[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
arbitrary precision floating point
I've started to put together an interface to the system described
below from Allegro CL. If you are interested in seeing the
results and/or participating in some way, send me a note.
The original code, in Fortran, is unchanged. For you to use
it you would need to be able to get hold of a fortran compiler.
.... from the tech report...
MPFUN: A Portable High Performance Multiprecision Package \\
David H. Bailey \\
April 3, 1991
The author has written a package of Fortran routines that perform a
variety of arithmetic operations and transcendental functions on
floating point numbers of arbitrarily high precision, including large
integers. This package features (1) virtually universal portability,
(2) high performance, especially on vector supercomputers, (3)
advanced algorithms, including FFT-based multiplication and
quadratically convergent algorithms for $\pi$, exp and log, and (4)
extensive self-checking and debug facilities that permit the package
to be used as a rigorous system integrity test.
This paper describes the routines in the MPFUN package and includes
discussion of the algorithms employed, the implementation techniques,
performance results and some applications. Notable among the
performance results is that the MPFUN package runs up to 15 times
faster than another widely used package on a RISC workstation, and it
runs up to 170 times faster than the other package on a Cray
The author is with the NAS Applied Research Office, NASA Ames Research
Center, Moffett Field, CA 94035.