03/06/84  fortran hexadecimal mode

This segment describes Fortran's hexadecimal mode and how it is used.


Explanation of hexadecimal mode:

A single word floating point number is represented on Multics as a
signed 8-bit exponent followed by a signed 28 bit mantissa.  The value
of the number is normally the mantissa multiplied by the value of 2 to
the power of the exponent.  This gives a numerical range between
1.469E-39 and 1.701E+38 inclusive.  In hexadecimal mode, the value of
the number is the mantissa multiplied by the value of 16 to the power
of the exponent.  The range of hexadecimal numbers is between
1.864E-155 and 8.379E+152 inclusive.


Invoking hexadecimal mode for Fortran:

The control argument "-hexadecimal_floating_point", or "-hfp" for
short, instructs the compiler to generate code that uses hexadecimal
floating point numbers.  This option can also be specified in the
source in a "%global" statement.  (e.g.  "%global hfp").  The default
option is "-binary_floating_point" or "bfp".

This feature is only supported by the DPS8 hardware.  Compilations
using the "-hfp" control argument and the execution of any such
programs, requires "rw" access to >sc1>admin_acs>Fortran_hfp.acs.  This
segment must be created by site personnel if they wish to make use of
HFP.


Use of hexadecimal mode:

Binary floating point numbers cannot be mixed with hexadecimal floating
point numbers.  The "hexadecimal_floating_point" option always applies
to the entire compilation unit.  Programs compiled in hexadecimal mode
differ from those compiled in binary mode only in their use of real,
complex and double precision numbers.  Integers are unaffected.


Pitfalls:

If Fortran programs compiled in hexadecimal mode pass or receive
floating point arguments of Fortran programs compiled in binary mode,
the results are unpredictable.  Also, as Fortran is the only standard
Multics language that supports hexadecimal floating point arithmetic,
hexadecimal arguments cannot be reliably passed to any program compiled
in another language.

In gaining the extra magnitude, hexadecimal numbers suffer from less
precision.  Generally, hexadecimal mode results in a loss of 3 bits of
reliable precision in the mantissas of floating point variables.


                                          -----------------------------------------------------------


Historical Background

This edition of the Multics software materials and documentation is provided and donated
to Massachusetts Institute of Technology by Group BULL including BULL HN Information Systems Inc. 
as a contribution to computer science knowledge.  
This donation is made also to give evidence of the common contributions of Massachusetts Institute of Technology,
Bell Laboratories, General Electric, Honeywell Information Systems Inc., Honeywell BULL Inc., Groupe BULL
and BULL HN Information Systems Inc. to the development of this operating system. 
Multics development was initiated by Massachusetts Institute of Technology Project MAC (1963-1970),
renamed the MIT Laboratory for Computer Science and Artificial Intelligence in the mid 1970s, under the leadership
of Professor Fernando Jose Corbato. Users consider that Multics provided the best software architecture 
for managing computer hardware properly and for executing programs. Many subsequent operating systems 
incorporated Multics principles.
Multics was distributed in 1975 to 2000 by Group Bull in Europe , and in the U.S. by Bull HN Information Systems Inc., 
as successor in interest by change in name only to Honeywell Bull Inc. and Honeywell Information Systems Inc. .

                                          -----------------------------------------------------------

Permission to use, copy, modify, and distribute these programs and their documentation for any purpose and without
fee is hereby granted,provided that the below copyright notice and historical background appear in all copies
and that both the copyright notice and historical background and this permission notice appear in supporting
documentation, and that the names of MIT, HIS, BULL or BULL HN not be used in advertising or publicity pertaining
to distribution of the programs without specific prior written permission.
    Copyright 1972 by Massachusetts Institute of Technology and Honeywell Information Systems Inc.
    Copyright 2006 by BULL HN Information Systems Inc.
    Copyright 2006 by Bull SAS
    All Rights Reserved