PCjs Machines

The file ASM. is a tutorial on the assembly language for the IBM-PC's
8088 processor and was distributed by IBM itself.  It is a concise
description of the essentials of IBM-PC assembly language programming.
NEWCLOCK is a fixit device driver addressing the 24 hour rollover
problem as well as mis-stamped file times.  UASMLS is a utility that
will enable the creation of an almost-assembler-ready listing from the
output of a DEBUG dis-assembly.  The PCTALK files are overlays to the
original PCTALK III release and support VT100 or VT52 terminal
protocols.  TERM is overlay for the IBM ASYNC.BAS program.

Special Requirements:  Some programs require a version of BASIC.

How to Start:  Type GO (press enter).

Suggested Registration:  None.

File Descriptions:

BASCOM   DOC  Four patches for the BASIC Compiler.
CABLE    DOC  Procedure for making a cable for the Sony Profeel.
FPLOT    BAS  Plots math functions in polar and Cartesian coordinates.
FPLOT    DOC  Documentation file for FPLOT.BAS.
KBDFIX   BAS  Fix for keyboard buffer program in September Softalk.
NEWCLOCK DOC  Documentation file for NEWCLOCK.SYS.
NEWCLOCK SYS  Fix for DOS 2.0 clock update function.
PC-TALK  DEF  PC-Talk utility file.
PC-TALK  DIR  PC-Talk utility file.
PC-TALK  KEY  PC-Talk utility file.
PCT3VT   BAS  Vidtex, VT52, VT100 & ANSI cursor control to PC-Talk III.
PROFEEL  DOC  Review of Sony KX 1211 HG monitor/TV.
TERM     BAS  Adds auto-dial and auto log-on to IBM ASYSC. Comm. 2.0.
TERM     DOC  Documentation file for TERM.BAS.
TREND123 DOC  Documentation file for TREND123.MAC.
TREND123 MAC  LOTUS 123 macro for straight line/exponential curves.
UASMLS   DOC  Documentation file for UASMLS.EXE.
UASMLS   EXE  Formats DEBUG unassemble output file.

Here are a few of the patches from IBM related to the
IBM PC BASIC Compiler

-----CHAIN with COMMON-----

The use of CHAIN with COMMON when CHAINing from a large program
with a large data area to a small program with a large data
area may produce an incorrect "Out of memory" error.

  Here is the patch procedure:
A>rename basrun.exe basrun.tmp
A>b:debug basrun.tmp
-e1951
xxxx:1951  3B.72 C1.09 77.3b 03.c1 E9.77 CA.05 FB.e9
xxxx:1958  A3.ca 7A.fb 07.00 0B.00 FF.a3 74.7a 38.07
-w
Writing 7C00 bytes
-q
A>rename basrun.tmp basrun.exe

-----MOD-----

MOD (and others??) does not return the correct value under the
following conditions:

  1) Using a binary non-commutative operator such as MOD, IMP,
     of \ (integer division with the backslash)
  2) The left operand is simpler than the right operand
  3) The result is required in the BX register

The following example shows the problem:

 10 A=3 : B=4 : C=10
 20 PRINT "10 MOD (3 + 4) = "; C MOD (A + B)

In the unpatched compiler, it will return 7 (the right operand,
A + B).  The correct value is 3.

    The patch procedure:
A>b:debug bascom.com
-e7450
xxxx:7450  F6.86 87.47 02.02 00.24 08.08
-w
Writing A280 bytes
-q
A>



-----ARRAYS and OPTIMIZING-----

When using arrays and making calculations such as:
    T(J,L) = G(J)*E(L)+T(J+1,I)
the compiler may give an Internal Error or the calculation
may be performed incorrectly.  I think the error is in the
optimization process.  I wanted to compare the generated code
before and after the patch but the unpatched compiler bombed
on that statement with "Internal Error".

   The patch procedure:
A>b:debug bascom.com
-e8ab3
xxxx:8AB3  0C.1f
-w
Writing A280 bytes
-q
A>



-------------

How to get patches for a problem with IBM software.

   If you are having a problem with one of IBM's software
products for the PC, you should do the following.  Try to
isolate the simplist conditions that will cause the problem.
If the problem is with a language product, isolate it to
one line (if possible) and have a very short program
available that will demonstrate the problem.  Then, take
this information to your friendly dealer who sold you the
product.

