home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky sci.math:17183 rec.games.abstract:652
- Path: sparky!uunet!mcsun!sun4nl!tuegate.tue.nl!rw6.urc.tue.nl!rcbaaw
- From: rcbaaw@rw6.urc.tue.nl (Angelo Wentzler)
- Newsgroups: sci.math,rec.games.abstract
- Subject: Re: Game of pentominos
- Message-ID: <rcbaaw.724692264@rw6.urc.tue.nl>
- Date: 18 Dec 92 15:24:24 GMT
- References: <1992Dec15.154734.23894@odin.diku.dk> <pete.03it@bignode.equinox.gen.nz>
- Sender: root@tuegate.tue.nl
- Reply-To: rcbaaw@urc.tue.nl
- Followup-To: sci.math
- Lines: 62
-
- pete@bignode.equinox.gen.nz (Pete Moore) writes:
-
- >Torben AEgidius Mogensen (torbenm@diku.dk) wrote:
- >>martel@marvin.mr.sintef.no (Paulo Martel) writes:
-
- >>>After several tries I gave up a combinatorial analysis of the game of
- >>>pentominos. Would someone point me to a reference, or briefly explain
- >>>how one could compute the total number of solutions for a grid of a
- >>>given size (6x10, 5x12, 4x15, 3x20).
-
- >>I saw a paper once that reported the number of solutions to each of
- >>these rectangle sizes. It used a heavily optimized machine code
- >>program to exhaustively search for all solutions. I remember that for
- >>the 3x20 case there are only two solutions barring reflections and
- >>rotations. These are quite easy to find by hand. The number of
- >>solutions for the 6x10 case was quite large, but I don't recall the
- >>number. I also don't recall the title or author of the paper.
-
- >I can't give an authoritative reference, but according to
- >some notes I made back in my young & innocent days when I used to note
- >information without references, there are:
- > 2 3x20 solutions
- > 2339 6x10 solutions
- > 1010 5x12 solutions
- > 368 4x15 solutions
-
- >If your pentominoes are 3-dimensional (i.e. each could be constructed by
- >joining 5 cubes together) rather than 2-dimensional, you can also form a
- >3x4x5 rectangular prism from the 12 pentominoes.
-
- Gee, this must be my lucky month! I like pentominoes, wrote a program
- once (who didn't) and I rank them right up there along with life,
- rubik's cube and tiled squares (See: tiling problem in this group :-)
-
-
- C.J.Bouwkamp, a professor here at the TUE has computed all solutions
- to this prism as well as to "the steps" which is a three dimensional
- staircase. He's just as mad about pentominoes as I am.
-
- The catalogues are:
-
- "Catalogue of solutions to the rectangular 3x4x5 pentomino problem"
- C.J.Bouwkamp, Eindhoven University of Technology.
-
- "Packing the steps with solid pentominoes"
- C.J.Bouwkamp, Eindhoven University of Technology.
-
- The books are available here, but not in "computerised" format.
-
-
- By the way, has anybody ever tried to write a program to compute all
- (N+1)-ominoes from the set of N-ominoes? It's not difficult:
- Just take one N-omino at a time and fit an extra square onto it. This
- produces one extra (N+1)-omino. Fit a square on all possible positions,
- (just "roll" it along the edge) then proceed with the next N-omino.
- This leaves you with a *bag* of (N+1)-ominoes, of which you now have to
- make a *set* (remove multiple copies).
- It would be smart to check for multiples right at the start though...
-
- It would take a lot of computation, but not an awful lot of time.
-
- Angelo Wentzler
-