Info-IBMPC@USC-ISIB.ARPA (Info-IBMPC Digest) (01/21/86)
Info-IBMPC Digest Monday, 20 January 1985 Volume 5 : Issue 7
This Week's Editor: Eliot Moore <Elmo@USC-ISIB>
Today's Topics:
HACK sources
Escape from Cygnus X-1 : part one
Black Hole: part two
More creative names, please
Epsilon and Lightning
DISNDATA
Network Research Corp. FTP for IBM PC (Query)
"out of environment space", config.sys, autoexec.bat
Definicon trig routines
Mathematical and Statistical Software for IBM PC
Tax Software
MicroSoft Windows
MASM Bugs
Zenith Z150 Hard disk trouble
Cluster Size for 20 MB Hard Disk (3 Messages)
----------------------------------------------------------------------
Date: 20 Jan 1986 14:22:38 PST
Subject: HACK sources
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
The HACK sources will become available for ftp'ing when we successfully
de-ARC them. We've had some problems with this recently. If you're on
the Arpanet, you can ftp the ARChived HACK files right now. Just do a
MULTiple GET HACK*.* and then de-ARC them yourselves. The de-ARC'ed
files may be available next week.
Koji
-------
------------------------------
Date: 20 Jan 1986 12:03:48 PST
Subject: Escape from Cygnus X-1 : part one
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
The following programs have returned from the software void and are
available for ftp'ing. More programs will be returning in the very
near future.
DOS-HELP This subdirectory contains 52 DOS 2.1* help files, which
(subdirectory) cover DOS commands and concepts. Very useful for beginners
and handy for experienced people as well.
David Glaser <DG%LEXINGTON@COLUMBIA.EDU> 1/18/86
NEATLIST.BAS After NEATLIST.BAS is appended to a BASIC/BASICA program
and is executed, it generates a structured listing which
indents for IF-THEN-ELSE , WHILE/WEND, FOR and 2 spaces
for GOTO/GOSUB. At the end of the program, it lists the
number of lines, number of statements and the number of
REMarks. Remarks are indented and prefaced by an '*'.
Multiple Remarks are contiguous. Single or Beginning/
Ending REMarks are given an additional space.
<John Shaver, JSHAVER@APG-3> 1/17/86
NETIO.ASM MONITOR.ASM, used in conjunction with the device driver
monitor.asm NETIO.ASM allows DOS console I/O calls to be transmitted
netio.doc across an IBM PCNet local area network (or other NETBIOS
compatible network). The documentation file contains
patches to Kermit that allow it to be run over the network.
This code is a good source language example of how to
access IBM's net BIOS.
<Eric DeSilva, DESILVA@NBS-VMS> 1/19/86
SDL31.ASM Sorted Directory List. User has control over many
options. Another handy utility. Ages 5 and up.
<W. C. Bodycomb - original author>
<Updated by Ted Shapin, BEC.SHAPIN@USC-ECL> 11/21/85
-------
------------------------------
Date: 20 Jan 1986 13:23:00 PST
Subject: Black Hole: part two
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
The following programs have been added to our library:
EXTND.INC This include file contains macros for use with MASM
so that you can use the new instructions in the NEC
8088 clone - NEC V20/V30. Found in a BBS.
<John Ruschmeyer> 1/18/86
NLIST.PAS This is a menu-driven, ASCII file listing program for
Epson-compatible printers. It's written in Turbo
Pascal (version 3.0) and provides a number of printing
options. Written by Nate Liskov.
<Barbara H. Liskov, LISKOV@XX.LCS.MIT.EDU> 1/18/86
WHERE.PAS This is another "WHEREIS"-like file locator, written
in IBM Pascal. It starts at the root directory and
then tries all subdirectories. Wildcard indicators
such as '*' and '?' are accepted in the search string.
<Mike Johnson, MJOHNSON@ATHENA.MIT.EDU> 1/18/86
-------
------------------------------
Date: 20 Jan 1986 14:26:20 PST
Subject: More creative names, please
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
Howdy.
Ever since we added some of the DECUS Unix-like command programs last
year, other people have been submitting their own Unix-like command
sources. This is fine. The problem is that many people are using the
same names for their programs, like "PC-GREP," "PC-MORE," and "PC-WHEREIS."