   There should be someone at the dealership who can
understand your problem.  If he can't see anything wrong
with your sample program, he will have to contact IBM.
He will call the IBM hotline and tell them what product the
problem is with.  The support people are usually all busy,
so they will take the phone number of the dealer and will
call him back when someone is available (i.e. you will have
to come back/call back later for the answer).  When the
dealer gets in touch with IBM, they will try to find a
patch that deals with the area of your problem.  If one
is found, IBM will print the update description and procedure
and send it to the dealer.  These are titled "IBM Personal
Computer Programming Clarifications" and are usually a couple
of pages long, printed on an IBM/Epson printer.  If they
can find no match for your problem, they will ask for an
example of the problem, and work from there.


        Dave Alverson
        72155,1560
        September 15, 1983

-----------------------------------------------------------------------------
        Some additional patches courtesy of the SVCS Library SIG.
-----------------------------------------------------------------------------

The use of the INPUT# statement with quoted strings of length 1
may produce random results.

A>RENAME BASRUN.EXE BASRUN.TMP
-E6236
XXXX:6236  3C.b1 22.ff
XXXX:623B  75.3c 10.22 80.75 FA.10 2C.80 75.fa 0B.2c E8.75
XXXX:6240  7E.0b 00.e8 BA.7c 22.00 22.ba E8.22 48.22 D4.e8
XXXX:6248  72.46 40.d4 B1.72 FF.3e
-W
Writing 7C00 bytes
-Q
A>RENAME BASRUN.TMP BASRUN.EXE


A>B:DEBUG BASCOM.LIB
-ECBD7
XXXX.CBD7  3C.b1
XXXX.CBD8  22.ff 75.3c 10.22 80.75 FA.10 2C.80 75.fa 0B.2C
XXXX.CBE0  E8.75 00.0b 00.e8 BA.00 22.00 22.ba E8.22 00.22
XXXX.CBE8  00.e8 72.00 40.00 B1.72 FF.3e
-ECCA1
XXXX.CCA1  39.3b
-ECCA8
XXXX.CCA8  3F.41
-W
Writing 19600 bytes
-Q
A>

Test program:
10 OPEN "TSTPROG" FOR OUTPUT AS #1
20 FOR I=1 TO 10
30 WRITE #1,I,"A","BB","CCC","DDDD"
40 NEXT
50 CLOSE
60 OPEN "TSTPROG" FOR INPUT AS #1
70 IF EOF(1) THEN END
80 INPUT #1,J,A$,B$,C$,D$
90 PRINT J,A$,B$,C$,D$
900 GOTO 70

Program should print:
1 ABBCCCDDDD
2 ABBCCCDDDD
3 ABBCCCDDDD
4 ABBCCCDDDD
5 ABBCCCDDDD
6 ABBCCCDDDD
7 ABBCCCDDDD
8 ABBCCCDDDD
9 ABBCCCDDDD
10 ABBCCCDDDD

Test program for the ARRAY patch given above:

10 DEFINT D,F,N
20 DSD=0
30 FOR N=0 TO 7
40   FSEC=FSEC+1
50 NEXT
60 PRINT "The result is  ";512*(1-DSD)*FSEC,"The result should be 4096"






        These instructions are to help IBM PC users with the Color Graphics
Adapter construct a cable/interface for a Sony Profeel monitor.  These
are not complete instructions, but corrections to the instructions and
schematics that appeared in PC World, Volume 1, Number 3, pages 320-321,
and Softalk, August 1983, pages 8-9.   Sony now lists a cable for use
in interfacing with the IBM PC.  It is part # PC701.  It retails for
80 dollars.

        This cable/interface will only produce the high intensity colors.
Use of the IBM CG Adapter pin #6, used for setting intensity, is NOT
applied.  Use of that pin was NOT in the magazine instructions, and this
author is not qualified to figure out the correct use of that signal in
interfacing.  It is not known whether the Sony PC701 cable gives both
intensities.  I am waiting for a reply from Sony.

        These corrections have been determined through comparison of the
pin-outs of the Color Graphics Adapter printed in the adapter's
Inventory Checklist, the Sony 12" Profeel monitor's pin-ins as printed
in the Sony PX-34 instructions that came with the set, and finally, the
schematics printed in the two issues of the above mentioned magazines.
This author takes ABSOLUTELY NO RESPONSIBILITY for the results of this
cabling procedure.  You are onw your own.  These instructions are only
supplied in order to share my experience with you.

