Full text of "Computer_Notes_1978_03_07"

See other formats

computer

ocriPLE & PRODUCTS

TO MEET ^ K ,r F p - s "
1 CHALLENGE

. t com-

^ rl " a mir lead m . *» servicing r r $100

, to extend «« "L marketing, ceS s oi o

namic divi ds w weS , "^niinistration,

? Wte coking ^ r eS Se today to:

5nS « » ail your

oaift DeS

ass*

9610 DeSoto AjgJ^ 9 l311

•(Ve are an AJJ 1

PERTEc Cn»?£ NTE RS. Entirf b ' e for 50«J e i h P year 'or

iSS ruTEBconp ° R -^

Products or

tR coRPo RAr/ «« i m a gazine enrec computer ^

In This Issue

The Latest with CN

System Timing Modification for the MITS/Altair88-DCDD Floppy Disk. 4
by Tom Durston

A. Copy/ Rewrite Procedure 4

by Gale Schonfeld

B. Single Drive BASIC Diskette Rewrite Procedure 5

by Charles Vertrees

C. Single Drive DOS Diskette Rewrite Procedure 6

by Drew Einhorn

D. Easy Floppy Disk Alignment Check 6

by Tom Durston

More on the KCACR

By Doug Jones

field starts with the Sync Bit 250 to
350|us (old timing) or 350 to 500|us
(new timing) after the beginning of the
sector. The logic 1 pulses after the
beginning of the sector are caused by
the notse written by the Write circuit
being turned on when that sector was
written. There should be a long period
(250-400us) of all logic from the
noise pulses to the Sync Bit.

For optimum timing, Read Clear
should go low halfway between the
noise pulses and the Sync Bit. The
Read Circuit will generate errors if the
noise pulses occur after Read Clear
goes low or if the Sync Bit and Data
occur before Read Clear goes low.

If necessary, the Index Sensor may
be temporarily adjusted to allow
proper reading of a diskette by
centering the low time of Serial Read
Data as described earlier. Note the
original position of the Data, so the
Index Sensor may be returned to
normal. Check both inner and outer
tracks of the diskette in order to
compensate for skew in the data.

Program on page 27

About the Authors

Tom Durston is the MITS Engineering
Program Director and is involved primarily
with peripheral interface design. A MITS
employee for five years, Durston studied
Electrical Engineering at the University of
Virginia and the University of New Mexico.

Gale Schonfeld has been employed by MITS
for two years and is the Software User
Specialist. She is currently pursuing a
Bachelor of Science degree at the University
of New Mexico in Electrical Engineering/
Computer Science.

Chuck Vertrees is the Director of Software for
MITS. He has a B.S. in Electrical Engi-
neering from the University of New Mexico
and is currently studying for a Masters in
Computer Science.

Drew Einhorn holds a degree in Mathematics
from the University of Oklahoma. He has been
employed by MITS for two years as a
scientific programmer.

Theannouncement of the new MITS®
KCACR board (Kansas City Audio
Cassette Recording) for their MITS/
Altair™ microprocessor was indeed a
welcome relief for me and for a still
ailing papertape reader. With the
installation of this single board, a
world full of holes and spilled chad
has turned into neat little plastic
boxes each with a cassette tape. The
chaos of rattle-rattle-checksum error
has turned into absolute quiet, broken
only occasionally by an eject-click.

Regarding the hardware, the board
occupies one slot of the 680 expander
board. Its features include CMOS
logic for low power consumption, and
it uses total digital logic without a
single potentiometer or adjustment.
The input/output is at 300 baud,
allowing a speed tolerance of 20%.

The software that is supplied with
the KCACR is, likewise, quite good.
MITS' CSAVE BASIC is supplied on an
audio cassette tape and its features
still amaze me. A bootstrap loader
PROM chip that fits into one of the
PROM sockets on the main board is
also supplied. Since this chip has no
name, I will refer to it as the KCACR
MONITOR. A large portion of this
article will concentrate on this chip.

Since there are many things to
discuss about the KCACR and related
software, I have organized this article
into four sections, all intending to
help you gain the most from the
hardware and software of the KCACR.

This writing will appear at times to
be a collage of software tidbits that
have appeared over the last year in
Computer Notes, I would like to
give credit where it is due. My thanks
to Mark Chamberlin (I literally stole
his PUNBAS routine) and to Ron
Scales for his help on a rather sticky
interrupt problem.

I. Inverse Assembly of the KCACR
MONITOR

After putting this new PROM chip
on my 680 processor board, it was
nice to see its two primary functions
work well. A (J)ump to FD00 will allow
a load of a Motorola-formatted audio
cassette tape through the new port,
and a (J)ump to FD74 will allow a
properly-formatted dump of any se-
lected portion of memory. And it really
works quite well.

But curiosity started to get the
better of me. Exactly how does it
work, I asked myself. Are there any
useful subroutines in it that can be
called by other programs? Are there
any provisions for turning off the
motor on a checksum error? I wanted
to know the answers to these and
other questions.

I ran a 680 Inverse Assembly
("Inverse Assembler Makes Machine
Language Programs Understandable",
by Doug Jones; Computer Notes,
July 1977) on it and produced the
listing that is shown. The comment,
labels, and a bit of doctoring-up was
done using the EDITOR.

I received answers to my initial three
questions and they were "well", "yes",
and "no". It may not turn off the
tapedeck motor on a checksum error,
but there are some useful routines in it
that are easily called from an assem-
bly language program. If you spend a
few minutes and study the KCACR
MONITOR program, perhaps you will
spot some useful subroutines or learn
a new programming technique, such
as the following question illustrates
about the KCACR MONITOR. The
problem is, "If BADDR (address
$FD59) is a subroutine that required a
JSR to enter, how do you exit?"
II. Comparing the KCACR MONITOR

to the 680 MONITOR

Table 1 compares the addresses of
the major subroutines of both MON-
ITOR programs, and, interestingly
enough, both sets of subroutines
function identically except that they
address different ports. For example,
you wish to send a letter to the
teletype port

C6 XX LDA B #'(letter)
BDFF81 JSROUTCH

;680 PROM MONITOR address.
On the other hand, you wish to send
a letter to the KCACR
C6XX LDA B #'(letter)
BDFDF5 JSROUTCH

; KCACR MONITOR address.

The 680 PROM MONITOR manual
will give you register usage on all of
the other subroutines mentioned in
Table 1 . Beware, for there are some
hidden "GOTCHAs", at least they
always seem to get me. A call to INCH
does not return an 8-bit character;
parity has been stripped off of it. ..will
I ever learn?

Computer Notes Jan/ Feb 1 978

III. Preparing Your Other Software for
Use with the KCACR

You may wonder why there is a need
for converting your original BASIC or
EDITOR to KCACR. CSAVE BASIC is
good; as a matter of fact, I use it
99 percent of the time. But, naturally,
my favorite demonstration program
needs the extra few hundred bytes that
the original BASIC has in usable
memory. Also, since I am generally
trying to emerge from the papertape
world, the EDITOR and the EDITOR/
ASSEMBLY are naturals to convert to
KCACR format.

The PUNBAS ("680 Software News",
by Mark Chamberlin; Computer Notes,
November 1976) program was easily
converted to a PUNKCR program for
these purposes. Let me warn you of
several sticky areas.
BASIC
V1.0 R3.2 9/25/76

Load in the PUNKCR program.

Load BASIC, but do not initialize.

Make patches to BASIC'S CONT

statement (see "680 Software

News" article).

Dump BASIC to cassette by doing a

(J) 4000.
Later Revsions
Check last load line of BASIC tape

for address of last byte.

Load PUNKCR program.

Load BASIC, but do not initialize.

Adjust LDX statement at $4000 for

last byte address.

Dump BASIC to cassette by doing a

(J) 4000.
EDITOR

R1 .0 9/30/76

Load PUNKCR program.

Load EDITOR program, but do not

initialize.

Dump EDITOR to cassette by doing

a (J) 4005.
EDITOR/ASSEMBLER
R1.0 9/30/76

Load PUNKCR program.

Load EDITOR/ASSEMBLER, but

do not initialize.

Make patch correction ("Software

Tidbits", by Mark Chamberlin;

Computer Notes, April 1977) to

EOR statement by depositing an

$88 at address $03A7.

Dump EDITOR/ASSEMBLER to

cassette by doing a (J) 400A.

I suggest using fifteen-minute per
side tapes. I also suggest following
MITS' advice to dump the same thing
to both sides of the tape to save
rewind wear and tear. These are your
big programs, so you will need to buy
three tapes.

680

KCACR

ROUTINE

MONITOR

BADDR

FF62

FD59

BYTE

FF53

FD4B

INCH

FF00

FD62

INHEX

FF0F

FD36

OUT2H

FF6D

FDE3

OUTCH

FF81

FDF5

POLCAT

FF24

-no equivalent

Table 1. Subroutine Address Comparison

IV. Techniques of Using Other Soft-
ware with the KCACR

Table 2 shows a cross-reference of
INCH and OUTCH subroutine calls
and their respective addresses in both
MONITOR programs. For example,
you have finished a long session with
the EDITOR, and you wish to store
your buffer on cassette for future use.
Exit EDITOR with X$$
Set nulls .M00CE0010

Adjust OUTCH call .M01EEFFFD

.N01EF81 F5
Return EDITOR .J010A
After this point, you will not be able
to see echo, since it is being sent to
the KCACR port.
To dump, type FFEF$$

At some later date, you may wish to
reload this tape into, for example, the
EDITOR/ASSEMBLER.
Initialize E/Aby

doing
Exit the EDITOR

with
Adjust INCH call

.J 0107

X$$

.M0184FFFD
.N 01 85 00 62

Jump directly to
APPEND function J19FF.

Your load is completed when the
terminal begins to rattle in response to
some impulses on the KCACR. Next,
hit the computer RESET, and readjust
the INCH call .M0184FDFF

.N 0185 62 00
Return to EDITOR . J 01 0A

Using the full editing features,
check the first and the last few lines of
your buffer. It is likely that the first
line will be FFEF$$, which can easily
be killed.

Table 2 also shows the PEEK and
POKE cross-references for the same
subroutine calls. For example, if you
are in BASIC and if you wish to send
some of your PRINT statements to the
KCACR, do a POKE 2222,253: POKE
2223,245. To return the print to the
teletype port, do a POKE 2222,255:
POKE 2223,1 29.

Be alerted that the POKE/ PEEK
addresses shown here are for the
subroutine addresses; the JSR com-
mand ($BD) is found one address prior
to those. For example, if you wanted
to NOP the OUTCH call out of CSAVE
BASIC, you would have to NOP three
consecutive addresses beginning at
$08BB.

The technique in BASIC of alter-
nately writing to the teletype, the
KCACR, or NOPing the OUTCH call
(writing to the bit bucket) might prove
a useful technique for debugging a
program.

