanswer capability.
3. Battery consumption is as specified -- 16 to 20 hours running time
per charge.
4. Since it has a real 8086 and uses RAM disk, it is faster than a PC.
It runs the infamous sieve in 7.7 sec, compared to 11.7 sec on my PC
with an electronic disk.
Is there anyone else out there with one? Let me know and we can trade
stories, and set up a special interest group under Info-IBMPC if there
is enough interest. Anyone know where I can get a (maybe preliminary)
tech ref manual without buying the $295 Programmer's Tool Kit (I finally
found out: it contains an assembler, linker, debugger, all which I don't
really need, since I have the PC, and some documentation).
Randy Cole (Cole@ISIB)
-------
Date: Fri 14 Sep 84 16:08:52-PDT
From: ALFIERI@ECLB
Subject: dBASE II or III for Networks
Does anyone know if Ashton-Tate did finally release
a networking version of dBASE II, and, if so, how
it works?
Thanks in advance.
--vince alfieri
computing information services
usc
alfieri@usc-eclb
-------
Date: 12 Sep 84 9:01:19-PDT (Wed)
From: decvax!ittvax!flandren @ Ucb-Vax.arpa
Subject: Need Case Mapping Patch for Microsoft Assembler
About a year and a half ago someone posted
a patch procedure that circumvented the
"feature" of the microsoft pc assembler
that mapped all symbol names to lower case.
i would appreciate it if someone could supply
me with a copy of the patch. thanks.
address: ...!decvax!ittvax!flandren
-------
Date: 13 Sep 84 9:08:44-PDT (Thu)
From: hplabs!hao!seismo!uwvax!hagens @ Ucb-Vax.arpa
Subject: Control U
Does anyone have a way to patch bios keyboard handler so
control U has the same function as the escape key?
--
Rob Hagens @ wisconsin
...!{allegra,heurikon,ihnp4,seismo,sfwin,ucbvax,uwm-evax}!uwvax!hagens
hagens@wisc-rsch.arpa
-------
Date: Mon 17 Sep 84 16:03:56-CDT
From: John Otken <CC.Otken@UTEXAS-20.ARPA>
Subject: Looking for VI or NROFF clones
Could anyone tell me of clones for the Unix programs VI or NROFF?
I am interested in versions for CP/M-80 and PC/MS-DOS. John Otken.
P.S. OK. How about SCRIBE clones? Yes, I know about scribble.
-------
Date: 18 Sep 84 10:20:38 PDT (Tuesday)
From: Bicer.ES@XEROX.ARPA
Subject: Re: Looking for VI or NROFF clones
To: John Otken <CC.Otken@UTEXAS-20.ARPA>
There is a program called ROFF, available from BDS-C user's group in
Kansas. If I recall correctly, it was designed to run on dot matrix
printers and comes ready for EPSON printers (80, 100, ?).
Jack Bicer
-------
Date: 17 Sep 1984 18:23:20 PDT
Subject: Displaywrite 2 Printing
From: Dwight Baker <SWG.DBAKER@USC-ISIB.ARPA>
To: info-ibmpc@USC-ISIB.ARPA
cc: alfieri@USC-ECLB.ARPA
A while back a problem was reported using a Diablo printer
with DisplayWrite 2. I requested help from IBM using an
on line bulletin board with our corporate office, but no
solution was offered. I just saw an ad in the OCT 1, 84
issue of Infoworld that will provide a commercial solution
to the problem. The ad was on page 70 and offered drivers
for the Diablo, HP Laser Jet, and others for $149. I have
no connection with the company nor have I tried the product
but offer their name and address only as information.
Dw2's Em-U-Print $149. avail. from:
Koch Industries, PO BOX 812, Northbrook, Il 60062
312-228-0590
-------
Date: Mon Sep 17 1984 17:07:05
From: Marco Papa <papa%usc-cse.csnet@csnet-relay.arpa>
Subject: Lattice C Exit Call Problem
An easy way to get the error message:
Memory allocation error
Cannot load COMMAND, system halted
with Lattice C (versions 2.10 or 2.12, ANY memory model), is to "malloc" a
buffer (or allocate it with any one of the other memory functions) and then
write some data AFTER the end of the buffer. This will crash the system
with all memory models, since after release 2.10 of Lattice C the memory
allocation functions perform a call to the DOS Allocate function.
Marco Papa
USC - Computer Science Dept.
-------
Date: Mon Sep 17 1984 17:16:39
From: Marco Papa <papa%usc-cse.csnet@csnet-relay.arpa>
To: info-ibmpc@usc-isib.arpa
Subject: Flight Simulator Update
The update policy of Microsoft for its flight simulator program is just
terrible. The list price for the new version is $49.95. You can also get
it by sending your old version with the manual + $30.00 (for a total of
about $32.00).
The mail order price for the new one varies from $30 to $35 (see the ads in
PCWorld, for example). If you buy it this way you end up with two versions,
one of which you can legally sell or give as a present to a friend.
Microsoft, how can you call it an "update".
Marco Papa
USC - Computer Science Dept.
-------
Date: 13 Sep 84 11:19:00-PDT (Thu)
From: pur-ee!uiucdcsb!grogers @ Ucb-Vax.arpa
Subject: Attribute Bits in PC-DOS 2.0
You can set your prompt so that it will change the colors on the screen. All
the information is in the DOS manual but you must put several things together
to get this to work. First you must use the ANSI.SYS screen driver so in your
CONFIG.SYS file you need 'device=ansi.sys' (or some such thing). You then
can send escape sequences to the screen driver to change the color of the
characters. You can send these to your screen via your prompt. If I had a
manual handy I would give you an example but I don't. You should look at
the sections on the ANSI.SYS driver and setting your prompt.
-------
Date: 13 Sep 84 12:44:52-PDT (Thu)
From: cornell!vax135!ariel!hou5f!hou5g!hou5h!hou5a!hou5d!hogpc!houxt!ijk @ Uw-Beaver
Subject: Sorting PC Directories
Is there any way of sorting directories (i.e, equivalent to say dir | sort)
but from a low - level using DOS or ROM BIOS calls. Note that I
don't actually want to restructure the directory table, just print
it out in a sorted fashion. Nor do I actually want to start up
a complicated sort routine.
Thanx in advance.
Ihor Kinal
houxt!ijk
-------
Date: 12 Sep 84 14:10:30-PDT (Wed)
From: ucsfcgl!harrison @ Ucb-Vax.arpa
Subject: Databases for IBM-PC and Mini/Mainframes for Medical Research
We are doing a large scale research project on AIDS. We are currently
planning on doing data capture on IBM PC's, consolidating the data on a
mini (currently a PDP 11/70 running IAS operating system), then
selecting subsets of fields and records to be downloaded back to the
PC's for analysis. Some of that data will be shipped back up to an IBM
mainframe for analysis with SAS.
Are there any users out there
(a) doing AIDS research
(b) using MEDLOG
(c) with further suggestions
Peter Harrison
(415) 921-5060 or (415) 666-2951
ucbvax!ucsfmis!harrison
ucsfmis!harrison@berkeley.arpa
or whatever path got this message to you.
-------
Date: 15 Sep 84 11:30:24-PDT (Sat)
From: hplabs!sdcrdcf!trwrba!escher!noao!jacoby @ Ucb-Vax.arpa
Subject: Solution to PC TALK Problem
Thanks to those who replied to my query for help with
PC-TALK. The problem, in brief, is that the BASIC
version works fine, but the compiled version did not
perform file transfers properly. This is a problem
with the newer Hayes 1200B modem
The solution is to move the DTR jumper from the
default configuration across A-B to position B-C.
Why this would work in BASIC is still a mystery to me,
but may have to do with timing. In any case, if it works,
don't fix it.