As we all know that this is a PC-dedicated digest, there is no need to
give a program's name a "PC" prefix. If you're the original author,
please try to give your program a name that will have more chances of
being unique.
If you're submitting someone else's "PC"-prefixed program, try to contact
him/her (if it's feasible, of course) so that he/she can rename it....
*then* submit it to us. Otherwise, no fuzzy dice.
From now on all programs with "PC"-prefixed names will not be accepted
unless the submitter is absolutely sure that it was the first program to
be called so.
Let's be more creative with the names, eh?
Koji
Info-IBMPC program submissions
-------
------------------------------
Date: 20 Jan 1986 15:04:42 PST
Subject: Epsilon and Lightning
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: polish@CS.COLUMBIA.EDU
cc: info-ibmpc@USC-ISIB.ARPA
Todd Doucet informs me Epsilon doesn't indeed coexist with Lightning.
It runs fine with Sidekick but not Lightning. They are working on
Lightning compatibility and expect to have it soon. Sorry for the
mis-information.
-------
------------------------------
Date: 20 Jan 86 11:14:02 PST (Monday)
From: DHickey.ES@xerox.com
Subject: DISNDATA
In-reply-to: Your message of 19 Jan 86 16:05:33 PST
To: INFO-IBMPC@USC-ISIB.ARPA
cc: DHickey.ES@xerox.com
The DISNDATA disassembler sounds like just the product I need. Does
anyone have some data on the vendor (phone #, etc.)?
Dan Hickey
------------------------------
Date: Mon, 20 Jan 86 16:05:26 cst
From: dlnash@ngp.UTEXAS.EDU (Donald L. Nash)
Posted-Date: Mon, 20 Jan 86 16:05:26 cst
To: info-ibmpc@isib
Subject: Network Research Corp. FTP for IBM PC (Query)
Here at UTexas @ Austin, we have just obtained the FUSION package from
Network Research Corporation for the IBM PC. I've been working with it,
but have been unable to get the FTP server to work properly for it. The
user FTP and Telnet work, but not the FTP server. Here are the
symptoms: You can connect to the server and get this message:
220 Machine_Name FUSION FTP server (Version 3.1.7) ready.
The trouble starts when you try to log in. Before starting the server,
you set the PC-DOS environment variable F_USER=xxx where "xxx" is the
name which FUSION is to use as your user name. After connecting to the
server, if you try to log in as "xxx", a deadlock occurs. Under 4.2 BSD
FTP, you can use interrupt to break the deadlock, but the server does
acknowledge that the USER command was ever received. It simply acts as
if nothing has happened. If you try to log in as something other that
"xxx", then you get this message:
530 User yyy unknown.
You can log in as user anonymous and things will work as expected. You
can transfer files, etc. There are no access restrictions, since PC-DOS
doesn't know about file ownership.
I have been unable to find documentation about this in the manuals sent
with FUSION, and it really doesn't matter, since guest login is
available. I am simply asking this out of curiosity.
Thanks in advance for any info...
Don Nash
UUUU UUUU
UU UU UUCP: ...!{ihnp4,allegra,seismo!ut-sally}!ut-ngp!dlnash
UU TTTTTTTTTUUTTT APRA: dlnash@ngp.UTEXAS.EDU
UU TT TT UU TT
UU TT UU
UU TTUU
UUUUUUUU The University of Texas at Austin
TT Hook 'em Horns!
TTTT
------------------------------
Date: Sun, 19 Jan 86 19:44:23 est
From: malpass@ll-sst (Don Malpass)
To: INFO-HZ100@RADC-TOPS20, info-ibmpc@usc-isib
Subject: "out of environment space", config.sys, autoexec.bat
Because I recently installed a new vi editor on my Z-100 and
it called for several entries to config.sys and/or SET entries to
autoexec.bat, I've been playing with those two files. Among other
things, I added (and was forced to delete) SWITCHAR=- to allow slash,
instead of just backslash, to separate subdirectories [see p. 96 of
August 85 Dobbs article on make.] Things began to come unglued, and
I've deduced that it is almost certainly caused by filling up the
"environment space". At one point, when I had everything I wanted
defined, my boot message concluded with "out of environment space" or
some such. When I later tried to return from a shell-escape from
something it said "...COMMAND search directory bad" and forced me to
reboot since there was no COMMAND.COM in the default [ram]disk. One
time, invoking SET indicated that one of my variable strings had been
truncated. Another time, the whole list had been cut to two
variables, and PATH was equated to an empty string.