In summary, many ideas have been
presented in this article, some of
which are good and some you may
consider not so good. I hope you will
be able to improve on both. But, no
doubt about it, MITS has a good
product with the KCACR board.
Program on page 29

BASIC

CSAVE BASIC

INCH

OUTCH

POLCAT

$0420
$08 A E
$061 C

P1056
P2222
P1564

$042E P1070
$08BC P2236
$0627 P1575

EDITOR

EDITOR/ASSEMBLER

INCH

OUTCH

APPEND

$0169
$01 EE
.J 0689

$0184
$022E
.J 19FF

680 MONITOR

KCACR MONITOR

INCH

OUTCH

POLCAT

$FFO0
$FF81
$FF24

P255,000
P255.129
P255.036

$FD62 P253.098
$FDF5 P253.245
—no equivalent—

Table 2. INCH and OUTCH Cross Reference

Computer Notes Jan/ Feb 1 978

Union County Career Center, Update 1977

By James Gupton, Jr.

During 1977, Computer Notes car-
ried two articles dealing with the
high school students in Union County,
North Carolina and their experiences
in the construction of a MITS®/
Altair™ 680B microprocessor com-
puter. To the MITS/Altair computer
owner who receives Computer Notes,
there must be some questions as
to why high school students should
receive such prominence. The answer
is simple. The Union County Career
Center is the first vocational education
center in North Carolina to offer an
adult level curriculum with computer
construction and programming. Most
educational institutions that include
computer operation in their programs
have relied on "breadboarded" com-
puter circuits of limited program
capabilities, whereas the MITS/Altair
system was far better suited for the
tone of the Union County Career
Center's Electronics program. This
article will center on the student
assembly of the 680B-BSM 16K RAM
circuit board for adapting the 680B to
BASIC language programming.

There are no words to explain the
feeling of apprehension when en-
trusting almost $700 in circuit compo-
nents to the semi-skilled hands of
high school students. It is similar to
the feeling a father has the first time
his son asks for the family car! As a
teacher of electronics, I try to develop
manual skills in my students by
having them solder printed circuit
boards, and these young adults usu-
ally prove themselves to be quite
capable, if only given the chance. So I
entrusted my students with the as-
sembly of the 680-BSM memory
expansion circuit board.

The students needed to have the
proper tools for such delicate solder-
ing tasks. Panavice, Inc., by way of
Bert McCabe, their Executive Vice-
President, generously donated the
original "Panavice" with the Model 315
circuit board holder. This provided an
ideal method of holding the large
circuit board of the 680-BSM memory
circuit. The Ungar Princess soldering
iron with agold-plated micro-precision
solder tip proved to be well worth-
while in soldering the infinite number
of integrated circuit socket pins of the
RAM circuits. Even so, a magnifying
glass was diligently used to examine
each IC socket for solder bridges and
cold solder joints.

Upon the completion of the assem-
bly of the 680-BSM circuit board, the
students and I discovered a fault that
seems universal in the MPU computer
field. The engineers that write the
assembly manuals assume that the
assembler of the product possesses a
certain level of knowledge in computer
technology. Let me illustrate with the
680-B I/O port. The 680-B assembly
manual provides the assembler with a
variety of I/O port choices but offers
nothing to aid in the selection of the
proper I/O port. The 680-B assembly
manual states connector finger-test
points but does nothing to identify
which side or what end is considered a
starting point for pin counting.

As we at Union County Career
Center soon discovered, the 680-B
MPU computer alone is worthless,
unless one memorizes the entire ASCII
binary code. Programming instruc-
tions require alphanumerical designa-
tions that cannot be made without an
appropriate keyboard, so we procured
an inexpensive Radio Shack keyboard
and associated components, only to
learn that we must have a parallel
interface to use it. We stipulated a
cassette BASIC program in hopes that

there would be no need for an
additional interface. But, to our dis-
may, in order to use the cassette
program, we discovered that our 680-B
must also have the 680-B KCACR
interface. Should we wish to use the
Assembly language and editor or the
Editor programs included with the
680-BSM, we must then purchase a
tape reader and a parallel interface
circuit board. If not, these programs
will be useless to our 680-B system.
Presently, our 680-B computer just
sits there, doing absolutely nothing!

The important point is that we have
learned, the hard way, what is involved
in getting into microprocessor-based
computer programming. This has
been a valuable lesson for both
student and teacher. We have learned
that more is required than the basic
computer unit to run a program. We
now know that an ASCII keyboard and
its interface, as well as a CRT terminal
or printer, are essential. We also
learned that the KCACR interface is
needed to record any programs on
tape, and, primarily, we have learned
that there is no substitute for quality
in circuit boards and components, by
having Mike Jones at Computer Stores

Kevin Stewart solders RAM IC sockets to 680-BSM circuit board.

Computer Notes Jan/Feb 1 978

of the Carolinas test out the workabil-
ity of our computer with his periph-
erals in the store. We now know that
the student assembly of the 680-B and
the 680-B-BSM was 100 percent-
well, 99 percent (due to one bent lead
of an IC) perfect!

As the Electronics Teacher at Union
County Career Center, I must confess
to earlier doubts of my students'
capabilities. They deserve a great deal
of credit for operating the computer as
well as they did. We also thank Mike
Jones for his presentation to the
NCAEDS (North Carolina Association
Educational Data Systems), in which
the MITS project at Union County
Career Center was mentioned and
where the Reprints of Computer Notes
on Union County activities were so
well received. This level of micropro-
cessor-based computer activity has
previously been directed exclusively to

junior colleges and technical educa-
tion centers and is most unique at the
high school level.

There are but two more stops before
we at Union County Career Center can
program our 680-B MPU computer,
and they are the acquisition of a
KCACR interface and a terminal.
These will be the subjects of future
articles in Computer Notes. Perhaps
the obstacles we encounter and
overcome will serve as guides to the
upcoming generation of home com-
puter novices.

About The Author

James Gupton is a free-lance writer and an
electronics teacher at the Union County
Career Center in North Carolina.

For Sale

Monroe #326 Beta "Scientist" Pro-
grammable Calculator with Model
#392 Digital Tape Unit.

Tape drive is fully controllable by
program, permitting automatic read-in
of program overlays plus program-
mable storage/ retrieval of data. It is in
like-new condition, with full documen-
tation and fitted Attache' carrying
case.

Original co|st: $1300.
Asking priOe: $650.

Contact:

Gene Szymarjski

693 Rosedale Road

Princeton, N.J. 08540

(609)-924-8856

W^T^mij J

17 tfiM

r '4"<*iFir&£.n

/ ■. .Jill

■fjf tt'-r '« V«

'- Straws' '

Jeff Benton positions DIP switch on 680-BSM circuit board.

John Martin inserts integrated circuits into their respective sockets.

Computer Notes Jan/ Feb 1978

MITSTAItair™ CPU Modification

By Darrel Van Buer

Since its introduction, MITS®has had
various aspects of its computer design
criticized. One of the more severe
problems with the original design
concerns the circuitry used to gener-
ate the $1 and §2 clock signals for the
CPU chip. The MITS/Altair™ design
predated the availability of the 8224,
so a 74123 dual one-shot was used.
Problems with this circuitry have been
met with a variety of fixes, the most
unusual being to glue an aluminum
foil heat sink to the IC. Parasitic
Engineering Company offers a fix kit
based on a better quality one-shot.
Because I was interested in trying the
2.5 MHz and 3 MHz versions of the
8080 microprocessor, I studied the
board and the Intel data sheets to
learn how to substitute an 8224 clock
generator IC on the MITS board. While
it proved to be a rather complex
modification, it can be made in an
hour or two, as described here.

The modification involves the re-
moval of all the existing clock circuitry
and replacing it with the standard Intel
circuit and several components, to be
certain that the right signals are
available on the bus. The ready
latching circuit, however, was not
used, because this function was
already performed on the board and
would have increased the^ complexity
of the modification.

■■.'■' ':■ ■■■:■ ■

:&«

■- •! •

■*

2 3 4 5 6

Figure 1 . PC Land Cutting— Top Left Component Side

Before starting, you should care-
fully remove the CPU chip to a piece of
conductive foam to void any static
damage. The first procedure in the
modification is the removal of all
unwanted circuitry from the board.
Since none of these components will
be re-used, the main concern is that

% - >v * '

10 11

Figure 2. PC Land Cutting— Top Left Back Side

they be removed without damaging
the PC board or the plated holes. If
you have good de-soldering equip-
ment, such as solder suckers and IC
de-soldering tools, use them. If not,
the safest method for removing the
ICs, is to cut all the leads and then
unsolder each lead separately. The
parts to be removed are given later in
this article. If sockets were used for
the ICs, they must also be removed.-
All of these parts are located in the top
left corner of the board, between the
regulator and the CPU socket, and
above the large power bus running
across the middle of the board. All but
one of the parts in this area, capacitor
C2 — nearest the regulator, should be
removed. When this step has been
completed, the board should appear
as shown in Figure 1.

The second major step in the
modification is to cut through the PC
foil paths on the board in eleven
places. In making these cuts, remove
a small segment with a sharp knife or
scraper, taking care to avoid damaging
other parts of the board. Table 1
summarizes these cuts. The first six
cuts are also shown in Figure 1 , which
illustrates the top left area of the
board. The seventh cut is shown in
Figure 3. The four remaining cuts are
shown in Figure 2, which are on the
back of the board near the removed
parts.

Computer Notes Jan/Feb 1 978

i#«"

i *«*8v*aS.sas^fe**^

CUT 7 JUMPER 1 JUMPER 1

F/gure 3. Modifications to Top Right of Board

J7 C101 / " J2 J3 J4 I J6#J3 J2# J7
XTAU J5 J4 D101A

J J8

D101C

R101

Figure 4. Components and Jumpers — Top Left

The third major step is the addition
of the five parts listed in the Parts List.
Table 2 summarizes the locations for
these additions, and Figure 4 shows
the locations of these changes (IC Q 1
replaces IC Q in the same orientation).
C101 can be used to trim the crystal
frequency to high accuracy. If tuning

is not needed, a fixed capacitor in the
same range of values may be substi-
tuted. When mounting the trimmer,
note that several alternate holes can
be used to allow for size. Many
trimmer designs will require the
soldering of short pieces of wire to the
lugs before mounting to the board.

While not required, the use of a socket
for the 8224 is recommended.

The final step is the installation of
thirteen jumpers on the board. All of
the eight jumpers installed on the top
of the board run between plated-
through holes. Some of the holes are
those left by the removal of compo-
nents, and the others are extra holes
on the board for other connections.
The jumpers installed on the back of
the board have one or both connec-
tions made by wrapping the end of the
wire around the pin of an IC or socket
and then soldering it in place. The
locations of all jumpers are given in
Table 3, and Figure 3 shows the
location of jumper 1. Figure 4 shows
the positions for the remaining jump-
ers on the top side of the board. All
jumpers on the bottom of the board
are shown in Figure 5.

Figure 6 shows the finished conver-
sion. At this point, the CPU chip can
be returned to the board, and the
board can be re-installed. Trimming
the crystal frequency with capacitor
C101 is the only adjustment that may
need to be made. This adjustment is
not necessary, as the crystal will
generally oscillate within 0.1 % for any
setting. Critical adjustment can be
made with a high grade frequency
counter or by zero-beating with WWV
at 5 MHz or 10 MHz over a shortwave
radio. The 2 MHz frequency will vary
by as much as a few hundred Hertz, as
mainframe bus loading and instruc-
tion sequences change.