George
(ihnp4!noao!jacoby)
-------
Date: 18 Sep 1984 09:32:26-EDT
From: Peng.Si.Ow@CMU-RI-ISL1
Subject: Public Domain Programs
I am interested in acquiring copies of public domain programs. My most
urgent needs currently are for a PROLOG and a LISP (I have heard that
there is something called Xlisp available.). I would greatly appreciate
any help that I can get - just acquired my first PC. Thank you.
pso@cmu-ri-isl1
-------
Date: Tue, 18 Sep 84 17:00:51 edt
From: Paul Crumley <pgc%CMU-ITC-LINUS@CMU-CS-PT.ARPA>
To: BRACKENRIDGE@USC-ISIB.ARPA
Subject: Floating Point Performance on PC/AT
I have heard that some people have been plugging 16 MHz crystals into their
PC/ATs and are running them successfully. The 16 MHz clock is within the
80287's specs with CKM tied low. You may want to pursue this as it would give
you an effective clock of 5.33 MHz, a 12% improvement over the PC/XT and a
33.33% improvement over the stock PC/AT.
IBM says nothing about higher clock frequencies so I wouldn't trust
this to work with all PC/ATs. Happy experimenting.
Paul
-------
Date: 18 September 1984 23:51-EDT
From: Richard P. Wilkes <RICK @ MIT-MC>
Subject: Laserjet and Graphics
Anyone Out there been marginally successful trying to get some graphics
(even reasonable line-art -- like PC line drawing set) working with the
LaserJet? This really is a nice printer, but HP is having some {←REA{
problems getting additional fonts ({ther than Courier 10) for the
machine. It would really shine with a real proportional space font and
graphics. I am just beginning to wonder whether HP will EVER make these
available. After talking with the HP marketing rep who told me "the 12
pitch font is stuck in manufacturing -- technical problems. The Times
Roman font (which I want) hasn't had too much work done on it."
Doesn't sound good...
-------
From: goldberg%viking.DEC@decwrl.ARPA
To: info-ibmpc@usc-isib.ARPA
Subject: Example of Using the EXEC Function from Assembly Language
[This program can be found in the INFO-IBMPC program library as EXECTEST.ASM.
There is also a somewhat more complicated version of a Lattice C callable
EXEC function in the file EXEC.ASM -ed]
(Marshall R. Goldberg LJ02/E4 DTN 282-2325)
-------
Date: Wed, 19 Sep 1984 20:47 PDT
From: HALVORSEN@SU-CSLI.ARPA
To: info-ibmpc@usc-isib.arpa
Subject: IBM PC Expansion Chassis
Is it possible to buy an IBM PC expansion chassis without
having to get a disk drive with it. If so, how much should
I expect to pay?
Thanks
Per-Kristian Halvorsen
-------
Date: 13 Sep 84 14:05:00-PDT (Thu)
From: pur-ee!uiucdcs!seaburg @ Ucb-Vax.arpa
Subject: IBM PC Pascal Advice
Thanks for all the replies to my note. Many people suggested Turbo
Pascal, but since it can't link to subroutine libraries, we can't do
much with it. I've played with it before, though, and it is fun!
Others have suggested Microsoft Pascal, and we'll probably be going to
that. It can link to Microsoft Fortran programs (correct?), and this
is an advantage -- my boss likes Fortran. We're probably going to get
both Pascal and Fortran, so we'll be able to port
lots of our PDP-11 Fortran code over to PC's via the Microsoft
Fortran, and still be able to write new code in Pascal.
Any other suggestions? Thanks for all the help.
-------
Date: 13 Sep 84 12:29:36-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!ecsvax!unbent @ Ucb-Vax.arpa
Subject: Reading Command Line in Turbo Pascal
Can anyone tell me how to write a Turbo Pascal routine which will
return the balance of a command line (i.e., the parameter list) in a string
that I can then fool around with?
According to my MS-DOS programmers' reference, when DOS loads a COM
file (which is what Turbo produces), it builds a Program Segment Prefix at
offset 0. At offset 81H, there should be an unformatted parameter area
"containing all the characters typed after the command (including leading
and imbedded delimiters) with the byte at 80H set to the number of
characters". Also, all four segment registers contain the segment address
of the initial allocation block that starts with the PSP control block. SO
one would suppose that something like this would work:
program EchoCmdParams;
var
LinetoEcho: string[127];
I,HowMany,Value,Addr: integer;
begin
LinetoEcho := '';
HowMany := Mem[CSeg:$80];
for I := 1 to HowMany do
begin
Addr := $80 + I;
Value := Mem[CSeg:Addr];
LinetoEcho := LinetoEcho + Char(Value);
end;
Writeln(LinetoEcho);
end.
It doesn't. It doesn't work when I use DSeg instead of CSeg either.
What am I overlooking or doing wrong, please?
-- Jay Rosenberg <unbent@ecsvax>
-------
Date: 20 Sep 1984 13:30:58 PDT
Sender: BRACKENRIDGE@USC-ISIB.ARPA
Subject: Foreign Language Character Sets
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: decvax!mcnc!duke!phs!lisa@UCB-VAX.ARPA
You might try contacting Adriana Karin Vernon at the Kelar Corporation
5456 McConnell Ave. Los Angeles, CA 90066 (213)827-1200.
Kelar is US distributor for an Israeli company that makes word processors,
for Hebrew, Arabic, Farsi, Urdu and who knows what else. These programs
run on the IBM-PC and I understand can be customized to run other character
sets.
One of the features they are particularly proud of is the ability to mix
English with these languages. As the cursor must go in opposite directions
when writing Hebrew and English this is no mean feat.
They also support daisy wheel printers for these languages so you can get high
quality output. Some models have two print wheels to support the intermixed
English and foreign language. I have never seen one running, but I am told
it is quite a sight!
-------
Date: 13 Sep 84 1:21:23-PDT (Thu)
From: hplabs!hpda!fortune!and!dual!zehntel!zinfandel!bobk @ Ucb-Vax.arpa
Subject: Dbase II/III Dates
Certainly the addition of date manipulation capabilities is a welcome,
and necessary enhancements, however it still has several shortcomings.
Dates can only be indexed in chronological order. Common
business applications frequently like to see the MOST RECENT record first.
The manual seems to imply that meaningful (chronological) index keys can be
constructed from concatenated date fields and character fields. To do this
requires "unpacking" the julian date into a "yy/mm/dd" string, a function
they neglected to provide.
Dates aside, one other feature that I wish for is the ability
to carry forward selected fields (typically dates) when APPENDing records.
-------
Date: Thu 20 Sep 84 22:23:26-PDT
From: Carl Fussell <G.FUSSELL@SU-SCORE.ARPA>
Subject: Everex Display Controller
Address: Santa Clara University
Does anyone on the list have any information (or better yet, experience)
with a display controller by Everex? I am told it handles color or
monochrome, graphics, and has 1 serial and 1 parallel port on it.
I was wondering how well if works and how compatible/incompatible it
is with IBM's controllers.
thanx for any info...
Carl
-------
------------------------------
End of Info-IBMPC Digest
************************
-------
-------
∂28-Sep-84 0357 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #99
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 28 Sep 84 03:57:16 PDT
Date: 27 Sep 1984 19:45:26 PDT
Subject: Info-IBMPC Digest V3 #99
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Thursday, 27 September 1984 Volume 3 : Issue 99
This Week's Editor: Eliot Moore
Today's Topics:
MEDLOG
MSDOS 2.0 clock problems
Everex Display Controller
New PC Technical and Other Manuals
ZENITH 150 compatibility query
Bug in '78 8088
Turbo Pascal
Version Control System query
Basic 2.1 document problem
Linkers for MSDOS
Screen dump for Hercules card
Moby Resident Code
"Optimum" DBMS
Lotus on PCjr
AT hints
Multi-user Xenix on AT
ANSI.SYS and DOS 2.0
Computers in Psychiatry and Psychology
8087/80287 detection
HP LaserJet printer
hard disk query
ANSI.SYS and CLEOL
----------------------------------------------------------------------
Date: Mon 24 Sep 84 19:39:17-PDT
From: Gio <Wiederhold@SRI-AI.ARPA>
Subject: MEDLOG
To: info-ibmpc@USC-ISIB.ARPA
cc: ucsfcgl!harrison@UCB-VAX.ARPA, wiederhold@SUMEX-AIM.ARPA
We found MEDLOG a valid substitute for mainframe clinical databases.
