tcj_22 1986-0102.pdf

(4011 KB) Pobierz
1B8NI",a-WI
THE COMPUTER
Issu.
Num~r
22
JOURNAL~
S2.30US
For Those Who Interface, Build, and Apply Micros
NEW-DOS
Write Your Own Operating System
paleS
Variability In The BOS C Standard Library
Porting 80S C To CP/M 86
pa le21
The SCSI Interface
Introductory Column To A Series
pa le2S
Indexed Sequential Access Method Files
Using Turbo PascaiiSAM Files
pille27
The
AMPRO
Little Board Column
The Computer Corner
pa leS6
pa le42
heating the wire and applying rosin
core solder; then do the same again,
but put a little paste of liquid
flux
on
the wire before you start. You'll find
that the added
flux
makes for
'a
much neater and quicker job, and
also requires less heat from the iron.
Only a teeny tiny bit of
flux
is
required (more will just make a
mess); but the difference it makes is
tremen ous!
One inal, but very important
point:
ut why (and when) you
need to c
the
flux
residue off af-
terward. 'ney tell you that rosin is
non-eonductive...but that's not quite
true. It's non-conductive, compared
to a piece of wirei"but compared
to
a
ten-megohm resistpr, it conducts
quite nicely, thank you!
",
i/
For instance,
if
you're
trying
to get
a long time delay in a
555
timing
cir-
cuit, by using a fairly small
capacitor and a very
high
resistan-
ce, you had better clean off
all
the
flux
when you finish-or the conduc-
tivity of that "non-conducting"
flux
may upset your calculations con-
siderably. Or
if
you're working with
CMOS ICs, you may find that current
leakage from one of the power suply
pins to,' .an adjacent input pin,
througJI un-removed rosin
flux
residue,
can
shut· down the circuit
entirely!
1h
short, whenever you're working
with high resistances or low curren- /
/ ts, you had better clean all the old"
flux
off the board before you power,it
up, or you may find a nasty bug<'in
/
Df2~ ?~t-
Th~eomputer
Journal/Issue #22
your circuit!
Jock
Root
L.A.,CA
FORTH
\
Bill Kibler: I have just read your
Computer Corner column in issue
#17
and I
am
interested in your idea of
buildinga'
z-so
FORTH unit. I am
currentlY learning
64
FORTH from
HES /'On my Commodore
64
and
would like to put FORTH on a
z-so
ooard that I have already built.
I became interested in FORTH
~ause
I work with elevators which
are rapidly becoming computerized
and FORTH seems perfectly suited
to
this
field.
Any columns on
Z-BO
FORTH
". would be greatly appreciated. As
usual, you and everyone at The
Computer Journal are doing a great
job.
I
G.K.
\
~
j
\
New York
SETS\OU
Z Operating System. an a-bit OS
that
t1iesl Optimized
)8O/ZSO assembly language
code - full software dl!\l1!lopment system
with
prOllen
I
ble libraries of productive
subroutines - relocating (ROM and RAMI macro am;mb'r. linker. librarian. cross-
reference table generator. debuggers. translators and
dlsa~mblers
- ready to
free
~'
/
!
FREE!
mlzed OS enwon-
Data Acquisition and Control
you'
/
/
High
and rlexlbrllty' Proou([lWY results
ments. matching operator. taSkS and maChine
~formance
frln
t1ynamlcally cu
Real-tIme control kernel option allOWl qUick s ftware development for In striaI control
applications. other tools and utilities for office de -top personal compurrng functlo
local area
networks
[Q
Ethernet. AppleTalk Omnlnet. AI et. PC -Net (Sytek! - from micro to
Inframe
command. control and communications DI\!f;buted p(Q(esslng application programs ar aSlly
d~IOped
'
• Extreme organizational flexlbiliry. each drrect0'Y anorher enVIronment
• Multiple Commands per line
• ."-Ilases /complex serres of commands known by s,mple names/ With varrable pasSing
• Named Drrectorres With absolute password securr!)
• Full-screen command line editing With prt'VJous command recall and execution
• Shells and Menu Generators. With shell varrables
• Command-file search Paths.
dynamically
alterable
• Screen-{)rrented file manrpulalion and automatic archMng and backup
• 512 megabyte file sizes. 8 gigabyte diSks handled
• Auto disk reset fJhen changing floppies
• TCAP databasr!handles characterrstlcs of olot'r
50
compurers and terminals.
more easily added
• Tree-filruetwed online help and dOCumentalion subsysrem
• 76 syntax-Compalible support utilities
I enjoy reading your informative
journal.
Mr
Jerry Houston's article
on analog data acquisition and con-
trol systems was especially in-
teresting
to
me.
I wonder
if
he and!or others might
be interested in elaborating on ac-
tual
applications of ADC units, such
as those mentioned in his article.
As
you know, some of these devices are
relatively inexpensive and most can
be interfaced through RS-232 with a
number of different micros.
In
my
opinion, they present a unique oppor-
tunity
for us who interface micros
with the real world. I
am
sure that a
number of your current readers and
potential new subscribers would
be
. terested in
this
area.
"rl1aDl~
for your efforts on editing
and -publishing The Computer Jour-
nal.
Matthew K. Rogoyski, Ph.D.
Hotchkiss, CO
Your missing /Ink has been
found -
ZI
Now
fly
with eaglesl Fast response. emdent
resource utilization. link
to rest
of computing world - shop floor
to
executive suite.
micro
to
corporate mainframe. Call 415/948-3820 for literature.
(Continued on page
38)
l:.!J
Echelon, Inc.
~
101F1rstStreel. Suite4Z7. LosAlt08,CA94022. 415/94&-382tI
·
.
iY':Li
J
:'.~,
!'""\,..;.
I,
j
V
:~
A
\
\
The Computer Journal/Issue #22
Using C
Is CP/M
Dead?
to
say
to
myself, "this
is
neat, but
