Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!elroy.jpl.nasa.gov!ncar!gatech!ncsuvx!mcnc!ecsgate!stat.appstate.edu!c_s244010117
From: c_s244010117@stat.appstate.edu
Newsgroups: comp.sys.handhelds
Subject: Re: Questions on 48: diff eq's and algebra
Message-ID: <387.27a49897@stat.appstate.edu>
Date: 29 Jan 91 01:33:11 GMT
References: <1991Jan27.172808.20069@cs.dal.ca> <27a3d4cc:1826.1comp.sys.handhelds;1@hpcvbbs.UUCP>
Organization: Appalachian State University
Lines: 14

> the 48 uses patern matching of built in (and user defined) functions 
> to symbolically integrate/differentiate...A numerical method would 
> more than likely be necessary to solve diff-eq's...

This may be true, but there are ways of making it easier to solve higher order
d.e.'s! Using Wayne Scott's polynomial routines to find the roots of the
characteristic polynomial makes finding the homogeneous solution of a higher
order d.e. a breeze! If you have a non-homogeneous equation that can be solved
using undetermined coefficients, all you have to do is enter your d.e. and your
solution with undetermined coefficients [store this one in Y if using
dX(dX(Y))]. Make sure your d.e. is at stack level 1, and evalutate it until it
doesn't change. Equate coefficients and solve the resutling equations! Its
easier than it sounds, and it is really helpful. As far as first order d.e.'s, I
don't know how to accomplish these.........