Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84 SMI; site sun.uucp
Path: utzoo!linus!philabs!prls!amdimage!amdcad!decwrl!sun!dgh
From: dgh@sun.uucp (David Hough)
Newsgroups: net.micro,net.micro.68k,net.micro.16k
Subject: Re: Floating Point Comparisons
Message-ID: <2126@sun.uucp>
Date: Tue, 16-Apr-85 21:39:04 EST
Article-I.D.: sun.2126
Posted: Tue Apr 16 21:39:04 1985
Date-Received: Sat, 20-Apr-85 05:50:42 EST
References: <133@cfa.UUCP> <6370@boring.UUCP> <2574@nsc.UUCP> <227@osiris.UUCP> <930@cornell.UUCP>
Reply-To: dgh@sun.UUCP (David Hough)
Distribution: net
Organization: Sun Microsystems, Inc.
Lines: 13
Xref: linus net.micro:8739 net.micro.68k:657 net.micro.16k:290
Summary: 

In article <930@cornell.UUCP> solworth@gvax.UUCP (Jon Solworth) writes:
>
>	The NS16XXX times for longer addition time than multiplication are
>not unreasonable from the scientific programmer point of view.  Multiplies
>vastly outnumber adds in scientific code.
>
Really?  Are there any published papers with this result?  I was under
the impression that linear algebra algorithms typically take essentially
equal numbers of floating point additions and multiplications,
with comparatively few other floating point operations.  Most scientific
computation gets reduced to linear algebra sooner or later.

David Hough