The amount of noise present on
many lines of the MITS/Altair bus is
proportional to the length of the bus
that the signal has traveled from its
source. I have graphically witnessed,
with an oscilloscope in my system,
noise becoming unacceptable after 10
inches of signal travel. You can cut all
noise levels in half by locating the
CPU card in the center of the occupied
part of the bus, since neither signal
travels as far from source to destina-
tion. My system has been running
reliably for more than six months with
these modifications.

Figure 7 gives the schematic dia-
gram for the modified clock circuit.
Note that the 18 MHz oscillator output
has been brought out as the CLOCK
signal. In the standard MITS/Altair
system design, this pin is only 2 MHz,
so a slight modification in the con-
version is needed if any cards in your
system require the latter frequency. To
supply <f>2 to this pin, omit jumper J5,
listed in Table 3. Alternate jumper J5
belongs on top of the board from the
hole for pin 10 of IC P (one end of old
J5) to the other hole for the upper end
of C5 (one end of jumper J6).

Computer Notes Jan/Feb 1 978

PARTS LIST

IC Q 1 - 8224 clock generator

XTAL' -18 MHz

C101 - 3 to 10pF trimmer

R101 -2.2k

D101 -1N914

UNPARTS LIST (parts to be

removed)

ICN

- 7406

ICP

-7404

ICQ

-74123

XTAL -2 MHz

C3-

.01

C4-

10 pF

C5-

100 pF

C6-

20 pF

R29

-470

R30

-470

R31

- (none)

R32

-470

R33

-470

R34

- (none)

R37

■1k

R38

■330

R39

-Ik

R40

■330

R41

-13k

R42-

■6.2k

R43-

■680

A. Traces on top of board in part
removal area

1. Trace from P-14 to upper resistor
array

2. Trace from P1 4 to + 5V bus

3. Trace from Q-5

4. Trace from Q-4 to N-1

5. Trace from Q-1 1 to Q-1 6

6. Trace from Q-2 to Q-1 6

B. Trace elsewhere on top of board

7. The upper of two traces which pass
under SC17, just to the right of
SC17

C. Traces on bottom of board in parts
removal area

8. Trace from A-12 (CPU) where it
passes N-8

9. Trace from Q-1 3 below R43

1 0. Trace from Q-1 to Q-1 1

1 1 . Trace from Q-2 to Q-3

NOTES: P-14 denotes pad for pin 14 of
IC P. R43 denotes pads left by removal
of R43. Other references are similar.

Table 1. PC Land Cutting

About The Author

Darrel Van Buer was awarded an M.S. in
Computer Science by Iowa State University in
1975. He is currently studying for his Ph.D. at
UCLA and has been involved in personal
computing since the introduction of the MITS
microcomputers.

■

!%"•:':'■ ;■■■?*:■: ■■ ■ "■':' '/% f r ~. • ■ ''■■'v"S>'-' : \ :':■ ' ■ '■ r'lsrrp -i

•£■:-: . . .• '. v •-■- ".'■'-• ; .'- '/h-*.*' ■ ' ' ' <■- <■ ■■ •- ■'" .. - ■ -•--»*-- . - > :'-

... . \ .... . . . ■ ■ ■ •_ ... ..

fi^A^iVi^^

■S^'-irli

tfj^ii-i.ryiri/sii

J13 J10 J11 J12 J13

Figure 5. Jumpers— Back of Board

J12 J11 J10

a»''ttc

ifi

i. :. ■■ :.

Part
ICQ 1

XTAL'

C101
D101

R101

Figure 6. Finished Conversion

First Location

Same position and orienta-
tion as IC Q

Same position as XTAL
(insulate case from board)
Lower pad of R37 or R38
Cathode (banded end)
toN-2
N-1 4

Second Location

Lower pad of R39, R40, or C3
Anode to lower pad of R43

Unmarked pad above and to the
right of N-1 4, the left two to the
left of CPU

NOTE: Positions are the same as for Table 1.

Table 2. Component Additions

Computer Notes Jan/ Feb 1 978

C1D1 XTAL'

3-lDpF 18 MHz

-^f Ifr

Figure 7. Schematic Diagram

Jumper

First End

Hole to right of cut 7

nearSC17

P-3

P-4

P-14

P-10

Top pad of C5

Lower pad of C5

Top pad of R43

P-1 2

Hole slightly to left

(top view) of SC9

Hole in + 5V bus between

SC8 and SC9

N-6

N-8

Second End

Hole above pin 1 1 , IC K (821 2)

Top pad of C4

Lower pad of R41

Upper of two holes located

between SC5 and SC6

N-9

Hole between R101 and CPU

N-1

Top pad of R32

Wraparound pin 12 of CPU

Wrap around pin 5 of IC Q'

Wrap around pin 1 6 of IC Q'

Wrap around pin 1 1 of IC Q'
Wrap around pin 1 of IC Q'

NOTE: Same as Table 1.

Demonstration
Program

By Ken Knecht

A nice computer system is always fun
to show to computer-less friends.
Unfortunately, most programs are a
bit complicated for simple demonstra-
tions. By the time the StarTrek rules
have been explained, most people
have usually lost interest or are totally
confused. Therefore, I wrote the
following program, which makes a
rather good demonstration and per-
mits others to run the program
themselves.

Very little detail about the program
is required, other than to mention that
the "#" following some of the variables
indicates that it is double precision. If
your BASIC does not support this
function, omit the "#"s from the
variables and change lines 165, 230,
260, and 290 to read "NOT OVER 8
DIGITS" rather than "NOT OVER 16
DIGITS". Other than this change, the
program should run in any BASIC.

The program is very simple. It
consists of a string variable to
remember the user's name, some
simple math problems, counting let-
ters in a string, and the tried and true
"Guess the Number" game. The latter
seems to be the most popular.

In any event, when the program is
finished, you'll have fun answering the
many questions.

Program on page 32

About The Author

Ken Knecht currently heads his own company
called Kencom Corp. in Arizona. In addition to
this, he freelances as an author, computer
programmer, broadcast engineer, and tele-
vision system consultant.

Table 3. Jumper Locations

©VOLK

Computer Notes Jan/ Feb 1 978

A BASIC Memory Test

By Dave Culbertson

Rather than test a newly built mem-
ory board for your computer by
using the supplied memory test
routine written in machine or assem-
bly language, have you often simply
loaded BASIC, loaded a program, and
hoped that it would work? Many of us
are guilty of this. However, the
problem can be solved by using this
simple program that is visual in its
manner of testing your memory. The
program does not run at blinding
speed, but it continually displays what
is developing, which should help
computer users who do not under-
stand machine or assembly language
programming.

This program offers two unique
capabilities. Firstly, the program does
not stop when it encounters an error.
It simply prints the address and the
error. Secondly, it is possible to test
an address where no memory exists.
Again, the address and the error will
be printed.

I have written this program in MITS®
4.0 BASIC, but it can be modified for
other BASIC languages. Only standard
BASIC commands have been selected.
One should understand where memory
usage occurs in the computer before
using this program (Table 1). It is
assumed that you presently have 8K
bytes of memory operational and that
you have just completed a new 8K byte
board that you would like to test.
First, complete all of the manufac-
turers' recommended electrical tests.
If you do not have the equipment, a
local school or computer store should
be able to do these tests for you.
Next, use this program to test the
operational mode of your memory
board .

Since you only have 8K bytes of
operational memory, you will normally
be strapping the memory board for
address 8192. But, Irf this case,
temporarily strap the board to a higher
address, such as 24576. This would be
the sixth such 4Kbyte address assign-
ment. Remember to restrap the board
to the proper address when your test
is complete. This program will fit
within your 8Kbyte board, along with
the 8K BASIC and the stack. It is
necessary to strap the memory board
to a non-consecutive address, be-
cause the stack will move to the end of
your new memory board if you do not
separate the memory. If this happens,

you will not be able to test this area.
To attempt a test of this type will
disturb the BASIC interpreter, and this
will prevent some, or all, of the BASIC
commands from performing their nor-
mal functions. If the program crashes
as just described, you will need to
reload BASIC and reload the program.

Once you have strapped your mem-
ory to address 24576 and are ready to
begin execution, run the program.
You will be asked for the starting and
the finishing locations of the area to
test. The computer will then ask for
the complete or partial test. I suggest
that you run the partial test first, since
this runs faster. However, it only
checks to see if one number can be
written into each location. Try enter-
ing the test word #0 (zero). If all is
well, the computer will print out the
address and the contents of this
address. If there is an error, the
program will print the address, the
test word #, the resulting #, and the
word "ERROR". The original contents
of this address will be restored to this
location.

The test will continue in this manner
until complete. I use a teletype to
retain a hard copy record of the errors.
The good locations are printed on my
video terminal to save paper. If you do
not have two terminals, change line
#150 to read "150 REM" and change
line #170 to read "170 REM". These
changes will print all data on one
terminal. It is necessary to use the
partial test with the test word #0 (zero)
to determine if the new board will
accept all low inputs to be written into
each location.

Next, rerun the test using the partial
mode and test word #255. This is the
reverse condition that tests for all high
input (all 1's) to be written into
memory. If you have an error and want
a complete analysis, run the pro-
gram using the complete mode. The
program will try to write all combina-
tions (0 to 255) into each memory
location. The complete routine takes
about four seconds per location.
Errors are shown as they are in the
partial test, except the address will
only be printed once if an error is
found. The bad combinations of the
location will be printed with the
resulting word #. The complete analy-
sis may appear confusing initially, but
you should be able to analyze the area

of trouble by comparing your result
with the knowledge of your memory
board .

Many types of memory boards are
available, so the results of this test
may vary. It is important to know the
organization of the memory chips in
your new memory board. Some boards
use an organization of 1,000 times 1,
which means that the chip has 1 ,000
locations (addresses) that can store a
"0" (low) or a "1" (high) in each of
these locations. This type of board
will require 32 memory chips in order
to provide 8K bytes of storage. This is
a popular method presently used on
static memory boards.

The dynamic memory boards now
use fewer chips, since more locations
have been put into each chip. If you
have a 4K byte dynamic board, you
will find only eight chips. The popular
organization among this type of chip
is 4,096 times 1. The memory boards
store the information into the chips by
assigning a value to each of the eight
chips per address that are used. When
an address is selected and you would
like to read the information at that
location, the output is from these
eight chips. If one or more of these
chips is defective or if a location
within any of the chips is bad, an error
will be printed at this address when
this program is run. The value of each
of the eight chips is shown in Table 2.
Assume that the storage of informa-
tion within each chip has something
(1 or high) or nothing (0 or low) as its
only variations. If the chip has
something (1 or high), add its value
(Table 2) to the other chips with some-
thing (1 or high) in them. Thus, when
an address is read, you are really
seeing the combined output of eight
chips. If one of these chips having a
large organization is defective, many
addresses will be affected. For exam-
ple, if zeros (0) are stored into chips
through 6 and if ones (1) are stored
into chip #7, use only the chip #7 value
as the contents of this memory
location. The print-out would be 128.
If zero (0) is stored into chips 1 ,3,4,5,
and 7 and if one (1 ) is stored into chips
0,2, and 6, use only the values of chips
#0 = 1, #2 = 4, and #6 = 64. Add these
values (1 +4 + 64 = 69). The number 69
has been stored at this location. This
is the method used to store numbers
in memory of an eight-chip memory

Computer Notes Jan/ Feb 1 978

Address Description

to 6457 This area is used for the BASIC

interpreter.
6457 to 7676 This area is used for the "BASIC

MEMORY TEST" program.
7677 to 8192 This area is the overhead and stack

area.

