cos(3I)

NAME
     COS, DCOS, QCOS, CCOS, CDCOS, CQCOS, COSD, DCOSD, QCOSD - Computes
     cosine

SYNOPSIS
     COS ([X=]x)
     DCOS ([X=]x)
     QCOS ([X=]x)
     CCOS ([X=]x)
     CDCOS ([X=]x)
     CQCOS ([X=]x)

     COSD ([X=]x)
     DCOSD ([X=]x)
     QCOSD ([X=]x)

IMPLEMENTATION
     UNICOS, UNICOS/mk and IRIX systems

     CF90, MIPSpro 7 Fortran 90, MIPSpro Fortran 77

STANDARDS
     Fortran

     Fortran extensions:  QCOS, CDCOS, CQCOS, COSD, DCOSD, QCOSD

DESCRIPTION
     These functions evaluate y = cos(x).

     COS and COSD are the generic function names; the others are specifics.
     These are elemental intrinsic functions.  They accept the following
     argument:

     x         Must be of type real.  For COS, DCOS, QCOS, CCOS, CDCOS, and
               CQCOS, the argument must be in radians.  It is treated as
               (modulo 2 * pi).  For COSD, DCOSD, and QCOSD, the argument
               must be in degrees and is treated as modulo 360.

               See the RETURN VALUES section for more information on input
               data types and return values.

     Vector versions of these intrinsics exist on some systems.  Vector
     versions of the COS, DCOS, CCOS, and CDCOS intrinsic functions exist
     on UNICOS and UNICOS/mk systems.  On UNICOS/mk systems, the vector
     versions of these intrinsics are used when -O vector3 or -O3 are
     specified on the compiler command line.  A vector version of the
     single-precision and double-precision COS intrinsic exists on IRIX
     systems and is used when -O3 is specified on the compiler command
     line.

NOTES
     The COS, DCOS, CCOS intrinsic function names can be passed as
     arguments; the others cannot.

     On UNICOS and UNICOS/mk systems, the following return values occur:

     * On systems that support the IEEE standard for floating-point
       arithmetic,            25
       COS(x) = NaN if |x| > 2

     * On systems that do not support the IEEE standard for floating-point
       arithmetic,          25
       COS(x) = 0 if |x| > 2

     On IRIX systems, the following return value occurs:
                                      19
          COS(x) = NaN if |x| > pi * 2

     On all systems that support the IEEE standard for floating-point
     arithmetic, the following is true:

          COS(x) = NaN if x = NaN

RETURN VALUES
     COS and COSD return the single-precision cosine of their
     single-precision arguments.

     DCOS and DCOSD return the double-precision cosine of their
     double-precision arguments.

     QCOS and QCOSD return the quad-precision cosine of their
     quad-precision arguments.

     CCOS returns the complex cosine of its complex argument.

     CDCOS returns the complex double-precision cosine of its complex
     double-precision argument.

     CQCOS returns the complex quad-precision cosine of its complex
     quad-precision argument.

     The data type and kind type for single-precision, double-precision,
     and quad-precision values differ depending on your platform.  For
     platform-specific information, see the TERMINOLOGY section of the
     INTRO_INTRIN(3I) man page.

SEE ALSO
     Intrinsic Procedures Reference Manual for the printed version of this
     man page.