*nix Documentation Project
·  Home
 +   man pages
·  Linux HOWTOs
·  FreeBSD Tips
·  *niX Forums

  man pages->IRIX man pages -> f90/spacing (3)              
Title
Content
Arch
Section
 

Contents

 

SPACING(3I)					       Last changed: 1-6-98

NAME    [Toc]    [Back]

     SPACING - Returns the absolute spacing of real model numbers near the
     argument value

SYNOPSIS    [Toc]    [Back]

     SPACING ([X=]x)

IMPLEMENTATION    [Toc]    [Back]

     UNICOS, UNICOS/mk,	and IRIX systems

STANDARDS    [Toc]    [Back]

     Fortran 90

DESCRIPTION    [Toc]    [Back]

     The SPACING intrinsic function returns the	absolute spacing of real
     model numbers near	the argument value.  It	accepts	the following
     argument:

     x	  Must be of type real

     SPACING is	an elemental function.	The name of this intrinsic cannot
     be	passed as an argument.

NOTES    [Toc]    [Back]

     This routine is implemented on CRAY T90 systems that support IEEE
     arithmetic	by calls to scalar-only	library	routines.

RETURN VALUES    [Toc]    [Back]

     The result	type and type parameter	are the	same as	x.
			       e-p
     The result	has the	value b	  , where b, e,	and p are as defined in	the
     real number model,	provided this result is	within range.  If not, the
     result is the same	as that	of TINY	(x).  For more information on the
     real number model,	see the	MODELS(3I) man page.

EXAMPLES    [Toc]    [Back]

     Example 1:	 A CRAY	C90 system returns the following values	for the
     SPACING intrinsic function:

	  SPACING(3.0)=	2.8421709430404E-14
	  SPACING(3.0_4)= 2.384185791015625E-7
	  SPACING(3.0_8)= 2.8421709430404E-14
	  SPACING(3.0_16)= 1.009741958682895E-28

     The first is default type real, and the others are	defined.

     Example 2:	 An IRIX system	returns	the following values for the
     SPACING intrinsic function:

	  SPACING(3.0)=	2.384185791E-7
	  SPACING(3.0_4)= 2.384185791E-7
	  SPACING(3.0_8)= 4.44089209850062616E-16
	  SPACING(3.0_16)= 2.46519032881566189191165176650870697E-32

     The first is default type real, and the others are	defined.

     Example 3:	 A UNICOS/mk system returns the	following values for the
     SPACING intrinsic function:

	  SPACING(3.0)=	4.44089209850062616E-16
	  SPACING(3.0_4)= 2.384185791015625E-7
	  SPACING(3.0_8)= 4.44089209850062616E-16
	  SPACING(3.0_16)= 4.44089209850062616E-16

     Example 4:	 A CRAY	T90 system that	supports IEEE floating-point
     arithmetic	returns	the following values for the SPACING intrinsic
     function:

	  SPACING(3.0)=	4.44089209850062616E-16
	  SPACING(3.0_4)= 4.44089209850062616E-16
	  SPACING(3.0_8)= 4.44089209850062616E-16
	  SPACING(3.0_16)= 3.85185988877447171E-34

     Note that there is	no KIND=4 value	and that default is KIND=8.

SEE ALSO    [Toc]    [Back]

      
       
     MODELS(3I)

     Intrinsic Procedures Reference Manual, publication	SR-2138, for the
     printed version of	this man page.

SPACING(3I)					       Last changed: 1-6-98

NAME    [Toc]    [Back]

     SPACING - Returns the absolute spacing of real model numbers near the
     argument value

SYNOPSIS    [Toc]    [Back]

     SPACING ([X=]x)

IMPLEMENTATION    [Toc]    [Back]

     UNICOS, UNICOS/mk,	and IRIX systems

STANDARDS    [Toc]    [Back]

     Fortran 90

DESCRIPTION    [Toc]    [Back]

     The SPACING intrinsic function returns the	absolute spacing of real
     model numbers near	the argument value.  It	accepts	the following
     argument:

     x	  Must be of type real

     SPACING is	an elemental function.	The name of this intrinsic cannot
     be	passed as an argument.

NOTES    [Toc]    [Back]

     This routine is implemented on CRAY T90 systems that support IEEE
     arithmetic	by calls to scalar-only	library	routines.

RETURN VALUES    [Toc]    [Back]

     The result	type and type parameter	are the	same as	x.
			       e-p
     The result	has the	value b	  , where b, e,	and p are as defined in	the
     real number model,	provided this result is	within range.  If not, the
     result is the same	as that	of TINY	(x).  For more information on the
     real number model,	see the	MODELS(3I) man page.

EXAMPLES    [Toc]    [Back]

     Example 1:	 A CRAY	C90 system returns the following values	for the
     SPACING intrinsic function:

	  SPACING(3.0)=	2.8421709430404E-14
	  SPACING(3.0_4)= 2.384185791015625E-7
	  SPACING(3.0_8)= 2.8421709430404E-14
	  SPACING(3.0_16)= 1.009741958682895E-28

     The first is default type real, and the others are	defined.

     Example 2:	 An IRIX system	returns	the following values for the
     SPACING intrinsic function:

	  SPACING(3.0)=	2.384185791E-7
	  SPACING(3.0_4)= 2.384185791E-7
	  SPACING(3.0_8)= 4.44089209850062616E-16
	  SPACING(3.0_16)= 2.46519032881566189191165176650870697E-32

     The first is default type real, and the others are	defined.

     Example 3:	 A UNICOS/mk system returns the	following values for the
     SPACING intrinsic function:

	  SPACING(3.0)=	4.44089209850062616E-16
	  SPACING(3.0_4)= 2.384185791015625E-7
	  SPACING(3.0_8)= 4.44089209850062616E-16
	  SPACING(3.0_16)= 4.44089209850062616E-16

     Example 4:	 A CRAY	T90 system that	supports IEEE floating-point
     arithmetic	returns	the following values for the SPACING intrinsic
     function:

	  SPACING(3.0)=	4.44089209850062616E-16
	  SPACING(3.0_4)= 4.44089209850062616E-16
	  SPACING(3.0_8)= 4.44089209850062616E-16
	  SPACING(3.0_16)= 3.85185988877447171E-34

     Note that there is	no KIND=4 value	and that default is KIND=8.

SEE ALSO    [Toc]    [Back]

      
       
     MODELS(3I)

     Intrinsic Procedures Reference Manual, publication	SR-2138, for the
     printed version of	this man page.
[ Back ]
 Similar pages
Name OS Title
rrspacing IRIX Returns the reciprocal of the relative spacing of numeric model numbers near the argument value
fraction IRIX Returns the fractional part of the numeric model representation of the argument value
exponent IRIX Returns the exponent part of the argument when represented as a numeric model number
huge IRIX Returns the largest number in the integer or real numeric model
range IRIX Returns the decimal exponent range of a number according to the real or integer model
minexponent IRIX Returns the minimum (most negative) exponent of a real number in the numeric model
tiny IRIX Returns the smallest positive value for a real argument
precision IRIX Returns the decimal precision of a number in the real number model
floating IRIX 0- return scalar values associated with the argument derived from the model of floating point data.
getsm IRIX returns the current shading model
Copyright © 2004-2005 DeniX Solutions SRL
newsletter delivery service