This table assumes a MITS/Altair™
computer with 8K consecutive
memory.

Memory Chip #

Decimal Value

128

This table assumes 8 memory chips
used per bank.

Table 1.

Table 2.

bank. Using this method, the comput-
er permits any number from to 255 to
be stored. The computer is not able to
use a larger number.

If the program has been run, you are
now ready to analyze the errors
provided for you by the program. It is
assumed that the partial program was
run and that the "0" (zero) test word
was satisfactory. The error occurred
when the "255" test word was run, so
one or more of the eight chips must be
bad. Assume that the error occurred at
address 24576 and continued until the
program reached address 25576. Also
assume that, in all of these addresses,
the program showed 191 when you
tried to write the test word "255". If
191 is subtracted from 255, the result
is the number 64, the value of chip #6.
It is now known that chip #6 is either
dead or that it is not receiving the
correct voltages or signals from the

board. A complex problem has been
resolved by locating the area of the
trouble. Even if you decide not to
repair the board yourself, the repair
technician's time, trouble, and ex-
pense will be lowered.

Every program has its limitations,
and this program is no exception. If
your memory board has addressing
difficulties, three problems may oc-
cur. Firstly, this program may not
detect any error. Secondly, your
BASIC may bomb when you run your
regular program; or, thirdly, random
changes may be detected in your
regular program. This last condition
could also be due to a memory chip
that malfunctions intermittently. This
is the most difficult problem to
find— in which case, happy hunting!

The given program run shows an
actual problem. In this example, I

removed chip #6 from my memory
board and performed the BASIC
MEMORY TEST on it. The board will
not pass the partial test with the word
#0 (zero), but it will pass the partial
test with the word #255. When the
complete test program is run, it
indicates the number combinations
that cannot be written properly into
the memory. The decimal combina-
tions that read correctly are not
printed out in the complete test. If you
suspect the memory as the cause of a
problem you are having, this type of
program can save a great deal of time
and trouble.

Program on page 33

About The Author

Dave Culbertson graduated from the Spring-
field Technical Institute and is currently the
Vice-President of Custom Electronics, Inc. in
Massachusetts.

FDOS-III: The Latest from Pertec
Computer Corporation

FDOS-III, a powerful new Floppy Disk
Operating System for microcomput-
ers, is one of Pertec Computer
Corporation's newest additions to
their iCOM® product line.

FDOS-III offers the maximum in
flexibility and power with its relo-
catable assembler for Z-80 and 8080
code. All its console communications
are either in decimal or hex, thereby
simplifying program development.
The "BATCH" command allows auto-
matic chain operations, and the
system includes an optional operator
prompt feature for variable input
requirements. Data is stored and

recognized by FDOS-III, and it can use
all available disk storage capacity.

The new FDOS-III is available for
any iCOM Floppy Disk System operat-
ing on the 8080 or the Z-80. The
FDOS-III is fully compatible with
programs written under iCOM's
FDOS-II and allows immediate use of
any existing iCOM-compatible pro-
grams. The single command opera-
tions of FDOS-III give the user
disk-to-disk program editing and
assembling, disk-to-memory program
loading, disk-to-punch device trans-
fer, reader-to-disk transfer, disk-to-
disk transfer, named files, and many

other features.

FDOS-III also has relocatable driver
modules that provide easy access to
files, thus maximizing data handling
flexibility. The storage area on each
diskette is available for any number of
files of lengths ranging from a single
sector to an entire diskette. The files
may contain program source data,
program object data, or user-
generated data.

Files are specified by a 1—5
character file name, and any number
of files may be merged to create a new
file. Any file may be renamed or may
be deleted (FDOS repacks the disk-

Computer Notes Jan/Feb 1 978

Tic Tac Toe Modification

By John Trautschold

ettes automatically at the operator's
option to make the deleted file space
available). Also, files may be tagged
with attributes (i.e., a file may be
declared permanent, not allowing it to
be inadvertently deleted).

The resident FDOS-III is conve-
niently contained in a 1K PROM
located on the plug-in interface card.
The FDOS-III also contains its own
powerful disk-resident assembler and
editor. The microcomputer's monitor
remains intact, thus retaining all
existing non-FDOS operations.
A typical edit/assembly sequence
requires only a few minutes to
accomplish, and a string-oriented text
editor greatly simplifies file or pro-
gram modification.

"FDOS-III provides one of the most
powerful and complete development
packages available anywhere," claims
T. E. ("Gene") Smith, Division Vice-
President and General Manager of
PCC's Microsystems Division. "When
used with any of iCOM's family of
Floppy Disk Systems and compatible
plug-in interfaces, FDOS-III provides
an easy-to-use, reliable, fast, and
extremely efficient capability for aux-
iliary program and data storage.

"Using the iCOM program develop-
ment package, time is reduced by a
factor of 20 to 100 compared to
cassette or teletype. In sum, FDOS-III,
together with iCOM floppies, brings
new speed, convenience, and capa-
bility to users' development tasks,"
Smith stated.

Commands available with FDOS-III
include Copy, Alloc, Batch, Delet,
Pack, Delpk (Delet and Pack functions
in a single command), Edit, View,
List, Libo, Dump, Load, Merge, Print,
Renam, Run, Link, and Exit.

Also included are two new com-
mands, ASMBand SYSGN. ASMB, in
Z-80 or 8080 code, assembles the
contents of a source file and directs
the object output to the destination
file. SYSGN allows the user to store
I/O information in sectors on a system
diskette for use by FDOS-III, thus
minimizing the effort needed to bring
FDOS-III up on a custom-configured
machine.

FDOS-III is being marketed as part
of the PCC Microsystems Division's
iCOM Microperipherals® product line.
It is available from any of the more
than 70 iCOM dealerships
nation-wide.

The "Tic Tac Toe" software articles
from the August 1977 edition of
Computer Notes was very interest-
ing— and frustrating, to say the least!
When I loaded the BASIC program, I
discovered that, no matter how hard I
tried, I could not beat the computer!
The program was written in such a way
as to make the computer unbeatable;
the best that could be achieved was a
tie (as was mentioned in the article).
Even if the program could have been
beaten, there was no logic included
permitting the program to jump to the
"Player Wins" subroutine at line
number 1220. I have recently made
some modifications to correct this
problem as well as some that now
make it possible, but still difficult, to
win.

I have eliminated the lines that
establish the initial move for the

190D = INT(RND(1)*10/3)

191 IFD = 0ORD>3GOTO190

192 E = INT(RND(2)*10/3)

193 IF E = 0ORE>3GOTO192
i94IFC(D,E) = 0THENC(D,E,) = 3:

C$(D,E) = "C":GOTO210
195IFC(D,E)<>0 THEN 190
225 GOTO 1500

1500IFC(1,1) = 1 ANDC(1,2) = 1 AND

1510IFC(1,1) = 1ANDC(2,2) = 1 AND

1520IFC(1,1) = 1 ANDC(2,1) = 1 AND

1530IFC(1,2) = 1 ANDC(2,2) = 1 AND

1540IFC(2,1) = 1 ANDC(2,2) = 1 AND

1550IFC(1,3) = 1 ANDC(2,2) = 1 AND

1560IFC(1,3) = 1 ANDC(2,3) = 1 AND

1570IFC(3,1) = 1 ANDC(3,2) = 1 AND
1580 GOTO 230

This concludes the modifications to
the program. I hope that others will
enjoy this program as I have.

computer, because these always de-
faulted the computer to start in the
center square (which is nearly unbeat-
able as an initial move), as well as in
square 1 —3 (upper right square on the
board). To replace these eliminated
lines (190 and 200), I have written a
random number subroutine that ran-
domly places the computer's first
move in an empty square. After the
first random move, the other moves
follow according to the programmer's
logic. To repair the problem of having
no logic to determine if the player has
won, I have added a complete sub-
routine that is called in the new line
number 225. Line 1500 is the location
of the subroutine.

The following is a list of the new
lines to be inserted into the program
for proper operation:

C(1,3)
C(3,3)
C(3,1)
C(3,2)
C(2,3)
C(3,1)
C(3,3)
C(3,3)

1 GOT0 1 220
1 GOT0 1220
1 GOT0 1220
1 GOT0 1 220
1 GOT0 1220
1 GOTO 1 220
1 GOTO 1220
1 GOTO 1220

About The Author

John Trautschold has worked for five years in
television electronics and engineering. He
received his engineering degree from the
University of Wisconsin in Milwaukee, and he
enjoys working with computers both at home
and on the job. It has been three years since
he first acquired his MITS/Altair 8800.

Computer Notes Jan/Feb 1 978

Practical Programming, Part II

By Gary Runyan

This series is produced by the MITS®
Computing Services Department, and the
articles contain useful ideas for programming
MITS BASIC. "Practical Progamming, Parti"
appeared in the November 1977 issue of
Computer Notes, and it discussed the solution
to the problem of line counting.

CTRL-A, a feature of MITS®BASIC,
has become a powerful programming
aid, due to an undocumented feature
discovered by Donald Fitchhorn of
MITS. If CTRL-A is typed immediately
after EDITing a program line, the
edited line is returned as a command
to be edited. Thus, CTRL-A can be
used to shuffle program lines, break
apart multiple statement lines, and
isolate program errors.

A program line can be shuffled from
one place in the program to another by
typing the following sequence:

1 ) EDIT xxxx<CR>(xxxx = old line #)
2)Q

3) CTRL-A

4) lyyyy<CR> (yyyy = new line #)

5) xxxx<CR>

This moves the line that was at xxxx
to line yyyy and deletes line xxxx. The
original line can be retained by not
executing Step 5. If a new line is
needed that is slightly different from
the old line, ESCAPE can be typed in
place of CR (carriage return) as the
last character in Step 4. The editor can
then be used to modify the line before
placing it at yyyy.

CTRL-A can be used to break a
multi-statement program line into two
program lines without retyping either
of the new lines. A copy of the original
line is made using the above proce-
dures for copying a line. Then, the K
EDIT command is used to remove the
first half of the line from one copy,
and the H EDIT command is used to
remove the second half of the line
from the other copy. For example, to
change:

600LPRINTA:PRINTB
to:

600 LPRINTA:IFX<0THENGOSUB500
605 PRINTB
one would type:
EDIT600<CR>
Q

CTRL-A

I605<ESC>2KP<CR>

EDIT600<CR>

2SPHIFX<0THEN GOSUB500<CR>

To isolate a syntax error that has
been encountered while a program is
running, type a Q to exit EDITing
without losing the program variables.
Typing CTRL-A then restores the
program line for execution as a
command. The command line can be
modified at will without destroying all
the program variables and then exe-
cuted to test the modifications. Co-
lons can be replaced by single quotes
in a multiple statement line to isolate
the statement with the syntax error.
Obvious errors can be corrected and
tested immediately. For example, if
the line:

50A = 5:PRINTA:A = A + #7:PRINTA

is encountered while a program is
running, the following will isolate the
error, correct it, and continue the

program:

SYNTAX ERROR IN 50

50 (Type: Q)

(Type: CTRL-A)

!(Type:S:C'<CR>)

I (Type: CTRL-A)

!(Type:S'C:S:C'<CR>)

(Type: CTRL-A)

! (Type: S ? C:S:C' <CR>)

SYNTAX ERROR

(Type: CTRL-A)

I (Type: S#C3S'C: <CR>)

(Type: GOTO60)

After the syntax error is success-
fully corrected, one executes a GOTO
command (if the corrected line did not
branch back in) to continue program
execution. Continuing after correcting
is a good habit to adopt. Other bugs
are found without completely rerun-
ning the program. If variable values are
clobbered before an error is success-

fully corrected, the programmer must
decide if it is better to rerun or to
restore values (using direct com-
mands) before continuing with a
GOTO.