        The Sony Profeel and the IBM Color Graphics Adapter (CGA) differ
in their handling of the Vertical and Horizontal Sychronizations.  Sony
uses negative logic.  IBM uses positive.  To get the two to work with
each other, the polarity of the Vert and Horz signals must be reversed
between the two.

        The basic design of the schematics that appeared in the magazines
was correct.  The errors are all in the identification of the pin
numbers of the IBM CGA pin-outs.  The number labeling at the Sony end
was correct.

        In essence, the Red, Green, and Blue signals are passed from
the CGA directly to the Sony 34 pin connector.  This is just a straight
wiring.  The IBM signal pins were mislabeled in the schematic.  Pin #1
is shown as Ground.  This is correct.  Pin #2 is shown as the Red signal.
This is very incorrect!  Pin #2 is ALSO a ground.  The Red signal comes
from Pin #3.  The Green signal comes from Pin #4.  The Blue signal comes
from Pin #5.

        As indicated above, the IBM CGA pin #6 is for intensity and
IS NOT used in this cabling application (at this time - I am trying to
get data on how to use it correctly).  The IBM CGA Pin #7 is not used
by IBM and not used here.  I think it is a dead pin.

        The magazine schematics also mis-labeled the Pin Numbers for
pins #8 and #9.  As mentioned above, the Sony end IS LABELED CORRECTLY!
The IBM CGA pin #8 is the Horz Synch, NOT the Vert as listed.  The VERT
Synch comes from Pin #9 of the IBM CGA.

        Here are the correct Pass-Throughs for the interface:
IBM Pin # 1 (ground) to Sony Pin # 11 or Pin #13. (I used 11 because it
is the one Identified by Sony as Ground).

        IBM Pin # 2 to 74LS240 Inverter Pin # 1, and connect jumper from
74LS240 Pin # 1 to 74LS240 Pin # 10.  All of these are Grounds also.

        IBM Pin # 3 (Red Signal) to Sony Pin # 25 (Red).  IBM Pin # 4
(Green Signal) to Sony Pin # 26 (Green).  IBM Pin # 5 (Blue Signal) to
Sony Pin # 27 (Blue).  IBM Pins # 6 and # 7 are not used.

        IBM Pin # 8 (Horz Signal) to 74LS240 Pin # 2.  74LS240 Pin #18
to Sony Pin # 30 (Horz).  IBM Pin # 9 (Vert Signal) to 74LS240 Pin # 4.
Then 74LS240 Pin # 16 to Sony Pin # 31.

        Remember to jump the Vert and Horz signals to Ground via the
47 Pica Ferad Capacitors as indicated in the schematic.  This jump can
be made back to 74LS240 Pins # 1 or # 10, or to the cable wire running
from IBM Pin #1 to Sony Pin # 11 (or 13).

        There is one more wire that needs to be run.  It is a 5 volt
power supply than must be obtained and run to Pin # 20 of the 74LS240
Inverter.  The magazine instruction was to obtain the 5 V source from the
the IBM CGA's Light Pen connector.

        The Sony documentation indicated that a 5 V supply could be
obtained from the Sony Pin #1 or #2.  If the 74LS240 Inverter is placed
near the Profeel end of the cable the use of the monitor's 5 V source
makes much more sense.  The cable can be entirely self contained rather
than the sloppy arrangement of getting the power source from within the
computer.  This obviously allows easier connect/disconnect from the
computer when needed.

        My cable/interface came out well.  As I mentioned, I mounted
a 20 Pin Wirewrap socket (PAGE part # PWG20-2) to a mini-circuit board
(from neighborhood Radio Shack).  Radio Shack did not carry the 74LS240
Inverter.  I found it at another electronics supply house.  I mounted
the circuit board in a Radio Shack Deluxe Project Case #270-220.  I
had to slice off some internal ridges with an EXACTO knife, and had to
drill holes through which to route the cable.  Radio Shack stores now
also carry the DB 9 connector needed for the IBM end.  Be VERY CAREFUL
with the Sony 34 pin connector.  I don't know how difficult it would be
find another.