A paper comparing some experiences is in the 1984 Computers and Cardiology
conference, and I can mail a reprint on request. Warning: I have had
involvement with both the mainframe and the MEDLOG derivative, so beware.
Gio
------------------------------
Date: Wed, 26-Sep-84 19:58:57 PDT
From: Lauren Weinstein <vortex!lauren@RAND-UNIX.ARPA>
Subject: MSDOS 2.0 clock problems
To: INFO-IBMPC@ISIB
I've lately noticed that my XT system, when running DOS 2.0, sometimes
fails to advance to the next day when passing 00:00 hours. It is
rather sporadic and I've been unable to reliably duplicate the problem
in test situations. Does anyone have any comments or suggestions
concerning this minor mystery? Thanks much.
--Lauren--
------------------------------
Date: 25 Sep 1984 01:57-EDT
Subject: Everex Display Controller
From: JEWETT@USC-ISI.ARPA
To: g.fussell@SU-SCORE.ARPA
Cc: info-ibmpc@USC-ISIB.ARPA, jewett@USC-ISI.ARPA
I recently purchased a PC, and selected the Everex display
controller based on versatility. Although I only got a
monochrome monitor, I wanted to be able to run software which
uses the color adapter. The Everex trade mark for this board is
"The Graphics Edge"--it's advertised in PC World.
I'm a long way from having used all the features of this board,
but here are some initial comments:
- The color-to-monochrome conversion is really useful, but I
had to tweak the monitor (a Taxan) to keep it in vertical sync
when changing between modes.
- It's advertised to be compatible with Hercules graphics. In
fact, the Hercules driver for Framework does work--but I find the
IBM character set easier to read. Makes a big difference on pie
charts, though.
- It's got a 132 column mode which is supposed to work with
Lotus. I can't verify this, but Framework doesn't support it
yet.
- As usual for most products of this type, the documentation
that comes with it is lousy.
I hope that we'll see some tips from users more experienced with
this board on how to get the most out of it.
Tom Jewett
------------------------------
From: tekadg!jm%tektronix.csnet@csnet-relay.arpa
To: tektronix!Info-IBMPC@usc-isib.arpa
Date: Wednesday, 19 Sep 84 17:13:05 PDT
Subject: New PC Technical and Other Manuals
I would be interested in hearing from anyone who has found a
book or other publication that has helped them in the designing
of hardware add-ons for the PC. Someone told me that Norton
(of the Norton Utilities) has a book out, but my bookstores haven't
heard of it.
I would also appreciate finding out what the NEW PC technical manual
part numbers are. It seems that IBM decided to re-issue all their
hardware manuals, so they told their dealers to send back the old stuff.
They haven't released the new manuals, and all the information I can find
says they'll be available "in september". I get conflicting answers
when I ask what numbers to order, too. I have contacted dealers and
an IBM Products Store, but nobody knows what ALL of the various manual
numbers are. I have also gotten two different numbers for the same
manual.
Thanks.
Jeff Mizener
Tektronix Portables ADG
{ucbvax,decvax,purdue,uw-beaver,hplabs,allegra,ihnp4}!tektronix!tekadg!jm
tekadg!jm@tek
------------------------------
Date: 21 Sep 1984 19:38-EDT
Subject: ZENITH 150 compatibility query
From: DEVCTR@USC-ISI.ARPA
To: INFO-IBMPC@USC-ISIB.ARPA
Does anyone know exactly how compatible the ZENITH z150 is with
the IBM-PC? So far, I haven't found anything that won't run on
the Z150 that has been designed for the IBM-PC. I also
understand that the FLIGHT SIMULATOR program runs better on the
Z150 than it does on the IBM-PC from reports I read in BYTE on
tests at the last computer fair in Calif.
Fred
------------------------------
Date: 21 Sep 1984 0948-pst
From: Robert A. Lerche <XA.W51%STANFORD.BITNET@Berkeley>
To: <info-ibmpc@usc-isib.ARPA>
Subject: Bug in '78 8088
An important note for anyone who changes stacks in the 8088:
An early version of the 8088 chip has a bug that can cause
memory clobbers. According to the Intel iAPX86/88 book,
interrupts are disabled for one instruction following a MOV
into a segment register. Thus, the safe way to switch stacks
is to do a MOV into the stack segment register followed
immediately by a MOV into the stack pointer register.
The bug is that in 8088 chips marked "copyright 1978" the
automatic interrupt disable DOES NOT OCCUR. Thus, if an
interrupt occurs between the two MOV instructions, the
interrupt routine will store the registers in the new stack
segment but with the old stack pointer. Potential disaster!
The fix, of course, is to do your own CLI before switching
stacks.
DOS 2.0 device drivers are entered enabled for interrupts, so
this exposure exists there. Since timer interrupts occur
18.2 times per second in the PC, there's a good chance of a
crash if you change stacks in your driver without disabling
interrupts yourself. Take the word of one who has been
burned.
(P. S. remember to re-enable interrupts after you're done, or
your clock may run slow.)
There was an article about this in an early issue of the PC Tech
Journal. In that article, they gave a test which involved using
the debugger to trace a "move into segment register" instruction
followed by an "INC AX" (if I recall correctly). If the "INC AX"
executes (you can tell by examining the registers afterward) then
you have a later-model 8088. If the "INC AX" does not execute,
the trap interrupt occurred immediately after the "MOV", indicating
you have a '78 8088.
This bug was hell until I read that PC Tech Journal article. I wrote
IBM a letter complaining and got a very nice phone call from someone
who apologized and asked if everything was OK now that I knew the
fix. He also said they would consider putting a note in the next
version of the DOS manual, since the section on device drivers makes
such a big point of saying you should allocate your own stack if you
do anything other than saving the registers on the stack your driver
is entered with.
------------------------------
Date: Sat, 22 Sep 84 10:57 PDT
From: "C Midas Myth%SAI"@LLL-MFE.ARPA
Subject: Turbo Pascal
To: info-ibmpc@usc-isib.arpa
This is a reply to Jay Rosenberg's 13 Sept request for a Turbo pascal
routine that returns the parameter string on the command line when the
program is executed.
The example Jay gave worked fine for me when I compiled it. I have a plain
vanilla PC-I (c. August 1982).
Bear in mind that if you compile and run the program in Turbo memory mode,
(without creating an executable ECHOLINE.COM file), the command line
parameter string is whatever followed "turbo" when turbo was executed from
DOS. If you don't put any parameters on the command line, following "turbo",
there's nothing to echo. When you compile echoline to a ECHOLINE.COM file,
ECHOLINE writes the parameter string to the screen, then scrolls it off the
top of the screen immediately, so you have to look fast to see it.
Here's an easier way to do the same thing using Turbo's ability to place a
declared variable at a specified memory location and the fact the structure
of the DOS parameter string matches that of Turbo's string type. (I bet it's
not a coincidence.)
var
comm←line : string[127] absolute cseg:$0080;
begin
writeln; (* otherwise the line written out gets pushed off the screen. *)
writeln (comm←line);
end.
or you may prefer to "hide" the implementation dependent feature inside a
function:
const
max←char = 127;
type
msg←type = string[max←char];
(************************************************************************)
function comm←line : msg←type;
var
line : msg←type absolute cseg:$0080;
begin
comm←line := line;
end;
(************************************************************************)
begin
writeln; (* otherwise the line is pushed off the top of the screen. *)
writeln (comm←line);
end.
Dave Baxter
Development Software Associates
3711 Jackdaw St.
San Diego, CA 92103
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 22 Sep 84 14:59-PDT
Date: 17 Sep 84 5:12:17-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!mcvax!tnocsda!jaap @ Ucb-Vax.arpa
Subject: Version Control System query
Does anyone have any information about version control management systems
(like SCCS) to be used on small systems like IBM-PC and PC-AT?
If they cannot be found for use on a PC, can anyone give me a clue to suppliers
of such systems for bigger systems like a micro VAX or the like?.
I am looking forward to any replies.