CTRL-A can be used to isolate
ILLEGAL FUNCTION CALL, TYPE
MISMATCH, and other errors, as well
as syntax errors. One simply types:

EDIT [number of line in question]

CNTRL-A

to gain control of the line in question.

Once a programmer begins using
CTRL-A after exit from EDIT, he will
find that his whole set toward debug-
ging has changed. Rather than follow-
ing the old batch system approach of
guessing corrections from the listing
and rerunning, one will begin using
the computer to resolve the bugs.
Time lost to and ulcers caused by
debugging will be considerably re-
duced.

Initially, in the joy of a new-found
tool, you will use the "don't-leave-the-
terminal-until-the-bug-is-resolved" ap-
proach in excess. Eventually, after
several wild goose chases, you will
begin to discriminate between when to
sit back and really study the listing
and when to poke-around on the
terminal.

Some additional poke-around hints
are:

1. Use CTRL-A to execute lines that
print the values of variables.

2. Tack lines that will print variable
values onto the end of the program
to save constant retyping.

3. Edit extra STOPs into the program
to establish poke around points.

4. Edit in extra PRINT commands to
monitor the evolution of variable
values.

5. UseTRONandTROFF.

6. Edit in GOTOs to skip around
undesired outputting, or go
directly to problem areas.

About The Author

Gary Runyan is the Director of Computing
Services and has been a MITS employee for
three years. He has worked in the data
processing field for six years, and he holds a
Bachelor's degree in Electrical Engineering
from New Mexico State University.

Computer Notes Jan/ Feb 1 978

KNOW
THE

USER

Balding spot, from scratching head in bewilderment.

Abstracted expression, often obscured by
thick-lensed glasses, a result of endless
debugging.

■ Does not remove tie after work.

9 pencils, 2 mini-screwdrivers (standard and phillips),
small slide rule, and pocket calculator.

Clothing often reeks of solder.

Ring, to remind one of one's home and
family (now and then).

Thin, trembling limbs, indicative of neglect
to eat.

The advanced user often seems out of touch with
his surroundings and unable to speak about any-
thing other than his own system.

If you know a user, or have one in your family,
contact the address below. It may not help . . .
but it couldn't hurt.

PERTEC COMPUTER CORPORATION

20630 Nordhoff St.
Chatsworth, CA 91311

Please send me a subscription to Computer Notes.

□ $2.50/year □ $5.00/2 years □ $1 0.00/year for overseas

State:

Zip:

COMPANY/ORGANIZATION .
CI Check Enclosed

Computer Notes Jan/Feb 1 978

Modifying MITS® BASIC for ASCII I/O

By John Palmer

I am a certified electronics technician,
but most of my past experience has
been in radio and television. Conse-
quently, my M ITS®/ Altai r™ 8800 has
been an exciting challenge, and my
efforts have been aided by the
information in Computer Notes. Many
of the CN readers' comments indicate
that there are always newcomers to
the trade looking for information on
how to do elementary tasks, such as
making ASCII recordings on cassette,
so permit me to relate to you a few
pointers that I have learned. On
On modifying MITS BASIC for
ASCII I/O:
Hardware: 8800. with 16K, ACR, 2SI0,

and Model 33 TTY
Software: MITS 8K BASIC, Version

4.0, January 1977

I thought I might be the last person
to learn to enter BASIC'S I/O and to
change a few memory locations to
permit an ASCII program listing
(source code) to be either output or
input on a storage device other than
paper tape (a paper tape punch /reader
tends to be very expensive!).

Several problems that users of MITS
BASIC might encounter can be solved
by modifying the I/O routines in such
a way that the ACR cassette interface
replaces the terminal. If the user has
Extended BASIC, the console feature
will transfer I/O to the cassette.
Questions will arise if the program
was written in some version of BASIC
that does not have the console
command.

The following describes how and
why I make ASCII recordings in MITS
BASIC. A very simple batch to 8K
BASIC, Version 4.0, will place ASCII
characters onto the cassette when I
use my Model 33 teletype. After
loading the program, using CLOAD, I
then type:

POKE 1 362,21 1 :POKE 1 363,7

Upon hitting the return key, BASIC
then pokes these two locations in the
output routine, and what goes to the
teletype printer will go to the ACR.
The MITS Software Library has more
information on how to do both input
and output using the ACR and 4K
BASIC (which has no provision for
CSAVE and CLOAD).

But this simple method is only for
8K BASIC, Version 4.0. Before trying
this, be sure you have the same

Version of BASIC. Either the locations
are different, or there are not several
empty locations in the output routine.
Note that those two locations are
needed for the MITS 4P10 board, but
not needed otherwise.

Doing input is slightly more in-
volved, but there are three reasons for
troubling yourself.

1 . The output of one program may be
needed as input to another.

2. Cassette input will transfer a
program from one version of
BASIC to another. For example, if
you key in StarTrek in 8K BASIC,
you will find that you cannot load
it into the current version of
Extended BASIC.

3. Some types of errors due to poor
recording can best be corrected by
making a new recording in ASCII
and then using the new recording
as input. BASIC will put all lines in
proper order, provided that the
input speed is not too fast (put in
nulls, just to be sure).

To input an ASCII recording from
the cassette, one must either use the
POKE command or must stop the
microcomputer and alter memory lo-
cations with the front panel controls.

I am presenting a partial listing of
the routines that are used in MITS
8K BASIC, Version 4.0. for terminal
input and output. Note that output is
first.

The output precedes the input,
because both are 'called' routines that
are called from somewhere inside
BASIC. Furthermore, the front panel
will produce the same result as the
POKE command.

The following commands will trans-
fer input to the MITS ACR interface
using 8K BASIC, Version 4.0:

POKE 1367,6:POKE 1370,194:
POKE 1374,7

Before hitting RETURN, begin play-
back of a recording that was made in
ASCII mode.

To have BASIC return control to the
keyboard, either use the front panel to
restore the original input routine or
play a tape that was previously
prepared. Here is how to prepare the
f change-over' tape.

1 . POKE the two empty locations in
8K, Version 4.0, as shown earlier in
the article.

2. Type: NULL 3

3. Put a spare tape into the cassette
and begin recording (allow 15 sec-
onds for the leader).

4. Hit RETURN two or three times.

5. Type: POKE 1367,16:

POKE 1 370,202:POKE 1 374,1 7

6. Hit RETURN severaftimes.

If a typing mistake is made, you
must begin again.

The cassette will now have the
instructions needed to restore control

Split-Octal

OCTAL

Code or

Changes Needed for

Address

Data

Purpose

ACR Cassette Interface

Here is part of the output:

005 116

361

POP PSW

005 117

323

OUT

005 120

021

Data Port

005 1 21

365

Push PSW

005 1 22

000

NOP

323

005 1 23

000

NOP

007

005 124

361

Pop PSW

005 125

311

Return

Next is the in

put:

005 1 26

333

005 127

020

Status Port

006

005 130

346

ANI

005 131

001

MASK BIT

005 132

312

302 JNZ

005 133

126

STARTING

005 134

005

ADDRESS

005 136

021

DATA PORT

007

005 137

346

ANI

Computer Notes Jan/ Feb 1 978

to your keyboard (this is for a
keyboard that uses the MITS 2SI0 I/O
interface). Set aside your 'change-
over' tape.

When recording, it is good practice
to use at least three nulls to prevent
the tape from advancing ahead of
BASIC (when playing a tape, use
NULLO).

If you wish to merge two programs,
be sure that the two programs have
different line numbers. First, input the
program with lower line numbers.
Otherwise, BASIC must do too much
housekeeping, and it will fall behind.
MITS BASIC presently has no provi-
sion for merging files or programs
using CSAVE and CLOAD, and the use
of CSAVE and CLOAD is much faster
than ASCII.

When using a poor quality tape, a
line number may become garbled.
When loaded into 8K BASIC, such a
recording may cause trouble. The
following illustrates this:

LIST
10REM

Where is line 57 from? Why is it at
the end of the program? And why does
it repeat on and on and on ?

Any attempt to erase line 57 will
prove to be futile. Aside from peeking
inside the program buffer and trying to
erase the bad number, the only way to
cure this program is to dump it as an
ASCII listing.

To make an ASCII recording of an
existing program, do the following:

1 . POKE locations 1 362 and 1 363 with
211 and 7.

2. NULL 3.

3. Type the following (don't hit
RETURN yet):

PRINT:PRINT:PRINT:LIST

4. Start the recorder, type a few
spaces, and hit RETURN.

While BASIC is listing the program
on the printer, the same ASCII
characters are being recorded on
cassette. A standard teletype runs at
110 baud, yet the ACR interface is
normally 300 baud, which presents no

970 A = A + 4

980 IF A 12 THEN 450

999 END

57 &NH SJ%FORBV MID$= :A15,DLRO F.EIFM93

real problem. The cassette will have
some verrry loooong stop bits, but it
will playback adequately.

Be sure to leave a long leader at the
start and the end to prevent 'garbage'
from being fed into your computer's
input routine. The spaces and nulls at
the beginning will purge the ACR
buffer. First, play the tape, and, when
the spaces begin, start your computer
input. If you have not already done so,
it is advised to modify the cassette
machine to hear the playback while
the patch cord is in place. Try 47 ohms
across the mini-jack contacts.

Forthe more experienced, all of this
may be very elementary. But, for those
users like myself, I hope I have been of
some help. Incidentally, "NEW" may
be used when you don't want to merge
programs.

About The Author

Patrick Delaney is currently working as an
instructor of Digital Electronics at the Rhode
Island School of Electronics. He graduated
from the University of Rhode Island in 1970
with aB.S.E.E. and is now developing tutorial
programs for the M ITS lAltair 8800 computer.

MITS^ Newest Business System

The MITS® 300 Business System is
one of Pertec Computer Corporation's
major additions to their already exten-
sive product line.

The MITS 300 is a microcomputer-
based system that is complete with all
necessary hardware and software. It is
available in two configurations, one
with a hard disk (the MITS 300/55) and
the other using two floppy disks (the
MITS 300/25). The fully integrated
business system provides capabilities
for word processing, inventory con-
trol, and accounting functions, which
include a general ledger, accounts
payable, accounts receivable, and
payroll.

"Customers now can buy a totally
integrated system from a single
supplier," says T.E. Smith, Division
Vice-President and General Manager.
"We provide both hardware and soft-

ware and can assume responsibility
for the entire system. Also, service
facilities are available through the
PCC Service Division. And we are able
to provide extensive dealer support in
installing and starting up each
application."