         CompuServe 70055,474


Disk No:  114
Program Title:  Assembly
PC-SIG version: 1.2

The file ASM. is a tutorial on the assembly language for the IBM-PC's
8088 processor and was distributed by IBM itself.  It is a concise
description of the essentials of IBM-PC assembly language programming.
NEWCLOCK is a fixit device driver addressing the 24 hour rollover
problem as well as mis-stamped file times.  UASMLS is a utility that
will enable the creation of an almost-assembler-ready listing from the
output of a DEBUG dis-assembly.  The PCTALK files are overlays to the
original PCTALK III release and support VT100 or VT52 terminal
protocols.  TERM is overlay for the IBM ASYNC.BAS program.

Usage:  Programming/Education.

Special Requirements:  Some programs require a version of BASIC.

How to Start:  Type GO (press enter).

Suggested Registration:  None.

File Descriptions:

BASCOM   DOC  Four patches for the BASIC Compiler.
CABLE    DOC  Procedure for making a cable for the Sony Profeel.
FPLOT    BAS  Plots math functions in polar and Cartesian coordinates.
FPLOT    DOC  Documentation file for FPLOT.BAS.
KBDFIX   BAS  Fix for keyboard buffer program in September Softalk.
NEWCLOCK DOC  Documentation file for NEWCLOCK.SYS.
NEWCLOCK SYS  Fix for DOS 2.0 clock update function.
PC-TALK  DEF  PC-Talk utility file.
PC-TALK  DIR  PC-Talk utility file.
PC-TALK  KEY  PC-Talk utility file.
PCT3VT   BAS  Vidtex, VT52, VT100 & ANSI cursor control to PC-Talk III.
PROFEEL  DOC  Review of Sony KX 1211 HG monitor/TV.
TERM     BAS  Adds auto-dial and auto log-on to IBM ASYSC. Comm. 2.0.
TERM     DOC  Documentation file for TERM.BAS.
TREND123 DOC  Documentation file for TREND123.MAC.
TREND123 MAC  LOTUS 123 macro for straight line/exponential curves.
UASMLS   DOC  Documentation file for UASMLS.EXE.
UASMLS   EXE  Formats DEBUG unassemble output file.

PC-SIG
1030D East Duane Avenue
Sunnyvale  Ca. 94086
(408) 730-9291
(c) Copyright 1985,86,87,88.89 PC-SIG, Inc.


   This program is an IBM adaptation of an Apple program written by James
Fuller, that appeared in Creative Computing, January 1983.  The program
allows the plotting of a function defined at line 270 in Cartesian, or Polar
coordinates or as a 3 dimensional "Solid of Rotation".  The program also
allows the screen image to be saved for later recall.
   Three dimensional plots are quite time consuming (1-2 hours), although
often very interesting.  The program can be compiled which speeds up the
3d plots considerably considerably.
   Some suggested equations:

                DEF FN R(Q)=COS(7*SIN(2*Q))  in Polar coordinates
                DEF FN R(Q)=COS(Q)+COS(2*Q)+COS(5*Q)  any type of plot
                DEF FN R(Q)=COS(2*Q)+COS(X+BB/16)  3D plot
                DEF FN R(Q)=COS(4*Q)+(20/X^2+3))   3D plot

   Some interesting spiral graphics can be obtained with polar plots with
the following alterations:
                    Delete 750, 760
                680 FOR G=0 TO 18000 STEP INX
                770 IF G=0 THEN PSET(1.2*G+XOFF,YY)
                775 line -(1.2*XX+XOFF,YY)
        try DEF FN R(Q)=Q/3.14159 or R(Q)=COS(4*Q)
Use large increments (25-150), and small scale factors (0.05-0.5) for best
results.  Other examples were listed in the article.