The book says the environment space can contain about 200
bytes, which is roughly the size of the current reply to SET. At the
mement, I NEED every line that is there, and I'd like a few others. Is
this a common problem? Can I enlarge the environment space? Should I
be doing something else? I just ordered Genie, a Sidekick-like product
for the Z-100, and I'd be surprised if it did not call for even more
environment entries. I'm using MSDOS 2.11 [or 2.13 depending on which
banner I believe]. I've been too lazy to install 2.2x. Would it
help? Can YOU help? Tnx.
don [malpass@LL-sst]
------------------------------
Date: Sat, 18-Jan-86 14:09:18 PST
From: ihnp4!pelican!pete@ucbvax.berkeley.edu (Pete Carah)
Subject: Definicon trig routines
To: pelican!ihnp4!seismo!munnari!mimir.dmt!mvr
Cc: info-ibmpc@usc-isib.arpa, pelican!pete
Using the latest compiler/library release from DSI (about 1-1/2 weeks
ago), I have run some checks of the DSI sin/cos routines against a
matrix rotation, and the usual sin**2+cos**2 checks
(max(abs(1-s**2-c**2))), using 10,000 steps around a circle for the
sample set, and they indicate errors of order 4e-16 in the sin/cos
library routines, and errors of order 4e-13 in the rotation versus the
library, and in the rotation radius. This is about what I would
expect from a floating point processor that both works in IEEE double
format and stores intermediate results in that format.
The same compiled program linked with my previous CLIB (I got the board
and software just before Christmas) gave results with 2-3 times the
error on all of the measures above. Since the matrix multiply was
written in line, the starting sin/cos had to be in error by this much.
This is still a maximum error of 2e-12 or less, which are not "3/4
rectified wave" by any stretch of the imagination.
The same program gives much better results (all 0 errors) using the
Microsoft C v.3.0, and a native 8087 (which works in 80 bit numbers
internally, has native trig routines NOT using polynomial
approximations, and keeps intermediate results within the matrix
multiply to the same precision).
I have no idea where your errors came from, unless they are entirely
from the graphics package, which I haven't checked.
-- Pete
...{ vortex, ihnp4, scgvaxd}!pelican!pete
------------------------------
Date: Mon, 20 Jan 86 11:29:57 cst
From: pool@anl-mcs.ARPA (Pool)
To: Info-IBMPC@USC-ISIB.ARPA
Subject: Mathematical and Statistical Software for IBM PC
Cc: pool@anl-mcs.ARPA
The Numerical Algorithms Group offers the NAG Fortran PC50 Library
for various compilers on the IBM PC. It includes 50 routines from
the NAG Fortran Library, a library of 525 routines available for
systems ranging from workstations to supercomputers. For additional
information, contact
Numerical Algorithms Group
1101 31st Street, Suite 100
Downers Grove, IL 60532-1263
(312) 971-2337.
The NAG Pascal Library, a collection of 80 routines, will also be
available for the IBM PC in approximately the last week of March.
PC-GLIM, (Generalized Linear Interactive Modeling), will also be
available in late February or early March to provide yet another
statistical package!
Jim Pool
Executive Vice-President
NAG
------------------------------
Date: Mon, 20 Jan 86 10:46:02 EST
From: Andy_Mondore%RPI-MTS.Mailnet@MIT-MULTICS.ARPA
To: Info-IBMPC@USC-ISIB.ARPA,
JMSK%MC.LCS.MIT.EDU@MIT-Multics.ARPA
Subject: Tax Software
There is a program called PC-TAX85 which handles federal tax
returns. It is available on CompuServe in the IBM Pro Sig.
It should also be available on other bulletin boards, although
I'm not sure which ones.
------------------------------
From: DGilbert.ES@Xerox.com
Date: 19 Jan 86 16:10:26 PST
Subject: MicroSoft Windows
To: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
cc: DGilbert.ES@xerox.com, IBMPC^.ES@xerox.com
I too purchased MicroSoft Windows. I too was disappointed, even though
my mouse works just fine. Yes, there are some things I like about it.