Both configurations of the MITS
300 Business System incorporate a
MITS/Altair™ 880b turn-key main-
frame with 64K of Dynamic RAM, 1K
of PROM, and serial input/output
interface. Also included is a MITS/
Altair B-100 CRT terminal with a
12-inch, non-glare monitor. The CRT
displays 24 lines with 80 characters
per line and has a memory page of
1920 characters. The MITS/Altair
C-700 line printer, which is also part of
the basic configuration, is capable of
a bi-directional operation that allows
the printhead horizontal movement for

seeking the nearest margin of the next
line. The C-700 prints 60 characters
per second and 26 lines per minute.
Each configuration, comprised of
the mainframe, a CRT terminal, and a
line printer, also includes either a hard
disk or two floppy disks, a controller,
and BASIC language software. A
MITS/Altair A08 Accounting Package
and an Inventory Management Soft-
ware Package, although not included,
are available both with the hard and
the floppy disk systems at additional
costs.

The MITS 300 Business System is
being marketed as part of PCC's MITS
product line. It is available at the more
than 40 MITS Computer Centers
across the continent and by way of
PCC's Microsystems Division directly
on an OEM basis.

Computer Notes Jan/Feb

PERTEC COMPUTER CORPORATION'S new MITS 300 Business System is comprised of a
mainframe, a CRT terminal on a desk, and a line printer on a pedestal. Pictured here is the
MITS 300/55, which is the hard disk, rather than the floppy disk, system.

Favors

Captain Charles P. Connolly is a new
MITS® U ser and would like to ask for
your help. He is interested in contact-
ing anyone using BASIC to solve
substitution cryptograms. He would
be particularly interested if MITS
BASIC is being used, but any BASIC
without MAT statements will do
nicely. Please write to Capt. Connolly
at the following address:

2701 Park Center Drive
Apt. B-501
Alexandria, Va. 22302

Book Review

Presented here is a review of Dr. C. William
Engel's recently published book entitled
Stimulating Simulations. The small paperback
book is written in MITS®8K BASIC 3.2 [the
programs will also work with all higher
versions of BASIC] and contains ten rather
unusual simulations written for the enjoyment
of the computer hobbyist.
• Dr. Engel is a Professor of Mathematics
Education at the University of South Florida in
Tampa. His book sells for $5 per copy and $3
each for orders of ten or more. Send orders,
comments, or questions to:

Dr. C. William Engel

P.O. Box 16612

Tampa, Florida 33687

A Review

STIMULATING SIMULATIONS:

Ten Unique Programs in BASIC

The excitement of deep sea fishing,
the intrigue of a jewel robbery, and the
challenge of piloting a space ship on a
mercy mission are three of ten
simulations you can experience with
your computer. The interaction be-
tween computer and player is a
challenging one that forces the player
to make logical decisions in order to
succeed or, sometimes, survive.

These ten simulations can be found
in a ciearly-written, well-documented,
64-page book called Stimulating Sim-
ulations. Although the ideas are fairly
sophisticated, the programs are rela-
tively short (from 40 to 100 lines of
BASIC). Each program includes a

scenario, a sample run, a flowchart,
a listing of the variables, and sug-
gested modifications.

This book is a good starting point
for the computer hobbyist who wishes
to explore the use of the small
computer in simulating real events. A
brief description of each program is
given below.

"Art Auction" (48 lines)
One buys and sells paintings to
make a maximum profit. This is a
fast simulation and does not require
extra materials.

"Monster Chase" (48 lines)
A monster is chasing a victim in a
cage. The victim must elude the
monster for ten moves to survive.
This is a fairly quick simulation that
does not require too much thought.

"Lost Treasure" (74 lines)
A map of an island that contains
treasure is presented. The adven-
turer travels over different terrain
with a compass that is not very
accurate in an attempt to find the
treasure. This is a short simulation
that requires about fifteen moves. A
map is provided.

"Gone Fishing" (83 lines)
The object is to catch a large num-
ber of fish during a fishing trip. Half
of the catch spoils if the time limit
is exceeded, or if time is lost in a
storm. In addition, the boat sinks if
it is guided off the map. There are
also sea gulls and sharks to avoid. A
chart is needed to keep track of
good fishing spots.

"Space Flight" (68 lines)
The task is to deliver medical
supplies to a distant planet while

trying to stay on course without run-
ning out of fuel. Graph paper is re-
quired to plot the course.

"Forest Fire" (77 lines)
The object is to subdue a forest fire
with chemicals and backfires. Be-
cause the output is a 9X9 grid, a fast
baud rate to the terminal is desir-
able. The success of a firefighter is
based on the time needed to control
the fire and completely extinguish
it.

"Nautical Navigation" (70 lines)
This simulation requires the naviga-
tion of a sailboat to three different
islands, using a radio direction
finder. The wind direction is an
important variable. Graph paper,
protractor, and ruler are needed to
plot the course.

"Business Management" (92 lines)
In this simulation, raw materials are
bought, and finished products are
produced and sold. The cost of
materials and production and the
selling price vary each month. The
objective is to maximize the profits.
No extra materials are required.

"Rare Birds" (75 lines)
This is a bird watching simulation.
The object is to identify as many dif-
erent birds as possible. A record of
those identified is helpful, and a
bird-watching chart is provided.

"Diamond Thief" (83 lines)
One assumes the role of a detective
in this simulation. A thief has just
stolen a diamond from a museum.
Five suspects must be questioned
to determine the thief. A floor plan
of the museum and a chart indicat-
ing suspects and times are provided.

Computer Notes Jan/Feb 1 978

If you need real results from
your 8080 or 6800 based system

Then scan this
list of topics . . .

■ binary arithmetic

■ logical operations

■ organization of a computer

■ referencing memory

■ carry and overflow

■ multiple precision arithmetic

■ loops

■ shifting

■ software multiplication and
division

■ number scaling

■ floating point arithmetic

■ stack pointer usage

■ subroutines

■ table and array handling

■ number base conversions

■ BCD arithmetic

■ trigonometry

■ random number generation

■ programming of the 6820 PIA

■ programmed input/output

■ control of complex
peripherals

■ programming with interrupts

■ a software time of day clock

■ multiple interval timers in
software

■ data transmission under
interrupt control

■ polling

■ debugging techniques

■ patching a binary program

■ full source listing of a debug
program . . .

Order now . . . Start getting

real results from your

8080 or 6800 based systems.

Every one of these topics and many, many more are discussed
in the Practical Microcomputer Programming books. In chapter
after chapter and scores of formal program examples, the basic
skills of assembly language programming are developed step
by step. The examples are real and have been tested and
proven. They run, and more important, they teach. If you're
tired of generalities, reproductions of manufacturers data
sheets and books with examples that don't run, then there is
only one place to go, the Practical Microcomputer Program-
ming series from Northern Technology Books. At $21.95 each
they are the best bargain in programming information available
anywhere.

Northern Technology Books Box 62, Evanston, IL 60204

□ Practical Microcomputer Programming: The Intel 8080 $21.95

□ Practical Microcomputer Programming: The 6800 $21.95

□ check on US bank enclosed D money order enclosed

Illinois residents add $1.10 state sales tax.
Foreign orders add air mail postage if desired (.8 kg).
Please type or print

Name

Company,

Address

City

State_

Prepaid orders only

Zip-

Computer Notes Jan/Feb1978

— - — „__.. ... ....... . ... , .... , .....,„- - — _-„

<W>'*'"»-

10,000 Visit MINI/MICRO '77

By Marsha Sutton

The 1977 MINI/MICRO trade show
was held in Anaheim, California on
December 6—8, and the organizers
say it was quite a success. Total
attendance for the show was 9,917,
falling just short of the projected
10,000. The attendance figure includes
300 booth personnel, representing
nearly 180 companies from across the
nation.

The show was open for three full
days, during which time guests could
view the exhibits as well as attend the
technical sessions. The 20 sessions
consisted of 90 speakers, and the
program presentations ranged from
formal papers to panel discussions.
Topics included such areas as how to
begin a new company, small business
systems, microcomputers to help save
energy, and trends in mini-micro
software, small disk memories, CRT
terminals, and printer development.

Organizers of the conference were
pleased with the guests at the show,
claiming many prominent visitors
from Japan, Canada, and several from
Europe. Included among the guests
was LEON RUSSELL, the popular rock
performer. He appeared on the second
day of the show, looking very conspic-
uous in his sunglasses and cowboy
hat. When asked if he intended to
purchase a home computer some day,
he replied that he already owns a small
system (although he would not reveal
the type). He did say that his
applications include bookkeeping and
synthesizing of music. He was not,
however, using his system for compo-
sition, amplification, or production of
sound effects, which are some of the
latest innovative musical applications
for microcomputers.

Pertec Computer Corporation ap-
peared at the show in full force with
two booths, one for the Microsystems
Division and the other representing
the Pertec Division. The Pertec
Division booth displayed magnetic
tape transports and fixed, cartridge,
and flexible disk drives in an attractive
booth design. The Microsystems Divi-
sion (MSD) booth was also an
impressive display of both MITS® and
iCOM® Products.

PCC's Microsystems Division pre-
sented several new products at the
Anaheim show, all aimed at enhancing
and supporting the existing product

line. One of these products is the
MITS 300, a microcomputer-based
integrated business system. The MITS
300 is available in two configurations,
both of which are supplied with
complete hardware and software.

Visitors at the MSD booth were also
introduced to iCOM's Attach^™

microcomputer. The Attache is a desk-
top computer that is built around the
8080 MPU. Its basic configuration
includes a CPU board, keyboard, video
board, and turnkey monitor board.

MSD has also recently introduced
the FDOS-III, which is iCOM's new
Floppy Disk Operating System for

giim mB ti rara tiium £i

till '■s*Ilii#*';#'F :

W*L

: :

The MSD booth, with MITS Business System admirers on the left and onlookers of the
time-sharing BASIC demonstration

&&te

ill

MINI/MICRO in full swing, with PCC's Pertec Division booth at the end of the aisle

Computer Notes Jan/Feb1978

f : H' v ■ ■ ■-!»' '^Mii^W
Willi " > SM/^MM+m

IIP;

fB9BM!PP

■'^BHHH^^^BwiMBHHWf

HM«i

fne new M/TS Business System

iCOM's latest addition — the Attache

microcomputers. Compatible with
FDOS-III is DEBBI™ (Disk Extended
BASIC by iCOM), a comprehensive
BASIC language system that is easy to
use and offers expanded capabilities.
Demonstrations of MITS' Time-Sharing
BASIC were given regularly all three
days, attracting a large number of
people to the booth. A variety of other
MSD products was also on display for
the guests of the show.

The MINI /MICRO 78' show will be
held in Philadelphia on April 18-20,
and 40 percent booth space is already
reserved. The conference organizers
are anticipating another successful
show for 1978 and are projecting
increased interest and attendance for
the future as microcomputers reduce
in price and gain in popularity.

A crowd around the MITS OEM Products display

Computer Notes Jan/ Feb 1 978

>*K«WI<I|«MW^

Biiipmpw

Introducing the Compact Attache™ Computer

msmk

H*I

'VV

The Attache is an attractive desktop computer that was recently introduced by PERTEC
COMPUTER CORPORATION'S Microsystems Division.