                                        David Bush
                                        71046,654


╔═════════════════════════════════════════════════════════════════════════╗
║                   <<<<  Disk No: 114  Assembly  >>>>                    ║
╠═════════════════════════════════════════════════════════════════════════╣
║                                                                         ║
║   To print the documentation on this disk, type:                        ║
║                                                                         ║
║             MANUAL (press enter)                                        ║
║                                                                         ║
╚═════════════════════════════════════════════════════════════════════════╝

The following paragraph is a cogent explanation of the functioning of the Dos
2.0 CLOCK$ routine and the partial fix made by Larry Rosen:

There are two different "date" problems, Earle. One is caused by the fact that
the BIOS updates the time and DOS updates the date. That is, just because the
time passes midnight and is reset to zero, the date is not changed until DOS
"notices" that condition. If <two> midnights are passed without anyone asking
DOS for the time or date, it can't know that any more than <one> midnight was
passed. The other problem is the way DOS determines whether midnight was
passed. Without my "fix" DOS uses the "midnight" flag returned by a TOD request
to the BIOS. The <first> request after midnight receives this indication, which
is then reset. The problem in DOS 2.0 is the fact that the builtin CLOCK$
routine is not going to be the first requestor after midnight since other
routines also request the TOD, especially the diskette and hard disk routines
which are more popular. My "fix" causes DOS to detect the date change whenever
the current TOD is less than the previous TOD. The technique is not perfect,
but it's better than the old way.

SVCS Library SIG comment: DOS 2.1 functions correctly, no fix required.


        First, let me thank those SIG members who helped me with the process
of gathering information and then tying it together.  In no way could I have
gotten the job done without that help.

        As I type this text, I am viewing a brilliant two-color Wordstar.
The Sony is fired up and works beautifully.  It is the 12" PROFEEL, Model
Number KX-1211HG.  Last night I left a reply on this SIG to Dave Crane
and advised him that the image and text quality of this monitor was
better than the Amdek II RGB, but maybe not as good as the IBM, Quadram,
or Princeton.  I lied!!

        Maybe I had been up too late on several consecutive nights.
Maybe I had read too many specs (such as the Sony spec on PROFEEL Horz
Resolution that only claimed "..greater than 350 lines...").  Today I
got a chance to compare this jewel side by side with the same Princeton
that I judged side-by-side with an IBM.  All Five people judging it
judged the PROFEEL as good, and three of the five judged it better!

        Here is an interesting quote from the Sony PROFEEL KX-1211HG's
Operating Instructions Book.  "...The line system is switched
automatically from the normal 525 lines to 625 lines according to the
input from the microcomputer...".  Why Sony doesn't publish this spec
is beyond me, but the text and graphics quality speak for themselves.

        Another real plus, this 12" monitor gives me an eight and three-
quarter inch wide 80 column screen.  My Zenith ZVM-121 green monochrome
display, also 12", gives me only seven and one-half inches of width
(full 80 columns).

        The PROFEEL does not have the richer background of the IBM, but
the Sony seems to have more brilliant color than the rest of the field
(sans IBM).  The text is clearer and sharper than I ever hoped for.  And
the one big plus for me --- I can reach to the front panel and toggle
from RGB signal to TV/VCR signal, all with one easy touch.  And the TV
picture quality is unexcelled.

        I am going to write and upload cabling instructions to the SIG
XA area.  The schematic that appeared in PC World Vol 1, No 3, page 321,
and Softalk, August 1983, both had errors (identical).  I will document
those errors and give corrections.

        A major limitation IS left.  The color intensity is FIXED.  The
cable DOES NOT handle the Intensity signal-out from the Color Graphics
Adapter.  That pin-out #6 of the adapter is not used at all in the
cabling.  I do not possess the knowledge on how this intensity-out works.
The PROFEEL has a "Level Set" pin on it's pin-ins.  I have not tried any
connection between the two as I am afraid of hurting one or the other.
I am looking for a cure to this and will let the group know
when I discover it.  If anyone knows, please give me the info or a ring
at (918) 250 9770.

                                ADDENDUM

        Some additional information --- I am adding this one day later than
the original part of this text was created.  Today I was able to do a
side by side comparison with the IBM Color Display.  It was done at a
local computer store.  Everyone's eyes bugged out.  The two displays were
identical in clarity and quality.  Brightness and color were exactly the
same.  The ONLY difference was the background.  The Sony 12" Profeel
had just as black a background as the IBM.  The difference is that the
IBM has more of a flat black or non-glare appearance.  The actual
display screen of the Sony is larger, but I did not have the opportunity
to take measurements as I did vs. the Zenith green screen monitor.

