Return to nag_lib CVS log

Up to [CSRG BSD Unix] / 43BSDTahoe / new / help / src / f77

Annotation of 43BSDTahoe/new/help/src/f77/nag_lib, revision 1.1.1.1

1.1       root        1: .TI F77/NAG_LIB "Sep. 4, 1985"
                      2: NAG Library of Mathematical and Statistical Procedures
                      3: 
                      4: The NAG library, which includes about 500 Fortran callable subroutines,
                      5: is available for VAX UNIX.  There are both single and double precision
                      6: versions; check with your system administrator to see if your system
                      7: has the NAG library, and how to access it.  Your administrator can also
                      8: tell you where the documentation (6 volumes) is kept.  The rest of
                      9: this write-up assumes you have the double precision version,
                     10: accessed as '-lnagd'.
                     11: 
                     12: While NAG includes mainly mathematical and statistical routines,
                     13: the library also includes sorting routines in chapter M01 and
                     14: machine and mathematical constant routines in chapter X.
                     15: To use the double precision version, specify '-lnagd' to f77:
                     16: 
                     17:        f77 prog.f -lnagd
                     18: 
                     19: If the NAG library does not include a subroutine for what you are
                     20: doing, see "help f77 libraries" for information on other libraries.
                     21: 
                     22: The NAG manual is written to be applicable to both single and double
                     23: precision versions of the library.  Generally you will have to supply
                     24: double precision arguments.  Where the documentation indicates "real"
                     25: in italics, use 'double precision' on VAX UNIX.  If you have written
                     26: your program in single precision, see "help f77 to_double" to see what
                     27: changes are needed to use double precision.
                     28: 
                     29: There are five common errors in using NAG routines:
                     30: 
                     31: .IP "1."
                     32: Using single precision when double precision is required.
                     33: 
                     34: .IP "2."
                     35: Not checking the error flag.  Most routines have an error flag that
                     36: NAG sets to an error value when something goes wrong; it is your
                     37: responsibility to check it.
                     38: 
                     39: .IP "3."
                     40: Omitting arguments.  NAG routines often have many
                     41: arguments that are required even if you don't use the result.
                     42: 
                     43: .IP "4."
                     44: Not dimensioning arguments.  Especially make sure work areas are
                     45: appropriately dimensioned.
                     46: 
                     47: .IP "5."
                     48: Supplying incorrect 'row dimension' of matrices.  Many NAG routines
                     49: have a calling sequence of the form:
                     50: 
                     51:              call sub( a, ia, n, ... )
                     52: 
                     53: where 'a' is a matrix, 'ia' is it's first dimension as declared in
                     54: the calling program, and 'n' is it's order.
                     55: \&'ia' and 'n' are often different and must be specified correctly.

This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.