Pertec Computer Corporation's Micro-
systems Division recently introduced
a powerful desktop computer called
the Attache™ The Attache weighs
25 pounds and is built around the 8080
MPU. Its basic configuration includes
a CPU board, video board, turnkey
monitor board, and a full 64-character
alphanumeric ASCII keyboard.

Standard features of the Attache
include Light Emitting Diode (LED)
indicators for on/off and systems
status, a reset switch for return to the
PROM monitor, and a monitor PROM
that controls computer operation from
the keyboard. Also standard is a video
output jack for providing full upper
and lower case character generation,
16 lines of 64 characters each, and a
choice of black on white or white on
black character display with cursor
control.

The Attache's circuitry uses the
S-100 bus configuration with a 10-slot

board capability. Also standard with
the system on the turnkey board is 1 K
RAM with extra sockets for three
256-byte PROMs. An Audio Cassette
Recorder (ACR) SIO board is another
of the Attache's standard features, as
is a 16K Dynamic RAM Memory Board
that uses less than three Watts of
power and has an access time of 350
nanoseconds.

In addition to its list of standard
features, the Attache also offers high
reliability due to forced air cooling
over the vertically mounted cards. Its
power supply provides 10V at 10A
(regulated to 5V on boards) with pre-
regulated plus/minus 18V at 2A. The
Attache also features greater possible
expansion, because only three of the
ten slots are used by required boards
(the CPU, video, and turnkey monitor),
leaving seven slots for expansion.

Floppy disk systems and software,
including iCOM's® FD3712 Dual Disk

Desk Top IBM-formatted system or the
FD2411 Microfloppy with interface
supported by FDOS-III and DEBBI™
(Disk Extended BASIC by iCOM), are
available as options for the Attach§.
Other options include an audio cas-
sette recorder (KCACR) board,
110-9600 baud RS232 port, 16K byte
memory board expansion for up to 64K
of usable RAM, a 16K BASIC ROM
board with autojump start, and CSave
and CLoad cassette routines that are
included in BASIC. A ten-key pad for
high-speed data entries in business or
statistical applications and plug-in
compatibility for many versatile S-100
boards are additional options.

The Attache - is contained in a stylish
white cameo case and is priced below
competitive systems. The Attache
business computer is available at the
more than 40 MITS Computer Centers
across the continent.

Computer Notes Jan/ Feb 1 978

Machine Language to BASIC Converter

By Richard Ranger

An annoying but necessary step in
using the machine language interface,
DEFUSR, in MITS®BASIC is the
conversion of the machine language
program into POKE statements within
the calling BASIC program. Using the
following program, MITS BASIC users
may utilize the machine language
subroutines to enhance the capabil-
ities of their computers.

Machine language subroutines that
can be interfaced to BASIC through
the use of DEFUSR have been written
for a number of different functions,
from multi-precision addition to fast
analog to digital conversion and
storage. A few of these programs have
appeared in Computer Notes, while
others are scattered throughout the
operation and checkout procedures of
various manuals for MITS peripherals.
Generally, memory size is limited
during initialization. The machine
language program is placed above this
initialization limit, so that any opera-
tion within this subroutine will not
affect BASIC. This routine is normally
accessed using the DEFUSR function
of MITS BASIC, and, since the syntax
for this statement varies from version
to version, you should refer to the
manual to find the correct syntax for
calling the DEFUSR function
subroutine.

The purpose of this program is to
eliminate the need to toggle in the
machine language subroutine each
time a new routine is used. Without
this program, it would be necessary to
toggle in the subroutine before calling
it with any BASIC program or to
convert each octal location and in-
struction to decimal and then into a
statement of the form:

POKE (address), (instruction).

Using the following procedure, the
machine can write its own BASIC
program that contains all the neces-
sary POKEs to duplicate the machine
language subroutine. By running this
POKE program, the machine language
subroutine is quickly POKEd into
position before it is needed by the
main or calling program.

If you are using disk BASIC,
proceed according to the following
instructions. First, bring up BASIC,
initializing with at least one sequential
file and limiting its size so that your
particular machine language program
will reside in its appropriate location
(usually above the BASIC interpreter).

You must either toggle in the machine
language or use any method available
to enter the machine language pro-
gram initially, so the converter pro-
gram will be able to use the PEEK
function of BASIC to acquire the data.
After this has been accomplished,
LOAD the converter program, and
RUN it. At this time, you will be
required to enter the beginning and
ending locations of the machine
language program (in decimal) and a
temporary file name for the POKE
program. The converter will begin
PEEKing the locations containing the
machine language routine and will
create a string comprised of a line
number, the characters "POKE", ", ",
":", the address, and the contents of
the PEEKed location. This string of
characters is then written on the disk
in ASCII under the temporary file name
AND. AND may be merged with any
other program which does not contain
the same line numbers.

This method of creating machine
language subroutines that can be
interfaced with BASIC allows you to
write several different routines, merge
their corresponding POKE programs
into a larger BASIC program, and call
them much the same as BASIC
subroutines are called.

If you do not have a disk but still
require the use of machine language
subroutines, the temporary POKE
program must be written in ASCII but
placed on a medium other than floppy

disk. This problem may be resolved in
two different ways, depending upon
whether you have access to a teletype
with a paper tape punch and reader or
if you are limited to a cassette
recorder and mag tape.

If you do have access to a teletype,
load the machine language program as
before and delete lines 30, 35, and 140
from the converter program. Line 110
of the converter must be changed to
read: 110 PRINT T$. Enter the con-
verter program, make all the necessary
changes, type RUN, turn on the paper
tape punch, and type a carriage return.
The computer will then print the POKE
program on paper tape. After this has
been done, this ASCII paper tape may
be merged with the main BASIC
program by loading the main program
and then reading in the paper tape
program through the paper tape
reader. Again, note that the line
numbers of the POKE program and the
main program must be different.

If you do not have access to a
teletype or a floppy disk, your POKE
program must be saved in ASCII on a
cassette recorder. To accomplish this,
load the machine language as before
and be sure that BASIC has been
initialized with a "C" when WANT SIN-
COS-TAN was asked. (This write-up
assumes that the reader is using a
version of BASIC that incorporates the
CONSOLE command.)

Delete lines 30 and 35, and change
or add the following lines accordingly:

Continued on page 28

Easy Floppy Disk Alignment Check - continued from page 7

10 PRINT: PRINT"PIP - VER 4. 0"

20 CLEAR 0:X=FREC0>- 1500: IF X< THEN CLEAR 600 ELSE IF X>32000 THEN

CLEAR 3200O ELSE CLEAR X

30 DIMT2C 15>:F0RY=OT015:T2CY)=- 1 : NEXTY: PRINT"* "J :LINEINPUTBS

40 IFBi=""THENCLEAR200:EN€)

50 IF LEN(BS)»3 THEN CS= RI GHTS ( BS, LEN C BS ) - 3) ELSE CS=BS

60 BS=LEFTS(BS, 3)

70 IFBS="DAT"THEN680

80 IFBS="C0P"THEN87O

90 IFBS="LIS"THEN800

100 IF BS="CNV" THEN 1040

110 IFBS="DIR"THEWF=- 1: G0T027O

120 IF 3S="SRT" THEN F=0: DIMASC 255) : G0T027O

130 IFB£<>"INI"THE.gpRINT"ERR": G0T02O

140 G0SUB 760

150 AS=STRINGS< 1 37, 0) : M I DSC AS, 136* 1>=CHRS<255)

160 F0RT=6T076

170 F0R S=0 T0 3!

180 MIDSCAS, 1,2>=CHRSCT) + CHRS< C S* 17)AND3I )

190 G0SUB 6OO:DSK0S AS, s

200 NEXT S, T

210 T=7O:G0SUB 600 "DIRECTORY TRACK

220 A$=CHRSC70)+CHR$C0)+CHRSC0)+CHRS( 128)+CHRS( 1275+CHRSC0)

230 AS=AS+CHF.SC0)+CHRSC255>+STRINGSC 1 27, 0> +CHRSC 255)

240 DSK0SAS,

Continued on page 28

Computer Notes Jan/Feb1978

-:,'■ ' •' ' * .:-;-:

amm

■mwr

iimiPWi*

Easy Floppy Disk Alignment Check - continued

Machine Language to BASIC Converter - continued from page 27

;y else aso)=n$+'

250 PRINT:PRINT"D0NE"
260 G0T02O

270 G0SUB76O:0PEN"0", 1," RR", A

280 PRINT* 1- 1:CL0SE1:KILL" RR",A

290 PRINT

300 PRINT"DIRECT0RY DISK";A
310 PRINT: 1 =
320 F0RS=OT0 31
330 AS=DSKISC 17*SAND31>
340 AS=LEFTS(AS, 135)
350 AS= RIGHTS (AS, 128)
360 F0R T=0 T0 7
370 BS=LEFTS(AS, CT+r")'*16)
380 BS=RIGHTS(BS, 16)'
390 NS=L£FTSCBS,8)
400 BS= RIGHTS CBS, 8 >

410 X=ASC£BS) :BS=RIGHTS(B$, 7) :Y=ASCCB$)
420 BS=RIGHTS(B$,6):Z=ASC(BS>
430 IFASC(NS) = 0THEN470
440 IFASC(NJ)=255THEN490
450 R$="S":IFZ<:>2THENRS="R"
460 IF F THENPRINTNS; " ";R$,\" •"; XJ "
RI+" "+STRSCX)+" J "+STRSCY): 1=1+1
470 NEXTT
480 NEXTS

490 IF F 0R 1 = THEN PRINT: G0T0 20
500 IF 1=1 THEN 560
510 SW=0

520 F0R J=0 TB 1-2

530 IF AS(J)>ASCJ+1) THEN SWAP AS C J ) , ASC J+ 1 ) : SV=- 1
540 NEXT

550 IF Stf THEN 510
560 F0R J=0 T0 1-1
570 PRINT AKJ)
580 NEXT
590 PRINT: G0T02O
600 IFT2CA)<>— 1THEN640

610 IFCINP(8)AND64) = OTHENT2(A) = O:G0T064O
620 VAIT8,2, 2:0UT9, 2
630 G0T061O

640 IFT2(A) = TTHENRETURN
,650 D=l: IFT2CA)>TTHEND=2

660 WAI T8, 2, 2:0UT9, Ds T2CA) = T2CA>-2* CD- 1.5)
670 G0T064O

680 INPUT"TRACK"J T: IF T< THEN 20 ELSE INPUT"SECT0R"; S
690 G0SUB76O: G0SUB6OO
700 AS=DSKI$CS):F0RI = OT0LENCAS>-1
710 T1S=0CTS(ASCCRIGHT$CAS,LEN(A$)-I>))
720 T2S=LEFTSC" 000", 5-LEN C.T I S> ) +T 1 S: PRINT T2$;
730 IF I M0D 8=7 THEN PRINT

740 NEXT I: PRINT
750 G0T0 680
760 A=VALCC$)

770 IFA<O0RA> 15THENPRINT"ERR": G0T02O
780 0UT8, 128:0UT8,A
790 RETURN
800 G0SUB76O

810 CS=RIGHTSCCS,LENCCS)-l+CA>9))i I FASCC CSX>4054THENPRINT"ERR": G0T02O
820 CS=RIGHTS(CS,LENCCS)-1)
830 0PEN"I", t,CS,A
840 IFE0FC I >THENCL05E1: G0T02O
850 LINEINPUT#1,AS
860 PRINTAS:G0T084O
870 G0SUB76O:B=A

