Help Files: Fortran77: ISML Library

     HELP                      Sep. 15, 1984              F77/IMSL_LIB



          IMSL Library of Mathematical and Statistical Procedures

     The IMSL library, which includes over 500 Fortran  callable  sub-
     routines,  is  available for VAX UNIX.  There are both single and
     double precision versions; check with your  system  administrator
     to  see if your system has the IMSL library and how to access it.
     The rest of this write-up discusses assumes you have  the  double
     precision version.

     While IMSL includes mainly mathematical and statistical routines,
     chapter  U  of the library includes utility routines for printing
     histograms, for plotting functions, and for printing arrays.   To
     use the double precision version, specify '-limsld' to f77:

             f77 prog.f -limsld

     (You may need to use fort instead of f77 on some systems.  If so,
      you may also need to remove all .o and .a files before recompiling.)

     The IMSL manual includes an introduction, table of  contents  and
     permuted  index  at the beginning of volume 1.  New users of IMSL
     should read sections 5.2 through 5.6.1 (about 5 pages) of the in-
     troduction before using the library.  If IMSL does not contain an
     appropriate subroutine for what you are doing, see "help f77  li-
     braries" for information on other libraries.

     Most of the library is written using double precision  variables.
     To  see  if a specific routine expects single or double precision
     arguments, look in the manual for the 'PRECISION/HARDWARE' infor-
     mation.   If it specifies 'DOUBLE/H32' or 'SINGLE AND DOUBLE/H32'
     then floating point arguments must be  double  precision,  if  it
     specifies  'SINGLE/H32' or 'SINGLE/ALL' then floating point argu-
     ments must be single precision.

     The manual is written as if all  the  subroutines  expect  single
     precision arguments.  Generally, for the double precision version
     of the library, you will have to supply  double  precision  argu-
     ments.  If you have written your program in single precision, see
     "help f77 to_double" to see what changes are needed to use double
     precision.

     There are five common errors in using IMSL routines:

     1.   Using single precision when double precision is required.

     2.   Not checking the error flag.  Most routines  have  an  error
          flag  that  IMSL  sets to an error value when something goes
          wrong; it is your responsibility to check it.

     3.   Omitting arguments.  IMSL routines often have many arguments
          that are required even if you don't use the result.

     4.   Not dimensioning arguments.  Especially make sure work areas
          are appropriately dimensioned.

     5.   Supplying incorrect 'row dimension' of matrices.  Many  IMSL
          routines have a calling sequence of the form:

                  call sub( a, nr, nc, ia, ... )

          where 'a' is a matrix, 'nr' is it's row order, 'nc' is  it's
          column  order,  and  'ia' is it's row dimension.  The orders
          'nr' and 'nc' are the number of rows and columns being used,
          while  'ia'  is  the  row dimension of the matrix exactly as
          specified in the dimension statement in the calling program.
          The row order and row dimension are often different and must
          be specified correctly.