The clock is NEAT, with its simulated analog hands. The ability to flip
an application instantly from window to window, aaaa, sorry. tile to tile,
is very nice. Yes, there are no overlapping windows, so tiles are perhaps
a more appropriate term.
It crashed when I tried to run Crosstalk XVI in a tile to get my mail while
doing other things, don't know why......happened when I quit Crosstalk, then
tried to run it again....may be operator error, but...
I havn't had a chance to run all its features yet. However, I also have
DesqView from Quarterdeck, and so far I have to agree with George. DesqView's
the best multitasker I've seen so far....sorry MicroSoft.
{ the opinions expressed are my own, at least to date, and are subject
to change. I'm sure MicroSoft will not be affected significantly by
it. }
Doug.
------------------------------
Date: Sun, 19 Jan 86 22:42:58 pst
From: jhass%ucbiris@BERKELEY.EDU
To: info-ibmpc@usc-isib.arpa
Subject: masm bugs
For Schauble@MIT-MULTICS, the best books on the 8087 are:
(1) 8087 Applications and Programming, by R. Startz
(2) iAPX 86/88, 186/188 User's Manual (Intel)
(3) iAPX 286 User's Manual (Intel)
(4) Getting started with the NDP (AP-113, Intel)
****************************************************************
The new Microsoft Assembler (MASM 4.0) is well worth the price as
it is more than twice as fast as the previous version. (However,
I disagree with corwin@cdp.uucp who said it is twice as fast as
an assembler should be; it is only HALF as fast as it should be.)
Unfortunately, MASM still has a lot of bugs. Here is a list of
bugs that came to my attention from Roger Schlafly at Borland
International.
Microsoft Assembler 4.00 bugs:
(1) You cannot redirect the output to save the error messages
in a file.
(2) If you run out of disk space while the assembler is
assembling, the machine hangs.
(3) The /ML option does not work with the /Dsymbol option,
unless the /ML option precedes it.
(4) The instruction
fmul st(1),st(1)
is invalid, but MASM assembles it without reporting an error,
as if the instruction were:
fmul st(1),st(0)
(5) The following file assembles correctly with the command,
>masm bug /r;
but puts a bad opcode in the OBJ file if assembled with the
command:
>masm /ML bug /r;
;*************************************************************
; BUG.ASM
; File to demonstrate one of the bugs in MASM 4.0
s segment
assume cs:s
abCd2 dt 11223344556677889900R
org 20h
x proc near
fld abCd2
mov ax,7
ret
x endp
s ends
end
;*************************************************************
(6) You cannot include a file that has been edited with Wordstar.
The problem is that Wordstar puts end-of-file characters (hex 1A)
at the end of the file, but MASM chokes saying "extra characters"
if it sees more than one.
(7) MASM sometimes incorrectly gives a "value out of range"
error, as in the following:
;*************************************************************
false equ 0
true equ not false
db -1 ; This line is OK.
db true ; MASM 4.0 chokes on this line.
;*************************************************************
(8) This is not a bug, but when I read the ad that said, "define
symbols from the command line," I certainly expected that I would
be able to define the symbols to be something. For example,
/Dxmax=15
on the command line should be equivalent to
xmax equ 15
at the beginning of the file. This is what Microsoft C and a
number of other compilers and asemmblers do. Of course, I could
just put into my file a lot of ifdefs, e.g.,
ifdef xmax12
xmax equ 12
endif
ifdef xmax13
xmax equ 13
endif
ifdef xmax14
xmax equ 14
endif
ifdef xmax15
xmax equ 15
endif
ifdef xmax16
xmax equ 16
endif
ifdef xmax17
xmax equ 17
endif
and then just put /Dxmax15 on the command line if I wanted xmax
to be 15. Is this how the feature is intended to be used?
(9) This is perhaps not a bug either, but I find MASM's treatment
of far labels very strange when they are declared EXTRN inside a
segment. In such a case, MASM attempts to force the label to be
relative to a segment other than the segment it was defined to be
in. It seems to me that a far EXTRN label inside a segment should
be the same as outside the segment.
(10) SYMDEB still misses breakpoints a lot. I don't have a simple
example to prove it, but that bug has been in for a long time so
I assume Microsoft knows about it.