Thanks in advance,
Michel G. van Hamburg
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 22 Sep 84 22:55-PDT
Date: 19 Sep 84 5:29:51-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!ittvax!ittral!mccullou @ Ucb-Vax.arpa
Subject: Basic 2.1 document problem
Way back in February of this year, I purchased DOS 2.1 from an authorized
IBM distributor. Unfortunately, there was nothing to upgrade my BASIC manual.
There was however a form to be filled out promising such an upgrade package.
I filled out the form, followed all the instructions, and mail everything
to the address indicated. To make a short story long, the form said, "Please
wait 4 to 6 weeks for delivery." I have waited 7 months and seen nothing!
I tried all the IBM 800 phone numbers I could find - it wasn't their area.
I tried the information operator in Wallingford CT - IBM has too many offices
in that area and long distances charges would soon outprice the original cost
of the DOS package!! I tried mailing a nasty-gram to the same PO box - no
response. I'm now working with the manager of the store that sold me the DOS
package but his IBM representative is not helping very much. My last resort
will be to demand a refund from the store. That will be futile because I want
DOS version 2.1.
My questions are:
- Am I just lucky, or have others had this problem?
- The form said the package (I hope to get eventually) will upgrade BASIC 1.1
to BASIC 2.0. Does anyone know of differences between BASIC 2.1 and 2.0?
I know the file sizes are different. I bought BASIC 2.1 and I deserve the
documentation for it, not documentation for 2.0!!
- Does anyone know what is happening at PO Box 3160, Wallingford, CT 06494
besides a big letter vacuum that sucks up request forms never to be hear
from again? I understand the same thing is being done at the same address
for the new FORTRAN compiler.
Please respond directly with help or suggestions.
Cliff McCullough
ittvax!ittral!mccullou
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 23 Sep 84 13:58-PDT
Date: 14 Sep 84 11:52:20-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hpda!fortune!wdl1!jbn @ Ucb-Vax.arpa
Subject: Linkers for MSDOS
For large C programming on the IBM PC, the following seems to be
the best combination:
Lattice C
Plink, from Phoenix Software Associates
PolyMake, from Polytran.
If you are using the large memory model, the latest versions of Lattice and
Plink are essential. Early versions of the overlay loader in Plink were very
buggy. Programs in excess of 60,000 lines of C have been built with the above
combination.
I have no recommendations on debuggers; I don't find them necessary.
Phoenix sells one which understands their overlay structure, but I have not
seen it in use.
John Nagle
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 23 Sep 84 21:02-PDT
Date: 19 Sep 84 6:06:48-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From:
ihnp4!houxm!mhuxj!mhuxn!mhuxl!ulysses!gamma!epsilon!mb2c!uofm-cv!ro062 @ Ucb-Vax.arpa
Subject: Screen dump for Hercules card?
Does anyone have a low cost graphics screen dump for the Hercules card and
an Okidata printer? Anything close?
Martin Thompson
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 26 Sep 84 4:59-PDT
Date: 20 Sep 84 8:59:48-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!uwvax!hagens @ Ucb-Vax.arpa
Subject: Moby Resident Code
I have a large piece of code that wish to remain resident in memory.
I plan to do this via the dos hex 31 function (terminate, but remain
resident).
I have other programs which need to run, and occasionally let
this resident piece of code run. My mechanism for this would be:
The user program does an int xxx (where xxx is some unused interrupt).
The handler for int xxx puts a fake interrupt frame on the stack such
that when the handler does a return from interrupt, control returns
to the resident piece of code.
The resident piece of code does its thing for a while, and then
takes the other interrupt frame (the one created by the original
int xxx) and does a return from interrupt, thus returning to the
user program.
Now, none of this has been coded yet. What I am asking is:
1. Has anyone tried this before?
2. Does anyone know of any restrictions that dos has on resident code?
(It doesn't look like it, from the documentation).
3. Any comments on this idea?
I realize that registers need to be saved/restored in the correct order.
Thanks in advance for trying to understand what I just said,
--
Rob Hagens @ wisconsin
...!{allegra,heurikon,ihnp4,seismo,sfwin,ucbvax,uwm-evax}!uwvax!hagens
hagens@wisc-rsch.arpa
------------------------------
Date: Wed, 26 Sep 84 15:39:26 EDT
From: Pierre duPont <pdupont@bbnj>
Subject: "Optimum" DBMS
To: info-ibmpc@isib
I am interested in getting some info & reactions to a database
package called Optimum from Uveon in Colorado. Has anyone had
good/bad experiences with this package? Are there serious
limitations?
I do NOT subscribe to info-ibmpc, so please respond directly
to me at bbn-unix. If there is interest and sufficient response,
I will summarize to info-ibmpc.
Thanks,
pdupont@bbn-unix
------------------------------
Date: Sat, 22 Sep 84 9:33:52 EDT
From: Dave Farber <farber%udel-eecis3.delaware@udel-relay.ARPA>
To: info-ibmpc@usc-isib.ARPA
Subject: Lotus on PCjr
My account got mussed up so I'm using my Dad's.
(manny@udel-eevax)
As soon as I got a 128K memory expander for my PCjr, I tried to bring
Lotus 1-2-3 up on it. Everything worked fine except for one thing:
I couldn't display any graphs; the screen displayed semi-gibberish
instead.
This was due to some hardware incompatibilities between the PC and the
PCjr's color adapters; namely, the ports are different.
So, I looked in the graphics display driver (use IBM1G2.DRV, on the
utility disk) and found the offending code, and replaced it with
code that would use the ROM-BIOS to change display modes.
To make the change (for Lotus 1-A), do the following:
DEBUG GD.DRV (GD.DRV should be a copy of IBM1G2.DRV)
A 1b7
MOV AX,4
INT 10
[return]
F 1bc 1c9 90 (this replaces the rest of the offensive code with NOP's)
W
Q
- Manny
If anyone has questions drop me a line.
p.s.
One could also use the extra code space to select the PCjr's palette.
Does anyone out there know any other PCjr tricks, like how to get
Bluebush Chess to work with the PCjr's display?
------------------------------
Date: 24 Sep 1984 2335-PDT
From: STERNLIGHT@USC-ECL.ARPA
Subject: AT hints
To: info-ibmpc@isib
Depending on when you get delivery you might try to get the new IBM
graphics board for the color monitor which is supposed to deliver in
November. Or you might see if the dealer will take back the existing
board in a trade when the new one comes out. Or you might borrow or
buy a used one. The new one, announced September 10, has higher
resolution & works with the old display and the new one coming in
January (also announced September 10).
Unless you plan to run Xenix or PC/IX, don't buy more than 640k of
memory. The hard disk is so fast you don't need a ramdisk, and PC-DOS
3.0 and 3.1 cannot address memory above 640k, so any more is useless
unless you plan to run Xenix. The 128k upgrade for the 512k hard disk
Model 99 (or is it Model 90; I forget) is about $340 and the 512k
upgrade is about $1000.
Make sure you get an 80287 math chip. The AT does not test correctly
for the absence of such a chip, so all programs that are supposed to
test for it and run math correctly if it is absent don't. But they
all work OK if it is present. I have in mind Fortran and C in
particular, compiled with the library that tests to see if you have
the 8087 on a PC family and if not, uses the 8088 for arithmetic.
If you never plan to write disks to be used on PC/PC-XT/PC Portable,
don't bother getting a second disk drive of the 320/360k variety. The
1.2 meg floppy drive reads such disks just fine, and writes on them
fine except that once you write them, IBM won't guarantee they are
readable on any machine except the AT.
Buy a couple of boxes of disks as soon as possible. They are in very
short supply and hard to find. Dysan has a new disk, the UHR II,
reorder #802914 that works perfectly at 1.2 meg on the AT. Otherwise
get the IBM disks for the AT (red label on disks). Nothing else works
at 1.2 meg no matter what the box says. In particular, the Capitol
disks claimed to work at any density don't.
Machines are now coming out of the pipeline nicely in Los Angeles.
But accessories are hard to get. Get a serial cable or serial adapter
as soon as possible. The AT has a 10 pin RS-232 DB-9 connector but
unlike the Macintosh it is a male socket on the machine. Nothing fits
except IBM or a custom-made cable. The serial adapter IBM sells is
just a very short serial cable with a proper RS-232 plug on the other
end that fits modems and serial printers. The serial cable just has a
longer cord. Get one of each; they're not that expensive and when you
need one, you need it badly.
