REAL(3M) Last changed: 1-6-98
NAME
REAL, FLOAT, SNGL - Converts to type real
SYNOPSISREAL ([A=]a [,[KIND=]kind])
FLOAT ([A=]a [,[KIND=]kind])
SNGL ([A=]a [,[KIND=]kind])
IMPLEMENTATION
UNICOS and UNICOS/mk systems
STANDARDS
Fortran 90
DESCRIPTION
REAL, FLOAT, and SNGL are elemental functions for the CF90 compiler.
arg can be of type complex, integer, or real.
These functions convert specified types to type real, such as a (or a
for complex arguments). r
Type conversion routines assign the appropriate type to Boolean
arguments without shifting or manipulating the bit patterns they
represent.
Argument range
REAL:
Real:
2450
|x| < infinity (infinity is approximately 10 )
On UNICOS/mk systems and on CRAY T90 systems that perform IEEE
arithmetic
308
infinity is approximately 10
Integer:
46
|x| < 2
Integer on CRAY T90 systems that support IEEE arithmetic:
63
|x| < 2
Complex:
|x | < infinity
r
FLOAT:
Integer:
46
|x| < 2 (Cray PVP systems, 46-bit integer)
64
|x| < 2 (Cray PVP systems, 64-bit integer, and
CRAY T90 systems that support IEEE arithmetic)
Boolean:
46
|x| < 2
SNGL:
Double-precision:
64
|x| < 2
NOTES
The name of this intrinsic cannot be passed as an argument.
RETURN VALUESREAL returns the real equivalent of its complex, integer, or real
argument.
FLOAT returns the real equivalent of its integer or Boolean argument.
SNGL returns the real equivalent of its double-precision argument.
SEE ALSO
Intrinsic Procedures Reference Manual, publication SR-2138, for the
printed version of this man page.