880 CS=RIGHTS(CS,LEN(CS)- 1+(A>9)): IFASC CCS)<> &054THENPRIN T"ERR": G0T32O
890 CS=RIGHTS(C$,LENCCS)- 1 ) : G0SUB76O: C=A
900 PRINT"FR0M "; B; " T0 "S.CS
910 INPUTAS: IFASCCAS)OASC<"Y")THEN20
920 F0RT=OT076
930 0UT8, 128 :011TB, C

940 A=C:G0SUB6OO:0UT8, 128: 0UT8, B: A=B: G0SUB600
950 F0RS=OT031

960 0UT8, 128:0UT8,B:BS=DSKISCS)

970 F.S=DSKI$CS):IFFS<>BSTHENPRINT"REREAD":G0T096O
980 0UT8, 128:0UT8, C

990 DSK0SBS, S: CS=DSKI SC S> : I FC$< >B$THENPRINT"REWRI TE": G0T09 5O
1000 NEXTS
1010 NEXTT
1020 PRINT"D0NE"
1030 G0T02O

1040 G0SUB 760 'ENABLE DISK
1050 F0R T=6 T0 76

1060 G0SUB 600 'P0SITI0N T0 TRACK T
1070 F0R S=0 T0 31

1080 AS=DSKISCS): IF ASC (HI DSC AS, 3, 1 ) )<>0 THEN 1120
1090 IF MIDSCAS, 136, l)=CHHSC255) THEN 1120

1100 MIDSCAS, 136, 1) = CHRSC255)

1110 DSK0S A$, S

1120 NEXT S
1130 NEXT T: G0T0 20
0K

CHRNGE LINE 168 TO REHD :

100 T*=K*+P*+R*+S*+X*+0*+P*+B*+S*+V*+Q*+P*+C*+S*+Z*+CHR* < 13 >
CHRNGE LINE 110 TO READ:

110 FOR J=l TO LEN<T*> : PR=flSCCMID*CT*, J,l»
HDD LINE 112:

112 WHIT 6,128,128: OUT?, PR : NEXT J
CHRNGE LINE 140 TO RERD :

140 REM THIS IS THE CONSOLE COMMAND FOR R 2SIO
RDD THE FOLLOWING LINES:

142 T*=" CONSOLE 16, " +CHR* < 12 >

144 FOR 1=1 TO 100 :REM fl SLIGHT DELflV

146 NEXT I

148 FOR K=l TO LENCT*>

150 PR=RSC <: M I D* < T*, K, 1 '> >

152 WRIT 6, 128, 128 : 0UT7, PR

154 NEXT K

At this time, start the cassette
recorder (record mode), and, after a
few seconds, type RUN, followed by a
carriage return. The added parts of the
program will allow the computer to
place the POKE program on cassette
tape and will follow it with a CON-
SOLE command to the main terminal
in use. If an I/O card other than a 2SIO
is used for this terminal, line 142 must
be changed in accordance with the
appropriate console register setting
for that particular I/O card (see page
34 of your BASIC manual). After the
POKE program has been made on
cassette, it may be merged with the
main BASIC program by first LOADing
the main program into the computer,
then typing CONSOLE 6, 3, followed
by a carriage return. The computer will
now take in data from the input port
#7, and, when all of the POKE program
has been entered, it will CONSOLE
back to the main terminal. (Note again
that the line numbers of the POKE
program must be different from the
line numbers of the main program.)

In all of the procedures just out-
lined, the entire program, main BASIC
plus the POKE program, may be saved
together as one main program after
they are both in the computer's text
buffer. The unmodified conversion
program set up for disk BASIC users is
also given in this article.

Program on Page 29

About The Author

Richard Ranger, a MITS engineering techni-
cian, is a Navy veteran who worked in
airborne reconnassance. He is currently
studying at the University of New Mexico for a
degree in Electrical Engineering.

Computer Notes Jan/Feb 1 978

Machine Language to BASIC Converter - continued

5 CLEAR 5B0

10 INPUT "STRRT LOCATION"; TRT

20 INPUT "STOP LOCATION"; STP

28 LINE INPUT "FILE NAME"; N*

35 OPEN "0", 1, N*.

40 K=10

50 FOR I=TRT TO STP STEP 3

60 X*=STR* C PEEK C I > > : V*=STR* C PEEK > : Z*=STR* < PEEK

70 A*=STR* : B*=STR* : C*=STR*

SS K**STR*<IO

90 P*="POKE" : S*=", " : 0*=" : "

100 T#=K*+P*+A$+S*+X*+0*+P*+B*+S*+V*+0*+P*+C*+S*+Z*

110 PRINT #1, T*

120 K=K+10

130 NEXT I

146 CLOSE 1

More on the KCACR - continued from page 8

FEF$$

NAM PUNKCR

OPT NOG
OUTCH EQU SFDF5
OUTCH EQU $FDE3
CRLF EQU $FFAB
STACK EQU $3FFF

ORG $00 F3

FCB $FF

ORG $4000

* ENTRY LINE FOR BASIC VI

.0 R3.2

LDX #$1AB2

BRA START

*ENTRY LINE FOR EDITOR Rl

LDX #$090B

BRA START

*ENTRY LINE FOR ASSEMBLER/EDITOR R1.0

LDX #$1C81

START STX HERE

LDS iCSTACK

BSR LEDTRL

LDX #0

STX BEGADR

LDX #$E6

BSR PUN

LDX #$100

STX BEGADR

FCB $CE

HERE FCB 0,0

BSR PUN

LDX #EOF

BSR PMESS

BSR LEDTRL

JMP CRLF

LEDTRL CLR A

GLR B

LED1 JSR OUTCH

DEC A

BNE LED I

RTS

PUN StX LASADR

PUNO LDX #FORM

CPX LASADR

BSR PMESS

BNE PUNO

LDA A LASADR+1

RTS

SUB A BEGADR+1

SENDIT JSR OUTCH

LDA B LASADR

INX

SBC B BEGADR

PMESS LDA B X

BNE PUN2

BPL SENDIT

CMP A #16

RTS

BCS PUN3

PNCH2 LDA B X

PUN2 LDA A #15

ABA

PUN3 STA A NUMBYT

PSH A

ADD A #4

TBA

JSR 0UT2H

INX

PUL A

BSR PNCH2

INX

BSR PNCH2

RTS

LDX BEGADR

FORM FCB $D,$A, 'S, "1,$FF

PUN4 BSR PNCH2

BEGADR RMB 2

DEC NUMBYT

LASADR RMB 2

BPL PUN4

NUMBYT RMB 1

STX BEGADR

EOF FCB $D,$A, 'S, "9,$FF

COM A

ORG $00 F3

JSR 0UT2H

FCB $03

DEX

END

00001

00002
00003
00004
00005
0000S
00007
00003
00009
00010
00011
00012
130013
00014
00015
00016
00017
0001S
00019
00020
00021
00022
00023
00024
00025
0002 S
00027
00023
00029

00030

00031

00032
00033
00034
00035
00036
00037
00038
00039
00040

00041

00042
00043
00044
00045
00046

0004 7
00048
00049
00050
00051
00052
00053
00054
00055
00056

0005 7
00058
0005 9
00060
00061
00062
00063
00064
00065
00066
00067
00068
00069
00070
00071
00072
00073
00074
00075
000 76
000 77
00078-
00079
00080
00081
00082
00033
00084
0008*
00086
0008 7
00088

FDF5
FDE3
FFAB
3FFF

00 F3

00F3 FF

4000

4000 CE 1AB2
4003 20 08

4005 CE 090B
4003 20 03

OUTCH
0UT2H
CRLF

STACK

400A
400D
4010
4013
4015
4018
401B

401 E
4020
4023
4026
4027
4029
402B

402 E
4030
4032
4035
4036
4037
4 03 A
403B
403D
403E
4041
4044
4046
4049
404 C

404 F
4052
4054
4056
4053
405A

405 D
405 F
4062
4063
4065
4067
406A
406C
406F
4071
4074
4075
4078
4079
407C
407E
407F
4082
4083
4035
4087
4088
4 08 A
408B
408C
408 D
4090
4091
4092
4093
4098
409A
409C
409D
00 F3
00 F3

CE 1C81
FF 4027
8E 3FFF
3D 20
CE 0000
FF 4098
CE 00 E6
8D 1 E
CE 0100
FF 4098
CE
00-

8D 13
CE 409D
3D 53
8D 03
7E FFAB
4F
5F

BD FDF5
4A

25 FA
39

FF 409A
CE 4093
8D 3D
B6 409B
B0 4099
F6 409A
F2 4098

26 04
81 10

25 02
86 0F
B7 409C
3B 04
BD FDE3
08

8D 23

8D 21

FE 4093

8D 1C

7A 409C

2A F9

FF 4098

BD FDE3

BC 409A

26 C3
39
BD
03

E6 00
2A F8
39

E6 00
IB

BD FDE3

0002

0001

START

HERE

FDF5

NAM
OPT
EQU
EQU
EQU
EQU
ORG
FCB
ORG

♦ ENTRY LINE

LDX
BRA

♦ ENTRY LINE

LDX
BRA

♦ ENTRY LINE
LDX
STX
LDS
BSR
LDX
STX
LDX
BSR
LDX
STX
FCB
FCB

BSR
LDX
BSR
BSR
JMP

LEDTRL CLR A
CLR B

LED1 JSR

DEC A

BNE

RTS

STX

LDX

BSR

LDA A

SUB A

LDA B

SBC B

BNE

CMP A

BCS

LDA A

STA A

ADD A

JSR

INX

BSR

LDX

BSR

DEC

BPL

STX

COM A

JSR

DEX

CPX

BNE

RTS

SENDIT JSR
INX
LDA B
BPL
RTS
LDA B
ABA
PSH A
TBA
JSR
PUL A
INX
RTS
FCB

PUNKCR

NOG

$FDF5

$FDE3

$FFAB

$3FFF

$00F3

$FF

$4000
FOR BASIC VI. R3.2

#$1AB2

START
FOR EDITOR R1.0

#$090B

START
FOR ASSEMBLER/EDITOR Rl .0

#$1C«1

HERE

# STACK

LEDTRL

BEGADR

#$E6

PUN
#$100

BEGADR

$CE

0,0

PUN
#EOF

PMESS

LEDTRL

CRLF

OUTCH

LED1

PUN
PUNO

PUN2
PUN3

PUN4

PMESS

PNCH2

FORM

BEGADR RMB

LASADR RMB

NUMBYT RMB

EOF FCB

ORG

FCB

END

LASADR
#FORM
PMESS
LASADR+I
•BEGADR+1
LASADR
BEGADR
PUN2
#16
PUN3
#15
NUMBYT
#4
0UT2H

PNCH2

BEGADR

PNCH2

NUMBYT

PUN4

BEGADR

0UT2H

LASADR
PUNO

OUTCH

X
SENDIT

0UT2H

TOTAL ERRORS 00000
ENTER PASS

$D,$A, 'S, '1,$FF

$D,$A, 'S, '9,$FF

$00F3

$03

Continued on page 30

Computer Notes Jan/Feb 1 978