```From: "Trevis Crane" <tcrane@uiuc.edu>
Path: news.mathworks.com!newsfeed.mathworks.com!webx
Newsgroups: comp.soft-sys.matlab
Subject: Re: MATLAB Programming Contest: November 3-10, 2004
Message-ID: <eeefd44.42@webx.raydaftYaTP>
Date: Thu, 4 Nov 2004 12:55:09 -0500
References: <eeefd44.-1@webx.raydaftYaTP> <eeefd44.41@webx.raydaftYaTP>
Lines: 83
NNTP-Posting-Host: 130.126.102.68
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Xref: news.mathworks.com comp.soft-sys.matlab:239593

The solver.m file that was supplied is completely random, so there's
no reason to expect it to pick out the "right" moves as they were
chosen on the rules page. Hence on average, that algorithm requires
more moves in order to arrange the furniture.

Concerning the increment tables -- what you call North, South, etc.
is completely irrelevant as long as you have a complete and
non-degenerate set. For ease of use in submission, just go with what
they supplied and then you won't have to worry about it.

trevis

Jin wrote:
>
>
> !!!A Big Confused Problem for me!!!
>
> I find so called "increment tables" in the testsuite(smaple), are
> not
> consistent with the explanation on the contest's homepage.
> (but no body submit this problem?)
> for instance, I run the following code(testsuit zip file is
> Updated^_^):
>
> ai = [ 4 0 0 3 ;...
> 0 0 0 0 ;...
> 1 2 0 0 ];
>
> af = [ 0 0 0 4 ;...
> 0 2 3 0 ;...
> 0 1 0 0 ];
>
> w = [ 5;10;20;1 ];
>
> mv = solver(ai,af,w)
>
> a result is:
> mv =
>
> 3 3
> 3 1
> 3 2
> 2 4
> 3 2
> 3 3
> 3 4
> 1 1
> 4 1
> 4 1
> 4 1
>
> but in the hompage:
> "
> One solution is
> move = [ 3 4
> 3 3
> 4 2
> 4 2
> 4 2
> 2 1
> 1 2]
>
> "
>
> see [2 4] vs [2 1]
> block#2 shold go where?
>
> In fact:
> according to the homepage, the true "increment tables" is :
> % N=1, E=2, S=3, W=4
> % By Matrix Direction
> I = [-1 0 1 0];
> J = [0 1 0 -1];
>
> not the testsuit solver's:
> % N=1, E=2, S=3, W=4
> I = [0 1 0 -1];
> J = [1 0 -1 0];
>