Sharp MZ-800 User Manual

5.5 Array Variables
An array is an arrangement of variables of the same data type, which are referred to by a common
name. Each variable of an array is identified by the common name, which is composed of a string
formed in the same manner as a variable name and followed by subscripts enclosed within parenthese,
e.g., A(X) and BS(x,y). An array with one subscript (such as A(X), B$(1) or P(100)) is called a one-
dimensional array, while that with two subscripts (such as A(x,y), B$(1,3) or P(50,25)) is called a two-
dimensional array. To use array variables in a program, the common name and the number of varia-
bles
included
in
the
array
must
be
declared
before
they
are
used.
For
details
see
the
explanation
of
the DIM statement in Chapter 6.
* Note Concerning Computational Error
Computational
error
must
always
be
taken
into
consideration
whenever
a
computer
is
used.
The
reason
for
this
is
that,
although
computational
error
can
be
reduced
by
increasing
the
number
of
digits
of
numerical
data
which
are
handled,
not
even
a
computer
can
handle
an
infinite
number
of
digits.
Further,
the
more
digits
are
involved
in
any
given
calculation,
the
greater
the
amount
of
time
which
is
involved
in
completing
it.
Therefore,
it
is
important
to
be
aware
of
the
sources
of
error
and
to
construct
programs
so
that
error
is
minimized.
(For
example,
use
the
sequence
‘5
*6/3”’
instead
of
“*5/3*6"’.)
Take
the
following
into
account
when
doing
calculations
in
BASIC
(1Z016)
for
the
MZ-800.
(1) Rounding error
Rounding
error
is
the
error
which
results
when
the
number
of
digits
to
the
right
of
the
decimal
place
exceed
the
number
of
effective
digits
which
can
be
handled.
For
example,
when
the
num-
ber
2/3
is
calculated,
the
true
result
is
0.666666666.
. .
(where
the
number
of
6s
is
infinite).
However,
if
the
number
of
effective
digits
is
8,
the
result
will
be
rounded
to
0.66666667.
(2) Error resulting upon conversion to binary form
Although
numbers
are
ordinarily
input
in
decimal
format,
they
are
internally
converted
to
bi-
nary
form
for
calculation.
According,
a
binary
number
with
an
infinite
number
of
digits
may
result
upon
conversion
even
if
the
original
decimal
number
only
has
a
few
digits.
For
example,
when
the
decimal
number
0.1
is
converted
to
binary
form,
the
result
is
0.00011001100.
.
.
.
Since
this
must
be
rounded
for
calculation,
a
certain
amount
of
error
results.
(3) Increase in relative error due to subtraction
When
one
number
is
subtracted
from
another,
the
relative
size
of
the
error
in
the
result
will
be
greater
than
that
in
the
original
numbers.
This
is
illustrated
in
the
example
below,
where
the
digits
which
include
error
are
marked
with
a
dot
(.).
An
error
of
+1
in
the
number
100012
corresponds
to
an
error
percentage
of
about
0.001%;
however,
relative
error
is
much
greater
after
subtraction,
since
11
+1
corresponds
to
a
relative
error
of
about
10%.
100012
~ 100001
i
(4) Error due to approximation
With
a
computer,
exponentiation,
trigonometric
calculations,
and
logarithmic
calculations
are
done
using
approximation;
in
consequence,
a
certain
amount
of
approximation
error
results
when
such
calculations
are
done.
5-6

Sharp MZ-800 Specifications

General

Brand	Sharp
Model	MZ-800
Category	Desktop
Language	English