Newsgroups: comp.ai.games
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!portc02.blue.aol.com!howland.erols.net!agate!nntpfeed.doc.ic.ac.uk!sunsite.doc.ic.ac.uk!lyra.csx.cam.ac.uk!warwick!bradford.ac.uk!leeds.ac.uk!news
From: csxgti@scs.leeds.ac.uk (G T Irwin)
Subject: Re: Tetris
Message-ID: <1996Dec16.224433.26817@leeds.ac.uk>
NNTP-Posting-Host: csgi44.leeds.ac.uk
Organization: The University of Leeds, School of Computer Studies
Date: Mon, 16 Dec 1996 22:44:33 +0000 (GMT)
References: <56e636$i3@news1.mnsinc.com><56eb9i$rnl@nef.ens.fr>
    <1996Nov18.135616.14531@leeds.ac.uk>
    <01bbd650$922beaa0$886c4fce@KMEISSNER.activision.com>
    <1996Nov25.112800.4173@leeds.ac.uk>
    <262954492wnr@robwilde.demon.co.uk>
    <329efebf.2202712@news.intranet.ca>
    <139344567wnr@robwilde.demon.co.uk>
Lines: 60

In article <139344567wnr@robwilde.demon.co.uk>, Robert Wilderspin <Rob@robwilde.demon.co.uk> writes:
> In article: <329efebf.2202712@news.intranet.ca>  ttoyooka@verisim.com (Takashi 
> Toyooka) writes:
> > 
> > Well, sometimes, you will want to slip a piece under another one, in which case,
> > you may need to move in one direction, and then back again.  However, usually,
> > that happens (to me) only if I'd made a mistake previously.  Still...
> 
> Yes, but the original poster was going on about moving to the left and then to the 
> right before the blocks drops down a level, so that it's just jiggling around in the 
> same place.  You don't do that when sliding underneath another block.
> -- 
> Robert Wilderspin
> 

The point I was making (perhaps you took what I said too literarlly) is that of
all the solution to this problem I have found none will concider sliding pieces
under overhangs. And _yes_ this situation can easily occur without making a
mistake.

e.g.	|	   |
	|	   |
	|	   |	It can happen on the very first move.
	|	** |
	|________**|

The fact is, if the computer decides a particular position is the best and it
is feasable that the shape can be manovered into that position, no matter how
complex it should be able to do it. How can you ever begin to design a computer
player of this game if even the most trival manovers cannot be achived by a
computer.

e.g.	|	  *|
	|     *	***|
	|  ********|
	|     *****|
	|*	***|
	|*****	***|
	|* ***	***|	What about this situation? I knwo its a little contrived
	|*  **	** |	but if by accident the computer ended up in this state
	|*****___**|    surely it should be able to get out.


Why bother spending ages designed the most scientifically amazing solution if
you don`t even spend any time allowing it to carry out ALL of its decsions.

Its like making the worlds finest sportcar only to decide that theres no point
putting a steering wheel in it because sportcar drivers drive on straight roads
most of the time!


Anyway, I have now solved the above problem, and am progressing to the later
stages of my project.

-- 
-----------
Greg Irwin                    
                                 
e-mail : csxgti@scs.leeds.ac.uk  
         csc4gti@sgi.leeds.ac.uk 