I am using the Ven-Tel PC-Modem half-card and it works fine; it is
Hayes compatible and runs 1200 baud as well as 300. You can then use
the serial/parallel card that comes with the Model 99 for a second
modem or a serial printer as well as a parallel printer.
When you first start up, unless the technician at the shop has done
his job, you will have to set the date and time in the built-in
clock/calendar chip. It's confusing but the ONLY way to do it and
have it take in that chip, not just in the current session, is to use
the configuration program on your diagnostic disk. (Yes, I know;
that's a weird place for it, but what can I tell you?) You will need
to know what type of hard disk you have installed if you have a Model
99, since the configuration program asks that. You can either ask the
tech, take the cover off the machine and look at the upper right hand
corner of the hard disk to see, or look through the front slots with a
flashlight to try to make it out. Mine is a type 20.
My machine has been running happily for three weeks now on its side
next to my desk, without the floor stand, which I have on order. The
machine hasn't fallen over yet.
MAKE SURE YOU GET THE KEY WITH THE MACHINE. IF THE TECHNICIAN AT THE
PLACE YOU BUY IT CHECKS OUT THE MACHINE, HE MAY FORGET TO PUT THE KEY
BACK IN THE BOX.
Make sure that any cards you get (modem, etc.) that are not AT cards
but PC family are put in the two slots without the extra 16-bit bus
connectors. You want to keep the 16-bit bus slots available for AT
cards.
There is already a file on a few RBBS systems called DOS30FIX.dat
which contains a couple of patches to DOS 3.0 to eliminate a couple of
inconveniences. I can send a copy to you when you get DOS 3.0.
Good luck; --david--
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 26 Sep 84 4:09-PDT
Date: 17 Sep 84 16:40:30-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!cmcl2!philabs!aecom!werner @ Ucb-Vax.arpa
Subject: ANSI.SYS and DOS 2.0
I believe a have encountered a bug (or at least a very annoying
feature) in PC-DOS 2.0.
I was trying to convert my keyboard into the Dvorak layout using
the ANSI.SYS key assignment.
The problem is that there doesn't seem to be enough assignments to
cover the whole keyboard. I wrote separate files for the first row, second,
and third. Each works individually to convert their own row, however when
typed consecutively (with a batch file OR as a concatenated file) they cause
the keyboard to hang. This happens midway through the third file, and is
independent of the order in which the files are typed to the screen.
How do I get around this problem? Why does this problem exist (too
small a buffer, maybe?) And, though I still wish to know the answer to this,
is there a better way to remap the keyboard to Dvorak from QWERTY.
Replies welcome from all over. Mail me at the address below, and
especially, ARE YOU OUT THERE, MICROSOFT?
Craig Werner
philabs!aecom!werner
"What do you expect? Watermelons are out of season!"
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 26 Sep 84 21:14-PDT
Date: 11 Sep 84 13:10:23-PDT (Tue)
To: info-ibmpc @ Usc-Isib.arpa
From: ihnp4!cbosgd!mark @ Ucb-Vax.arpa
Subject: ANSI.SYS and MS-DOS 2.0
ANSI.SYS is certainly not IBM only - the AT&T 6300 comes with it.
(By the way, the AT&T 6300 seems like a winner - it's about twice
as fast as the IBM XT, costs about the same, and is so compatible
that Venix and Xenix both run unmodified on it at twice the speed
of the XT.*)
However, ANSI.SYS is far from a wonderful way to write your software.
It's not very complete (for example, it can't do ED, that's "clear
to bottom of screen" in English, and it ignores the parameters to
EL as well) and it's awfully slow. Most useful IBM PC software is
forced to write directly into display memory.
* I work for AT&T but if you know me you know I'm not biased in favor
of AT&T products.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 26 Sep 84 23:39-PDT
Date: 16 Sep 84 21:01:18-PDT (Sun)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!sdcrdcf!randvax!jim @ Ucb-Vax.arpa
Subject: Multi-user Xenix on AT
Although the blurb you saw might not have made an issue of Xenix running
multiple users on the AT, it will. I saw an AT at the Dallas announcement
meeting running Xenix, and it had two PCs running off of its serial ports
as terminals. 3 is the limit for now - I guess that's a limitation on
the number of serial ports. However, the Xenix comes with a user license
for 16 users (if you want to run that many on an 80286).
Jim Gillogly
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 27 Sep 84 3:49-PDT
Date: 16 Sep 84 17:22:08-PDT (Sun)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!ecsvax!uevans @ Ucb-Vax.arpa
Subject: Computers in Psychiatry and Psychology
I'm gathering information on the use of computers (any size or type) in
the mental health fields. In our Department of Psychiatry at the University
of North Carolina at Chapel Hill, we are organizing a computer laboratory
for use by our faculty, staff, and students. I would like to maintain
a notebook in that room with examples of computer use in other Departments
with similar interests... Psychiatry, Psychology, Social Work, etc.
If you are involved in computer use in an area related to one of those
above, I would appreciate a summary (as long or short as you have time
and inclination for) of your applications. If you would like, I will
be glad to return a composite list of all replies I receive. Replies
by either US Mail or electronic mail will be greatly appreciated.
Thanks in advance.
Elizabeth A. Evans
University of North Carolina
Department of Psychiatry 208H
Medical School Wing D
Chapel Hill, NC 27514
(919) 966-5277
uevans@ecsvax
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 27 Sep 84 3:55-PDT
Date: 13 Sep 84 7:23:41-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!akgua!gatech!gitpyr!roy @ Ucb-Vax.arpa
Subject: 8087/80287 detection
Well,
As far as I can tell, the only way to detect the presence/absence of
the coprocessor is to read the system board switch(es) and determine if
a coprocessor is configured or not.
So as a guess, I would say that the AT switches read differently. I
know that the XT reads in an entirely different fashion than the PC, so
there seems to be a precedent....
--
Roy J. Mongiovi. Office of Computing Services.
Georgia Institute of Technology, Atlanta GA 30332
Phone: (404) 894-6163, (404) 894-4660 [messages]
...!{akgua,allegra,and,hplabs,ihnp4,masscomp,ut-ngp}!gatech!gitpyr!roy
...!{rlgvax,sb1,uf-cgrl,unmvax,ut-sally}!gatech!gitpyr!roy
------------------------------
Date: Thu 27 Sep 84 15:48:27-PDT
From: Jackie <Burhans@ECLD.#ECLnet>
Subject: HP LaserJet printer
To: info-ibmpc@USC-ISI.ARPA
I went to see a demo of the HP LaserJet printer today with an eye toward
purchasing it for use with our IBM PC in our word-processing
department. This message constitutes a review of the
printer from that perspective:
We went out to a computer store in Century City.
This store was an HP authorized dealer and had only HP equipment.
Nonetheless, we asked to see a demo of the LaserJet. The salesman
turned on the canned demo and sure enough, as fast and as
quiet and as pretty as you could want, out came several pages of
text with mixed fonts, bold, italics...even some graphics.
However, when I started to ask questions about software compatibility
and paper handling, I started to get some answers that weren't too
heartening. The salesman allowed as how any piece of software sending
straight ascii characters would print fine on the LaserJet but that
any special features or escape-sequences that a program sent might
cause problems if the software couldn't be configured to send
the right things. He could not, however, give me any idea of what software
would support the Laser Jet.
The most important drawback of the LaserJet as a production word-processing
printer is its lack of paper-handling accessories. It only has one bin
to feed paper from and only accepts standard (8.5x11,8.5x14) paper. It has
no envelope feeder although the salesman assured me you could feed
standard size envelopes manually (I don't think my WP department is going
to be too impressed).
Summary: nice machine, fast, quiet, good quality but not perhaps quite
geared for a production word-processing environment.
