[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference rusure::math

Title:Mathematics at DEC
Moderator:RUSURE::EDP
Created:Mon Feb 03 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2083
Total number of notes:14613

908.0. "Test problems for Mathematica" by PBSVAX::COOPER (Topher Cooper) Tue Jul 26 1988 12:32

    <<This note is being posted to both the Math and Maple notes files>>

    There is an off chance that I may have a chance to attend a demo of
    Mathematica the new, heavily hyped, algebraic manipulation system.
    What I'm looking for is some deceptively hard (i.e., easy to state
    but tricky to solve) algebra problems.  If I don't attend I'll forward
    any proposed problems to the people who are officially evaluating
    Mathematica for DEC.

				    Topher
T.RTitleUserPersonal
Name
DateLines
908.1What kind of problems was that?POOL::HALLYBThe smart money was on GoliathTue Jul 26 1988 16:4713
    How about:
    
"Solve X^n + Y^n = Z^n for positive integers X,Y,Z,n; n>2" I have a wonderful prrof of this...
    
    What exactly do you mean by "algebra problems"?
        
    How about multiplying two long irreducible polynomials together,
    and asking it to factor them?
    
    How about some Diophantine equations -- easy to create, easy to
    check, hard to solve.  Sometimes impossible, as in X�+X = 4Y+1.

      John
908.2Tricky, but not impossible or very long.PBSVAX::COOPERTopher CooperTue Jul 26 1988 17:3828
    The idea is to be able to give it some problems which a really good
    system would significantly aid the operator in solving within a
    reasonable amount of time (within the time reasonably alloted to
    a demo, say 10 minutes or so).
    
    In theory, algebra includes everything through calculus and
    differential equations.  Anything is fair game within that broad
    domain, including sensitive numerical problems.  However, no
    "package" may exist for solving particular classes of problems.
    
    For example, Diophantine equations (the possible to solve ones)
    would be excellent, except as of the date of the actual writing
    of the published manual, there is no package for solving Diophantine
    equations.  The only mention (according to the index) is that
    the function ExtendedGCD (which gives the GCD and the two multipliers
    which produce the two arguments as result in one step) is documented
    as being "important in finding integer solutions to linear
    (Diophantine) equations."  So that is a good suggestion, but, as
    it happens one which I probably can't use, except where the steps
    needed to solve it are likely to be obvious to a good demonstrator.
    (Maple, I might add, does have a routine for solving Diophantine
    equations of the proper, fairly general, form).
    
    The polynomial factoring problem is a good one which should be usable.
    
    Thanks.
    
    					Topher
908.3How about thisPOOL::HALLYBThe smart money was on GoliathTue Jul 26 1988 18:251
    A good one to try is integral exp(-x�) dx over the reals.