Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!zephyr.ens.tek.com!tektronix!sequent!mntgfx!mbutts
From: mbutts@mentor.com (Mike Butts)
Newsgroups: comp.arch
Subject: Re: Reciprocal approximation (was Intel 860 Architecture)
Message-ID: <1989Dec14.184714.1473@mentor.com>
Date: 14 Dec 89 18:47:14 GMT
References: <8991@cbmvax.commodore.com>
Organization: engr
Lines: 20

> In article <112400013@uxa.cso.uiuc.edu> afgg6490@uxa.cso.uiuc.edu writes:
>>
>>..> Reciprocal approximation
> 
> [list of machines that use NR approx to do divides...]
> 
All Floating Point Systems 64-bit machines (164, 264, M64-xxx) use
reciprocal approximation, as does the Mentor Graphics Compute Engine.

I recall being told (years ago, now) by algorithmic folks that much work in
scientific computing over the years has been directed towards finding
algorithms that minimize the need for division, because it is so much more
expensive (in time and/or HW) than multiplication.  Given the existence of
such, architects of scientific CPUs found reciprocal lookup tables to be far
preferable to FP divide hardware in speed and simplicity.
-- 
Michael Butts, Research Engineer       KC7IT           503-626-1302
Mentor Graphics Corp., 8500 SW Creekside Place, Beaverton, OR 97005
!{sequent,tessi,apollo}!mntgfx!mbutts         mbutts@pdx.MENTOR.COM
Opinions are my own, not necessarily those of Mentor Graphics Corp.