Xref: utzoo comp.graphics:5552 rec.games.programmer:858
Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!tank!uwvax!astroatc!philm
From: philm@astroatc.UUCP (Phil Mason)
Newsgroups: comp.graphics,rec.games.programmer
Subject: Re: 2D Maze in 3D
Message-ID: <1875@astroatc.UUCP>
Date: 5 May 89 21:04:10 GMT
References: <21795@santra.UUCP>
Reply-To: philm@astroatc.UUCP (Phil Mason)
Organization: Astronautics Technology Cntr, Madison, WI
Lines: 33

In article <21795@santra.UUCP> jmunkki@kampi.hut.fi (Juri Munkki) writes:
>I'd like to write a game where the player can wander in a maze and
>see the movement in 3D perspective. 
>I've thought of some algorithms, but someone must have come up with
>an absolutely amazing solution... I need a really fast algorithm.

I have a program written by Brad Myers (with Scott Brown) called "GetOut"
that very effectively simulates a 3-D display of a run through a 2-D maze.

Sorry, it only runs on the PERQ workstation (remember the PERQ?).  I do
have source; it was written in PASCAL.  It is copyrighted so I would not be
able to distribute it.

The PERQ is a microcoded 170 ns machine with hardware Raster-OP.  Display
size is 8-1/2 x 11 (portrait) or 14 x 11 (landscape).  I do not believe the
program used custom microcode.  Actually, the program is cleverly written
and is not overly complex.  The most difficult part (it seems to me) is the
generation of a one-exit maze with corridors and not very many loops.

You can generate realistic wire frame corridors with good heuristics.  Do
the hidden line elimination by heuristic methods as well.  Don't do brute
force calculations if you wish to run the maze in real-time.

The PERQ implemetation is great - too bad you can't see it run.




-- 
Kirk  : Bones ?                |  Phil Mason, Astronautics Technology Center
Bones : He's dead Jim.         |  Madison, Wisconsin - "Eat Cheese or Die!"
...seismo-uwvax-astroatc!philm |  I would really like to believe that my
...ihnp4-nicmad/               |  employer shares all my opinions, but . . .