Newsgroups: comp.sys.ibm.pc,comp.unix.i386
Path: utzoo!utgpu!jarvis.csri.toronto.edu!csri.toronto.edu!wayne
From: wayne@csri.toronto.edu (Wayne Hayes)
Subject: Re: What would you use for Fast Floating-Point?
Message-ID: <1989Jun29.141824.10236@jarvis.csri.toronto.edu>
Keywords: Floating-point,Coprocessors
Organization: University of Toronto, CSRI
References: <2105@randvax.UUCP> <1216@rlgvax.UUCP> <5346@pt.cs.cmu.edu>
Distribution: na

In article <5346@pt.cs.cmu.edu> ralf@b.gp.cs.cmu.edu (Ralf Brown) writes:
>In article <1216@rlgvax.UUCP> golds@rlgvax.UUCP (Rich Goldschmidt) writes:
>}  32 bit integer math: 2.4 MFlop  your time to convert to integer math
           ^^^^^^^^^^^^

Would someone explain this? If you can get such fantastic performance
with integer math, why doesn't everyone use it? (I'm assuming there's some
quick way to convert to and from integer representation to floating point.)
I've tried my own funny storage, BCD and the like, but it always comes
out slower than regular floating point.

------------------------------------------------------------------------------
"Open the pod bay doors, HAL."   "I'm sorry Dave, I'm afraid I can't do that."
Wayne Hayes	INTERNET: wayne@csri.toronto.edu	CompuServe: 72401,3525