        Anyone with nearby access to a video recorder/tuner would be
foolish to overlook this option.  Think of the possibilities.  The
combination of the PC / Profeel / and PC could be used as a terrific
computer training aid.  The student/teacher could toggle back and forth
between active computer and video-taped instruction as needed or
instructed.              -FP

         CompuServe 70055,474


        This is a program that is designed to be merged into the TERMINAL.BAS
program of the IBM Asyncronous Communications Support 2.0 to provide autodial
with a forty number directory and auto log-on capability for Compuserve and
very likely other services as well.  It also contains the lines neccessary to
add color to the program operation, and is formatted to work properly in 40
column as well as 80.  For auto log-on to Compuserve it requires a modem that
displays a screen message indicating that the carrier tone from the host
computer has been detected, for example the "CONNECT" message provided by the
Hayes Smartmodem.  The log-on sequence works by watching for "prompts" that you
have entered at the time of making your directory entry-- (actually, it watches
for the last three characters of the screen message-- "ECT" for "CONNECT",
"ID:" for "USER ID:", and "rd:" for "Password:".  Note that you must specify the
prompts exactly, with upper and lower case correct.  When you respond to the
input statements while entering the log-on information, you cannot enter
certain characters directly such as the "^C" (ASCII 3) neccessary to respond to
the "CONNECT" message from the modem.  Instead, enter the desired ASCII number
plus 128 by holding down the ALT key while typing the number on the numeric
keypad, then releasing the ALT key.  (For example, type 131 for a "^C" or 141
for a carriage return.  The character produced on the screen will actually be an
ASCII 131 or 141 but will be stripped of the added 128 when it is sent by the
communications program).  More instructions can be found by reading the program
listing.  For safety's sake, *DO NOT MODIFY YOUR ORIGINAL PROGRAM DISK!*.  The
program is actually written to use the Hayes Smartmodem and touch-tone dialing.
If your phone line will not support touch-tone, it will be neccessary to change
the "ATDT" in line 4430 to "ATDP".  This line and others may have to be further
modified if this program is to be adapted to another brand of auto-dialing
modem.  Please send questions, comments or suggestions to Robert Relf,
[70176,403].


                                   TREND123
                                 version 1.0
                       A Lotus 123 Curve fitting macro
               created by John C. Dannenfeldt (Compuserve 71435,561)

The file TREND123.WKS is a independent macro that may be used within any Lotus
123 spreadsheet.  This version enables you to create a set of data points with
the range name LINE or CURVE which fit any set of data points you range name
DATA with either a least squares straight line or an exponential curve:

     Linear (LINE)-         y=(c1 * x) + c2

     Exponential (CURVE)-   ln y=(c1 * x) + c2

Restrictions:
     The DATA range must be continous.. no blank cells.
     ...     ...    must be less than 257 points long.
     You must be at A70 before you load the macro into your worksheet.
     Your spreadsheet must not extend beyond A1..IV69 or it may be partially
        overwritten.

Instructions:
     Before running the macro you must:
          1) Define a continuous set of numbers range named DATA.
               They may be in either a row or column orientation but the
               program takes much longer to work on the vertical ranges
               because it must first transpose them into a horizontal range.

          2) Name the macro so that you can call it by typing:

               /RNC\t  then press (ENTER)
               A70     then press (ENTER)

          3) Load the macro into your spreadsheet-
             Now goto A70 and type:

               /FCCETREND123   then press (ENTER)

    To activate the macro type ALT-T and follow the instructions.  When the
    macro is done it will return you to its main menu and ask if you would
    like to do a linear or exponential curve fit or quit.  If you do a
    different curve fit the old LINE or CURVE will be destroyed.  If you quit
    you may then look at LINE or CURVE directly or even better graph it beside
    the original DATA that you were trying to fit the curve to.

Comments:
     I was forced to create some type of curve fitting regime for an
application I had and in the process developed part of the current macro.  The
next version will include some additional curve fitting routines such as
parabolic, moving averages and exponential smoothing.
     Some of the things I don't like about this macro are a result of the
limitations of the 123 macro language.  After finally figuring out how to
transpose a column into a row I certainly wished that 123 had a general
command for that function.  Hopefully the next version will have an easier to
use macro language.
     Feel free to make changes to the routine to fit your needs.  Some obvious
variants would be a basically vertical layout (which would make vertical
ranges faster to input) and locating the origin of the macro at a point more
suitable for your particular spreadsheet( remember that all references to
cells within a macro must be changed for it to work in a new location).
     If you got this macro in the form TREND123.HEX you must convert it to a
binary file called TREND123.WKS before trying to use it in a worksheet.

