Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site ut-ngp.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!unc!mcnc!decvax!linus!philabs!cmcl2!seismo!ut-sally!ut-ngp!lindley
From: lindley@ut-ngp.UUCP (John L. Templer)
Newsgroups: net.math
Subject: Re: real roots of polynomial
Message-ID: <1311@ut-ngp.UUCP>
Date: Tue, 12-Feb-85 17:38:01 EST
Article-I.D.: ut-ngp.1311
Posted: Tue Feb 12 17:38:01 1985
Date-Received: Fri, 15-Feb-85 05:48:36 EST
References: <7942@brl-tgr.UUCP> <9600001@uiucdcsp.UUCP>
Organization: U.Texas Physics Department; Austin, Texas
Lines: 22

From ashby@uiucdcsp.UUCP:

> One reliable and robust means of computing the roots of
> a polynomial is to find the eigenvalues of the polynomial's
> companion matrix. 
> 
> First set up the companion matrix; see any linera algebra
> text for a definition.  Then just compute that matrix's
> eigenvalues, via EISPACK for example.

I'm curious;  does this method fufill the requirement that it be
efficient?  I thought finding the eigenvalues of a matrix was not
all that simple a process.

-- 

                                           John L. Templer
                                     University of Texas at Austin

    {allegra,gatech,seismo!ut-sally,vortex}!ut-ngp!lindley

                 "and they called it, yuppy love."