------------------------------
Date: 26 Sep 1984 1655-PDT
From: LADKIN at SRI-CSL.ARPA
Subject: hard disk query
To: info-ibmpc@isib
cc: ladkin at SRI-CSL.ARPA
I'm about to buy two, preferably internal, hard disks for a couple of PCs
in the CSLab. There seem to be two kinds - the dealer-supported $2000+,
and the Qubie-type $1000 job. (in 10mb flavor). Are there any good
reasons not to buy Qubies? Can you direct me to further information?
Thanks,
Peter Ladkin
------------------------------
Date: Wed, 26 Sep 84 08:40:50 EDT
From: blue@NBS-SDC
Subject: ANSI.SYS and CLEOL
To: INFO-IBMPC@USC-ISIB
The DOS Technical Reference Manual says the sequence for erasing
the current line from the cursor to the end is ESC [k. The sequence that
works is ESC [K. The error is also in the COMPAQ documentation.
Jim Blue (blue @ nbs-sdc)
National Bureau of Standards
------------------------------
End of Info-IBMPC Digest
************************
-------
∂01-Oct-84 0136 INFO-IBMPC@USC-ISIB.ARPA Info-IBMPC Digest V3 #100
Received: from USC-ISIB.ARPA by SU-AI.ARPA with TCP; 1 Oct 84 01:35:43 PDT
Date: 1 Oct 1984 00:32:34 PDT
Subject: Info-IBMPC Digest V3 #100
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
To: Info-IBMPC: ;
Info-IBMPC Digest Monday, October 1, 1984 Volume 3 : Issue 100
This Week's Editor: Eliot Moore
Today's Topics:
MSDOS 2.0 clock problems
8087/80287 detection
AT Lattice C and large code
Xenix, UUCP on AT
AT hard disk type
Turbo Pascal (2 messages)
PCjr vs. Sanyo MBC555
Bugs in MS-DOS Kermit Bootstrapping Programs
File protection under DOS 2.0
Foreign languages on micros: a summary
Lattice C exec/fork bug?
VI or NROFF clones
Document Problem
Sidekick Compatibility Query
'True' BASIC
----------------------------------------------------------------------
From: <bang!crash!bblue@Nosc>
Date: Fri, 28 Sep 84 22:47:46 pdt
Subject: MSDOS 2.0 clock problems
To: info-ibmpc@usc-isib
Lauren - I have seen explanations for the date rollover problem in several
publications and have fixed it in a couple of ways. The best explanation I
have found for this appeared in the May 1 issue of PC Magazine in the PC
Tutor section, page 415. I'll paraphrase here:
...Apparently this is what happens. When the time rolls
over at 24:00, the timer interrupt routine sets an
overflow flag in memory at 0040:0070. When you call the
time-of-day routine (INT 1Ah) to read the time, this
routine resets the overflow flag to zero.
The time-of-day routine returns the overflow flag in
the AL register. When the DATE program calls on this
routine, DATE is smart enough to increment the date when
it finds the overflow flag on.
The disk-driver routines that are in the DOS (not in
the BIOS) also use the time-of-day clock. They might
use it to time out the disk drive appropriately or to
check for read errors. Unfortunately, these routines do
not change the date to the next date even though they are
told that the overflow flag is set. And once these
routines are run, the overflow flag is reset to zero.
If this happens before you ask for the date, the DATE
routine won't be told to change to the next day.
This certainly is a bug! And there's no simple way
to fix the software. The simplest answer might be to
create a new clock driver for PC-DOS 2.0 that does two
things:
First, the driver must substitute a new routine in
place of the time-of-day interrupt routine (INT 1Ah).
This new routine would not reset the overflow flag
unless the routine was called with an argument of 2.
(Presently, there are only arguments of 0 for setting
the flag and 1 for reading it.) Thus AH=2 would involve
reading the clock and also resetting the overflow flag.
Second, when the new clock driver is called for
reading the time or date, it will call the new
time-of-day routine with an argument of 2. Hence, the
overflow flag will be reset only when a program is
called that is able to change the date....
I have seen a public domain driver called CLOCKFIX.SYS (with an
associated .DOC file) that apparently does something similar. It seems
to work.
Hope this helps. --Bill Blue {ihnp4, sdcsvax!bang}!crash!bblue
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Sep 84 11:57-PDT
Date: 26 Sep 84 18:19:15-PDT (Wed)
To: info-ibmpc @ Usc-Isib.arpa
From: ihnp4!stolaf!umn-cs!digi-g!deg @ Ucb-Vax.arpa
Subject: 8087/80287 detection
It should be possible to determine if a coprocessor is resident by executing
an instruction that changes the state of something that the 8088 can access.
One instruction that would work is a store. The 8087 manual specifies
the maximum execution time for each 8087 instruction, so it should be easy
to compute a value for a timing loop. Since the initial contents of the
8087's registers are undefined, it probably would be a good idea to load
something into the register before it is stored:
FNINIT
<delay>
FNLD1
<delay>
FNSTP <address>
<delay>
; At this point, you can see whether the 8087 stored something into
; <address>.
Make sure that your assembler doesn't put a WAIT opcode in front of the 8087
instructions. There are probably shorter code sequences that would give
the same results.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Sep 84 16:11-PDT
Date: 28 Sep 84 10:51:19-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!rlgvax!geller @ Ucb-Vax.arpa
Subject: AT Lattice C and large code
To date there are no "C" compilers that support large model
coding on the IBM AT (Intel 286 based). IS THIS CORRECT? I
spoke with the author if the Lattice "C" compiler and he told
me of some design problems with the 286 that makes building
large model programs very difficult. I think there was some-
thing about a register changing function from the 8086 line.
I'd like to see some information on the net concerning large
model addressing - its problems, implementation details,
etc. Thanks.
David P. Geller.
rlgvax!geller
[this has nothing whatsoever to do with my employer, who, incidentally,
has been more than gracious enough to allow me the use of this infor-
mation proliferating medium. These views, comments, etc. are solely mine
and do not in any way reflect the policies or opinions of my employer.]
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 29 Sep 84 1:27-PDT
Date: 27 Sep 84 0:41:56-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!intelca!kds @ Ucb-Vax.arpa
Subject: Xenix, UUCP on AT
Ok people, it is nasty rumor dispel (?) time again. Many thanks
to the responses that totally obliterated the last nasty rumor:
the Xenix port that Microsoft is doing for the AT *does* run
in the 286s protected mode. Thanks to the respondents from
Microsoft and the Santa Cruz operation for lending a voice
of authority to all this, and also to the IBM product announcement
blurb. My confusion arose from the current (not 3 months hence)
availability of non-protected Xenix (the same as on the XT) for
the AT.
New nasty rumor: Xenix for the AT comes sans uucp. Apparently
because of configuration/support problems. I'm not sure whether
people here think this is a problem, what with all the potential
new uucp nodes this might create, but I certainly would think
twice about another software/hardware package if it at least wasn't
available somewhere. Anyway, here's hoping that it didn't get
"unbundled" never to return.....
--
I've got one, two, three, four, five senses working overtime,
trying to take this all in!
Ken Shoemaker, Intel, Santa Clara, Ca.
{pur-ee,hplabs,and,scgvaxd,dual,idi,omsvax}!intelca!kds
---the above views are personal. They may not represent those of Intel.
------------------------------
Date: 28 Sep 1984 18:05:36-PDT (Friday)
From: "Guillermo A. Loyola" <LOYOLA%ibm-sj.csnet@csnet-relay.arpa>
To: sternlight%usc-ecl.arpa@csnet-relay.arpa
Subject: AT hard disk type
CC: info-ibmpc%usc-isib.arpa@csnet-relay.arpa
The standard 20Mb hard disk on the AT is type 2 (two, NOT twenty). In
fact, I don't think 20 is a defined type (I don't have my Tech Ref
Manual here right now, so I'm not sure).
Guillermo.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 28 Sep 84 18:27-PDT
Date: 21 Sep 84 14:09:38-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!nsc!voder!gino @ Ucb-Vax.arpa
Subject: Turbo Pascal
My office mate, George Smith, has given me this copy of an
article in TUG Lines. I have run it and it works either in
memory or standalone (it even has a blank Writeln - see #373
referenced above).