 If you have any suggestions contact me on COMPUSERVE at the IBMPC SIG.


                           Uasmls

     Ever  use  DEBUG to dis-assemble a file and sit  around  for
hours  doing  a simetaneous dump to the printer,  all  the  while
wishing you could put it in a file instead,  and them get down to
actually  editing  the source code and re-assemblying,  with  you
changes? Pipes and filters are the answer.

     The  DOS manual says pipes "allow the screen output  of  one
program   to   be   used  as  the  keyboard  input   to   another
program....and piping is the chaining of programs with  automatic
redirection of standard input and output."

     Well,  exactly what is standard input and  output?  Keyboard
and screen, respectively. The redirection of standard output is a
cinch,  just  tell  DOS  to  send  the output  to  a  disk  file.
Unfortunately,  this kills the standard output to the screen,  so
you don't know what you are doing. Working in the dark, you might
say.  This is where the use of input redirection comes into play,
why not create a script of debug commands,  use that as  standard
input  and kick back and let the script do all the work.  This is
how it can be done for creating a debug dis-assembly file.

A>DEBUG
-F CS:0100 L1000 "0"   <=== Fills 1,000 bytes with hex "F"
-Q
A>DEBUG MYFILE.COM     <=== Load the file to be dis-assembled
-D                     <=== Dump  the  file,   note   addresses
                            containing data, continue Dumping
                            until you reach the end of the file
                            (section when contains contigious
                            FFFFFF's).
-Q                     <=== Quit debug
A>Copy con infile
U addr1  addr2         <=== Start and stop address
U addr3  addr4         <=== Start and stop address
etc.                   <=== Until all code addresses are specified
q                      <=== Tell debug to Quit
Ctrl Z                 <=== End of file mark

A>DEBUG MYFILE.COM <INFILE >OUTFILE

     Think of the pipes as funnels.  In this case INFILE will  be
used  in lieu of keyboard input to debug.  All output that  would
normally  be  sent to the screen will be piped to a  file  called
OUTFILE.  Of  course,  the names of these files can be changed to
any filename that you would like to use.

     You've  now  created  a  file  which  is  identical  to  the
unassembly that you would normally see on your screen. Now run the
UASMLS program. The program  will ask for the  input  file.  Enter
OUTFILE.  When  asked  for  the  output file  name  enter  a  new
filename,  usually with a .ASM extension.  The program will  then
read OUTFILE,  eliminate the hex addresses on the left and assign
labels  (in  the form Ln,  where n = 1 - 999) for all  jumps  and
calls, and send the results to the new file you have specified.

     If  you  get an error message "Error:  Referenced  code  not
found:",  it means that a jump or call was found which referenced
an  address  not found in the input file.  This means you  missed
portions of the code, you started unassembling in the middle   of
an instruction,  or you got into data areas


 Volume in drive A has no label
 Directory of A:\

BASCOM   DOC      4772   1-06-84   9:59p
CABLE    DOC      5317  12-11-83  11:30a
FPLOT    BAS      7231  12-11-83   2:54p
FPLOT    DOC      1406  12-13-83   7:29p
KBDFIX   BAS      1350  12-11-83  12:03p
NEWCLOCK DOC      1271   1-06-84   9:05p
NEWCLOCK SYS       384  12-11-83   2:42p
PCT3VT   BAS      7696  12-11-83   2:21p
PROFEEL  DOC      4022  12-11-83  11:36a
TERM     BAS      5686  12-11-83   2:23p
TERM     DOC      2140   1-06-84  10:06p
TREND123 DOC      3250  12-11-83   2:39p
TREND123 MAC      4224  12-11-83   2:32p
UASMLS   DOC      3158  12-11-83  11:40a
UASMLS   EXE     22656  12-11-83  11:55a
PC-TALK  DEF       640   1-11-84   9:44p
PC-TALK  KEY      5248   1-11-84   9:44p
PC-TALK  DIR       512   1-11-84   9:44p
GO       BAT        38  10-19-87   3:56p
GO       TXT       694  12-01-88   2:46p
MANUAL   BAT       147   6-08-88   3:02p
FILES114 TXT      1996   1-27-89   2:34p
       22 file(s)      83838 bytes
                       69632 bytes free