I read with interest your editorial
Is CP/M dead? Are hammers
how do I get to the system, and will
all of this spiff get in my way?" With
in Issue 21.
It
is possible
to
get most dead? Are nails dead? Is cooked food
of the things you wish using some C a thing of the past? There are some
the exception of the Ampro z..80
compilers (plus add-ons) but unfor- questions we writers should not
machines,
all
that spiff did get in the
tunately, the BDS C compiler does dignify with an answer, and would
way. Ampro allows me, the
not support them. Many C compilers not, if
it
weren't so much fun!
operator/developer, to decide how
" support the pre-processor directives
To be quite honest, I never saw
much spiff I want.
~Nasm
and #endasm. The use of these pure,
virgin,
CP/M until I just had to
These days I design systems for
dfrectives allows the programmers see what it looked like. Pure CP/M is
the disabled community. These are
to include in-line assembler code in an option on the Ampro Series 100
challenges greater than the space
his or
h~C
code. When the pre- systems. When I first booted it, I
program, and demand the highest
f~
these directives, the
thought my terminal program had
technology. I need a system that I
processor
code
is
marked'so that it will not be gone into high orbit (again). The
can tailor to the specific needs of the
individual. Not only must I be sure
optimized during"'ttIe optimization point
is,
who uses pure CP/M?
If
pass on the
compiler'~
this pass is CP/M
is
dead, then it has been dead
that the system may be used by the
present). This
is
a "'great im- for a long time, and will be dead for
individual with ease, but it must be
reliable.
provement over the
meth~
which decades to come.
It
is only we masochistic system
Now then, for the casual operator,
you have to use with BDS C. In te..rms
CP/M does not present a great deal
of "flash compiling", it is my
~
programmers who ever see CP/M.
of flash, nor pretty noises.
It
must be
pression that this could only be done".. We are the only ones who appreciate
using a single pass C compiler (with, )t,..albeit ina love/hate relationship.
remembered that CP/M was
designed as a business workhorse.
of course, a built in editor). The What
the'
user sees, and com-
problem with single pass C com-
muni~es
with is the Console Com-
For a person who just operates a
pilers is that they are sensitive to the mand Processor, (CPP). While it is
computer, or perhaps plays with an
order in which #define's are listed in
/8
"part of the-- standard CP/M Disk
assembler, CP/M type systems have
If
several #define://Operating System, <DOS), it is
little
to
offer. However, it
is
this type
the code.
statements refer to each other.
aiId
seldom allowed to remain as Digital
of computerist that
is
the most vocal
are incorrectly ordered, a single Reseach intended it.
"~rhaps
the
in what has been termed, "The DOS
pass compiler could end up with greatest gift to computerqom was
WARS." When running an ap-
unresolved references, which would Richard Conn's ZCPR,
~
CPP
plications program, the operator
.
",~,
never sees the operating system,
halt compilation. A good1llternative enhancement.
would be the use
of~C
interpreter,
I began computing in an industna
only the applications program.
several of which/are now on the world. The king of the
space'~~J
ese vocalists are judging the per-
market (although I believe all of program was the RCA CDPl802
"fQ
ance of these programs, not the
them are
ta~r
fed at the IBM PC en- microprocessor. This was a CMOS
0
.
g.systems. But, as I
think
of
vironment' ather than CP ). These chip, and has been available literally
it, th"
pIe do not read TCJ
gener
ve built in editors which for decades. People are just now
either,
b
dough-files, or BC
will
t back to the source code if discovering CMOS technologies,
Weekly.
s
x errors are found (similar to though few really understand them.
The best thing about CP/M type
bo Pascal's editor). They are Hence, my perspective
is
that of
hardware, as opposed to the CP/M
o generally syntactically com- machine
intellect,
robotics,
operating system, is that there is a
patible with major PC C compilers satellites, and deep space probes,
reasonably standard way of doing
(Lattice and Microsoft compilers where the machine must fend for it-
things.
If
I didn't have a fully
and occasionally some others). The self.
debugged set of routines to handle
My world does not generally in-
the disks, and terminal, I'd have to
problem here is that you are set back
the price of the interpreter (which volve spiffy graphics displays,
write them. That is just the reality of
can range from $100.00 to
$500.(0),
as . though I appreciate that type of
computing.
In
my work I use equipment
well as the cost of your compiler programming genius. Nor does
it
(which, for Lattice and Microsoft generally
involve
complex
designed to run CP/M. There are a
are not cheap).
mathematical process. I am, after
number of reasons for this choice.
Don Howes
all just a lowly chip mechanic.
First,
is
the price. Thousands of
As
I entered into the world of con-
people are maimed daily in
Pullman, WA
sumer computers I was spellbound
automobile mishaps. When a
by
all
the nifty features of CP/M
disability strikes money is an issue.
machines. After about an hour I had
In other fields,
it
is
the same con-

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika:

Zgłoś jeśli naruszono regulamin