[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Faster floating point



   Subject:      Faster floating point
I have a utility which may be of interest to anyone who needs faster floating
point math in MACL.  It is called FPC (Floating Point Compiler) and it takes
ordinary floating point expressions, compiles them into native coprocessor and
builds a lisp interface to the function.  Speedups of 10x or more are not
uncommon depending on the complexity of the function.  If anyone is interested
in this I will send you a copy in exchange for comments and suggestions.

Erann Gat

gat@robotics.jpl.nasa.gov