BTW, `TUG Lines' is the newsletter of the Turbo Users Group.
If you're interested, mail me & I'll extort their address out of
George.
-------------------- cut here ------------------------------
program params;
(*
** Submitted to TUG Lines by Paul Klarreich, Brooklyn, NY
**
** This program illustrates a method of finding the parameters
** to an MS-DOS command file. In any MS-DOS program, the parameters
** are found at offset $80 in the program segment prefix -- the first
** $100 bytes of the program segment. The byte at $80 gives the
** length of the string and the next "length" bytes are the actual
** characters.
*)
var
CmdTail: string[80] absolute cseg:$0080;
begin
writeln;
writeln('The command tail for this program is:');
writeln('-->', CmdTail, '<--');
end.
--
Gene E. Bloch (...!nsc!voder!gino)
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 29 Sep 84 11:59-PDT
Date: 27 Sep 84 15:54:00-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!cmcl2!acf4!leung @ Ucb-Vax.arpa
Subject: Turbo Pascal
I've not been using my TURBO pascal package for a while, and one thing that
I have forgotten how to do is to use TURBO to print out the pascal files.
I know I could do it and get it to number the lines because I've got a
couple of printouts, but for the life of me I can't remember how...
My fault for not keeping a note book, I know! Replies gratefully accepted at:
Alex Polozoff New York University
leung@acf4 <inhp4, allegra>!cmcl2!acf4!leung
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Sep 84 18:49-PDT
Date: 29 Sep 84 10:50:59-PDT (Sat)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!harvard!wjh12!bbncca!sdyer @ Ucb-Vax.arpa
Subject: PCjr vs. Sanyo MBC555
>From your configuration (128K PCJr/Tecmar Jr. Captain/Multiplan) it
sounds like you are taking advantage of this month's Sears Business
Systems special. I did the same, though I also purchased the new Sears
RGB/Composite/TV monitor (it's great! you can switch to TV during
C compilations!) The Tecmar Jr. Captain gives you 128K more, a
clock/calendar chip and a parallel port. This is a VERY nice deal
for $1300.
I am using this right now, and overall am wonderfully pleased. As some
of the readers of net.micro might know, I went out and purchased a
Sanyo MBC555 a few months ago. It is only semi-IBM compatible; pure
MSDOS-only programs (compilers, assemblers, etc.) will work fine, but
most other software which accesses the BIOS will simply not work.
The documentation is atrocious, and not a single communications program
I could find would work on it. I still can't figure out how to do
graphics (or even access the screen) from the lame documentation.
Enter the PCJr. I was immediately able to purchase a good VT100
emulator/XMODEM program (Mark of the Unicorn's PC/Intercomm).
The IBM PC Kermit distribution works fine, too. Either makes a
fine remote terminal, and I have had NO problems with XMODEM
or KERMIT file transfers at 1200 baud; techies at first worried
about the non-DMA disk drive--it locks out interrupts during
transfers, meaning that many comm programs might have trouble.
Though I grant this, it hasn't been an operational problem.
Most other software I can find works fine on an expanded 256K PCJr.
Naturally, you won't be as sure about compatibility as on a PC,
but most problems arise from going beneath the BIOS to access the
graphics or disk IO ports (which aren't in the same place.)
Few programs do this. Naturally, if a program requires two
disk drives, that will be a problem, too. But it's surprising
what you can do with a RAM disk and one disk drive.
The new PCJr keyboard is a great improvement over the old. As an
exercise in masochism, I used the PCJr as a remote terminal before
I received the new keyboard. Auggh. Now, my typing speed is back to
normal. Both are wireless, a useless feature unless you are using
the PCJr with a projection TV. The feel of the new keyboard is
a bit mushy compared to most; many people will still object to this.
I am not quite as picky. I have noticed one problem with using
PC/Intercomm (though not Kermit) at 1200 baud. A nonmaskable
interrupt is generated upon receipt of the start bit of a character
generated by the keyboard; the BIOS then acts as a software UART,
polling for bit transitions to generate the full scan code. This
seems to cause occasional problems when I type ahead of the display:
PC/Intercomm seems to get out of sync, and generates several characters
worth of fully filled white space (VT100 DEL chars.) This may be
due to its handling of UART overruns; I can't say yet. I intend to
report it to MotU.
Anyway, I have found that the new keyboard and memory expansion make the
PCJr a much more desirable machine than it was before. It also enjoys much
greater software and hardware support than the Sanyo, an equivalently
priced machine whose only advantages are 8087 support and dual disk drives,
both of which will surely be provided for the PCJr by the aftermarket. I
am not intent on damning the Sanyo excessively; it is a nice machine, but
the hardware and software situation for it is simply nonexistent right now,
or very immature. And even the most ardent "roll-your-own" hacker who
isn't interested in buying much software will be hampered by the
intransigence of Sanyo in providing technical documentation on the hardware
and software and the limitations of the current BIOS. Having lived with a
non-mainstream machine for quite a while (a Radio Shack Color Computer) it
is refreshing to simply be able to buy what I need to get the job done.
The PCJr suffers when compared to the PC, and those who feel that its
limitations are unacceptable and who can afford another machine should
buy another machine. But I think that you get a lot for the $800/$900
which the PCJr is selling for these days.
--
/Steve Dyer
{decvax,linus,ima}!bbncca!sdyer
sdyer@bbncca.ARPA
------------------------------
Date: Fri 28 Sep 84 12:48:34-EDT
From: Frank da Cruz <SY.FDC%CU20B@COLUMBIA>
Subject: Bugs in MS-DOS Kermit Bootstrapping Programs
To: Info-Kermit%CU20B@COLUMBIA, Info-IBMPC@USC-ISIB.ARPA
The Fortran-Basic program pair that is used for bootstrapping MS-DOS Kermit
initially from a mainframe to a PC does not work in certain cases. There
were two small problems -- one caused the Fortran program to hang waiting for
input (in Fortrans that do not supply default values for missing variables on
a READ statement), the other an erroneous GOTO that had the unfortunate
quirk of not breaking the program on systems whose Fortran happened to
initialize arrays with blanks... Anyway, the fixed versions are available
in KER:MSBOOT.FOR and KER:MSPCBOOT.BAS on CU20B and COLUMBIA-20. If you'd
rather not get the new files, the changes are simple:
In MSBOOT.FOR, after the statement 200 FORMAT(4A1), change
GO TO 30
to
GO TO 20
In MSPCBOOT.BAS, change line 100 to:
100 print#1,"O ,2" ' Char constants Oh,Space,Comma,Two.
Thanks to Jeff Ramsey of DGM&S, Delran NJ, for pointing out these problems.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Sep 84 2:27-PDT
Date: 17 Sep 84 19:01:27-PDT (Mon)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!akgua!uf-csv!leo @ Ucb-Vax.arpa
Subject: File protection under DOS 2.0
I am looking for a utility under IBM DOS 2.0/2.1 that will allow me to write
protect files and/or directories. My department just bought four more P.C.'s
and is going to have them networked together with the Blossom ring, with
a hard disk on the hmfic machine, and I desparately need a way to keep the
commies(er, um, uh, USERS) from zapping the disk on a daily basis.
I think I know enough to remove the format program, but how does one keep
people from saying 'erase *.* <cr> OOPS!'? I really think this is a
necessary utility, and will be eternally grateful if anyone can point
me to a utility that does this, public or private.
I will post or mail the responses to this query, as the volume of interest
dictates.
Thanks in advance,
--
Leo Wilson
akgua!uf-csv(!uf-csg)!leo
CSNET: leo @ ufl
USPS: University of Florida
512 Weil Hall
Gainesville, FL 32611
AT&T: (904)392-2371
"There are two kinds of aircraft: Fighters and Targets."
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 30 Sep 84 17:34-PDT
Date: 28 Sep 84 18:05:11-PDT (Fri)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!duke!phs!lisa @ Ucb-Vax.arpa
Subject: Foreign languages on micros: a summary
Article-I.D.: phs.939
<>
Four weeks ago I posted a request for information on foreign language
word processing on micros. In light of the surprising amount of interest
in the subject, I would like to summarize the responses I received for
the net.
