LabVIEW can represent numeric data types as signed or unsigned
integers (8-bit, 16-bit, or 32-bit), floating-point numeric
values (single-, double-, or extended-precision), or complex
numeric values (single-, double-, or extended-precision). When
you wire two or more numeric inputs of different
representations to a function, the function usually returns
output in the larger or wider format. The functions coerce the
smaller representations to the widest representation before
execution, and LabVIEW places a coercion dot on the terminal
where the conversion takes place.
For example, the For Loop count terminal is a
32-bit signed integer. If you wire a double-precision,
floating-point numeric to the count terminal, LabVIEW converts
the numeric to a 32-bit signed integer. A coercion dot appears
on the count terminal of the first For Loop, as
shown in Figure 3.
If you wire two different numeric data types to a numeric
function that expects the inputs to be the same data type,
LabVIEW converts one of the terminals to the same
representation as the other terminal. LabVIEW chooses the
representation that uses more bits. If the number of bits is
the same, LabVIEW chooses unsigned over signed.
In the example in Figure 4, a 32-bit
signed integer (I32) and a double-precision, floating-point
numeric value (DBL) are wired to the Divide
function. The 32-bit signed integer is coerced since it uses
fewer bits than the double-precision, floating-point numeric
value.
To change the representation of a numeric object, right-click
the object and select Representation from the
shortcut menu. Select the data type that best represents the
data.ut data types.
When LabVIEW converts double-precision, floating-point numeric
values to integers, it rounds to the nearest integer. LabVIEW
rounds x.5 to the nearest even integer. For
example, LabVIEW rounds 2.52.5 to
22 and
3.53.5 to
44.
Refer to the Data Types section of
Introduction to LabVIEW, of this
manual or to the LabVIEW Help for more information about data
types.
"A full introductory course on programming with LabVIEW."