(11) The new linker (3.05) changed the link map format, thereby
breaking most of the symbolic debuggers on the market. Is this a
deliberate attempt to stifle competition?
(12) MASM still requires a carriage return at the end of the last
line in the file, or it ignores the line. There is a similar bug
in DOS's execution of batch files, so I assume it is intentional.
------------------------------
Date: 20 Jan 1986 11:39:59-EST
From: Mark.Wilkins@FAS.RI.CMU.EDU
To: info-ibmpc@usc-isib.arpa
Subject: Zenith Z150 Hard disk trouble
A friend of mine has a Zenith XT compatible, and claims to be having some
trouble with the hard disk. He mostly does word processing, using
WordPerfect 4.1. He says that just recently the hard disk began taking
a considerably longer amount of time to read (and write?) documents.
He called his dealer, who said they could look at it and replace either
the controller for ~$300 or the disk for ~$900! I think that something
is just out of adjustment, because it is working, albeit slow.
Any ideas?
Mark Wilkins
wilkins@fas.ri.cmu.edu.ARPA
------------------------------
From: DGILBERT.ES@xerox.com
Date: 19 Jan 86 16:08:30 PST
Subject: Cluster Size for 20 MB Hard Disk
To: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
cc: DGilbert.ES@xerox.com, IBMPC^.ES@xerox.com
I just upgraded my IBM PC/XT to a 20 Megabyte Harddisk from Priority-One
Electronics ( Seagate half height internal ). While I was at it, also
changed from 2.1 PCDOS to 3.1 PCDOS. Its great to have 20 MB, but something
puzzles me....
The cluster size is 8 Kilobytes. I read somewhere that PCDOS 3.10 is
capable of a longer directory entry ( 20 bits instead of 16 bits for
number of clusters ) and was therefore able to use a much more efficient
2 Kilobyte cluster size for large hard disks, such as 20 MB. Nowhere
did I find any way to do this. Is it automatic on the AT? The FORMAT
command has no parameters for forcing cluster size. There is a note
in the 3.10 manual that formatting a 20 MB partition may not allow
PCDOS 2.1 to read the disk. However, 2.1 seems to read my hard disk
just fine. But I would rather have 2 kb clusters. ANYWAY I CAN DO THIS?
Thanks for any help you can give.
Doug.
------------------------------
Date: 20 Jan 86 10:14:39 EST (Monday)
Subject: Re: Cluster Size for 20 MB Hard Disk
In-reply-to: DGILBERT.ES's message of 19-Jan-86 16:08:30 PST
To: DGILBERT.ES@xerox.com
cc: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>, IBMPC^.ES@xerox.com
From: Bill <Crocca.wbst@xerox.com>
Doug,
I tried that on my 6064/5. Going to 3.1 reduced the cluster size but to
4 sectors; I was hoping for one! The version of 3.1 I tried was from
IBM. I'm still running 2.11 because the 3.1 was just borrowed for the
test. My disk is a 20MByte tandon and an OMTI controller.
~ Bill
------------------------------
Date: 20 Jan 86 10:15:29 PST (Monday)
From: DHickey.ES@xerox.com
Subject: Cluster Size for 20 MB Hard Disk
In-reply-to: DGILBERT's message of 19-Jan-86 16:08:30 PST
To: DGILBERT.ES@xerox.com
cc: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>, IBMPC^.ES@xerox.com
Doug,
You must go into fdisk under DOS 3.1 and deallocate the DOS partition.
Then reallocate it. Then a format/s/v. DOS 3.1 gets BPB information
(cluster size from the DOS partition boot sector. The # of bits in each
FAT entry is indicated in the partition table in the hard disk boot
sector (not the same one). Both these areas must have the correct
values loaded for the system to operate correctly. If the DOS 2.x
information is there DOS 3.1 will treat the drive as specified in the
BPB table.
After you have setup the 2k cluster sizes correctly, DOS 2.x will not
work (period). The 12 vs. 16 bit FAT entries would screw up DOS, but
the DOS boot initialization code would not recognize the partition as
valid of DOS 04 vs. 01 for flag byte.
Good Luck.
Dan Hickey
------------------------------
End of Info-IBMPC Digest
************************
-------