First, special thanks to each one who sent information and samples of
non-roman alphabet word processing. I regret that I was not able to
answer every response, but each of you are greatly appreciated.
Second, I have not been able to check out all of the various products
and claims made. Thus I am taking each response at face value, and
simply describing what has been presented to me.
Perhaps to no one's surprise, the consensus of opinion is that the
greatest potential for foreign language work is on the Macintosh.
The other major option for foreign language work (also no surprise)
appears to be the IBM PC.
Academic Font is a card/software package that handles Greek, roman,
and italic fonts on the PC. It claims to support several popular
dot matrix printers, as well as the Diablo 630 daisy wheel printer.
The samples I have seen (from an Epson FX-80 and an Okidata m-92)
are adequate, and handle full diacritical markings. I have not seen
how the screen display is handled. Supported word processors include
Final Word, Perfect Writer, and Mark of the Unicorn's Edix/Wordix.
University MicroComputers
1259 El Camino Real, Number 170
Menlo Park, CA 94025
Brit Company markets a scientific word processor that (reportedly)
generates characters "on the fly", and thus allows sophisticated
character combinations (i.e. accent marks, etc.) by overstriking.
My information here is scanty, but I understand that it requires a
Hercules graphics card and a good quality dot matrix printer (like
the Epson LQ1500). Apparently it prints all text in graphics mode
(and thus should come close to the flexibility of the Macintosh).
I believe Brit is in Philadelphia, but I am afraid I do not have
their address or cost information.
For Hebrew, Arabic, Farsi, and Urdu, an Israeli company that makes
word processing software for the PC is distributing through a
company in Los Angeles. I have not seen any samples of their
work, but I am told they are "quite proud" of their products
ability to handle left-to-right languages and right-to-left
languages simultaneously.
Kelmar Corporation
5456 McConnell Ave.
Los Angeles, CA 90066
attn Adriana Karin Vernon (213) 827-1200
Other possible solutions for the PC include generating printer
characters using "Fancy Font" (the sample text I have seen is
acceptable, but I assume that screen display is normal roman font).
Image Processing Systems (Madison, WI 50705) has a Proofwriter
word processor which, they claim, will handle foreign characters
via custom PROMs (I do not know if they also support printers).
The "ideal" solution seems to be the Xerox Star workstation. The
Star is the beneficiary of the Smalltalk philosophy (developed at
Xerox PARC) which is the basis of the Macintosh user interface.
An article in Scientific American earlier this year ("Multilingual
Word Processing" by Joseph Becker) outlines the Star's capabilities
(outstanding copy, comprehensive languages). Unfortunately, the
Star is very expensive ($10,000+ per workstation, plus Ethernet,
dedicated file and print servers, and a Laser printer). It is also
a closed box (the operating system is the best kept secret in town),
and incredibly slow (I am told all the software is written in a
dialect of Pascal). The Star is, however, the only "turnkey"
foreign language word processor of which I am aware.
Other possibilities include the Victor 9000. Ivo Welch at Columbia
University Linguistics Department is developing an Emacs/Scribe type
editor that will handle 254 characters and right-left cursor
movement. My own work with the Victor 9000 has produced DUKEFONT,
a BIOS screen patch that allows up to 16 font sets and screen
attributes to be displayed in word processing, communications, and
programming applications.
Finally, Steve Haflich at MIT has suggested that some graphics
terminals (notably the VT100) support downloadable character sets.
Thus a text editor in a UN*X environment could display foreign
language characters, and the result printed through some type of
customized *roff interface. I am not aware, however, of any
such product currently available.
Again, I thank all of you who were kind enough to respond to my
request for information. I remain interested in hearing of any
attempts to work with non-roman alphabets on micro computers,
and am more than willing to pass along any information of which
I am aware.
Jeff Gillette ...!duke!phs!lisa
The Divinity School
Duke University
Durham, NC
------------------------------
Date: Sun Sep 30 1984 15:57:37
From: Marco Papa <papa%usc-cse.csnet@csnet-relay.arpa>
To: info-ibmpc%usc-isib.arpa@csnet-relay.arpa
Subject: Lattice C exec/fork bug?
Has anybody figured out how to execute a subshell, that is, how to call
COMMAND.COM with the fork function of Lattice C version 2.12 ? Fork works
fine with normal .EXE or .COM programs, but whenever I try to call
COMMAND.COM I get errors like "Cannot find COMMAND" or "bad COMMAND".
Marco Papa
USC - Computer Science Dept.
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 28 Sep 84 23:50-PDT
Date: 23 Sep 84 19:55:43-PDT (Sun)
To: info-ibmpc @ Usc-Isib.arpa
From: hplabs!hao!seismo!rochester!jay @ Ucb-Vax.arpa
Subject: VI or NROFF clones
> A CP/M program called NRO was posted to the net a long time ago. I
> believe it is public domain. It is an nroff clone, written in C (I
> think Aztec, but I don't remember). Does anyone have the sources of
> this? Does anyone remember who wrote it? I think it may be due for
> reposting.
I have the sources for NRO, the program you mentioned. The title comments
are:
/*
* Word Processor
* similar to Unix NROFF or RSX-11M RNO -
* adaptation of text processor given in
* "Software Tools", Kernighan and Plauger.
*
* Stephen L. Browning
* 5723 North Parker Avenue
* Indianapolis, Indiana 46220
*/
It's pretty nice, especially if you like nroff. I have made a few
changes/additions to make nro look even more like it. I will send
it along to INFO-IBMPC, so look there in a little while . . .
Jay Weber
..!seismo!rochester!jay
jay@rochester.arpa
------------------------------
Date: Fri 28 Sep 84 09:38:56-PDT
From: ALFIERI@ECLD.#ECLnet
Subject: Document Problem
To: info-pc@USC-ISIB.ARPA
I think that his mail was lost or something, because
I sent the form and duly received (in about 3 or 4
weeks) the updated pages for the IBM BASIC manual.
I suggest that he try again!
--vince alfieri
computing information services
usc
alfieri@usc-eclb
------------------------------
Received: from Usenet.uucp by Sri-Unix.uucp with rs232; 28 Sep 84 6:15-PDT
Date: 20 Sep 84 6:13:21-PDT (Thu)
To: info-ibmpc @ Usc-Isib.arpa
From: decvax!mcnc!ecsvax!ncahec @ Ucb-Vax.arpa
Subject: Sidekick/monochrome query
Article-I.D.: ecsvax.3275
Does anyone know if the product "Sidekick" from Borland International
(home of Turbo Pascal, Turbo Toolbox, Turbo etc.) will run
satisfactorily on an IBM Monochrome monitor (with the standard
monochrome adapter card)?
Which leads to a more general question -
Does "windowing"-environment software run on a standard monochrome
configuration?
If so:
-- which ones do and which ones don't
-- what distinguishes those that do from those that don't;
Else:
-- never mind!
(endif)
Interesting responses will be posted. Thanx in advance.
---------
Jim Gogan (ncahec@ecsvax)
AHEC Program
School of Medicine
Univ. of North Carolina
at Chapel Hill
Chapel Hill, NC 27514
(919) 966-2461
------------------------------
Date: Friday, 28 Sep 1984 09:59:21-PDT
From: waters%viking.DEC@decwrl.ARPA (Lester Waters)
To: info-ibmpc@usc-isib.ARPA
Subject: 'True' BASIC
Has anyone heard about TRUE BASIC which is being produced by
John Kemeny (one of the original creators of BASIC) and Kurtz?
It is supposed to conform to ANSI proposed standards (i.e., be
Pascal-like supporting Procedures, etc.). As I understand it, Kurtz
is on the ANSI committee. There was an article in a recent
Popular Computing (which I haven't seen yet) about it.
Does anyone know if the ANSI proposed standard for BASIC is
floating around the Arpanet and/or any other network? I would
like to get my hands on it.
- Lester Waters -
WATERS%viking.DEC@DECWRL (Arpa)
...decvax!decwrl!dec-rhea!dec-viking!WATERS
(UUCP)
------------------------------
End of Info-IBMPC Digest
************************
-------