From: rellwood <rellwood AT aludra DOT usc DOT edu>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Collision Detection
Date: 30 Apr 1997 11:45:19 -0700
Organization: University of Southern California, Los Angeles, CA
Lines: 27
Sender: rellwood AT aludra DOT usc DOT edu
Message-ID: <Pine.SV4.3.94.970430112659.17180D-100000@aludra.usc.edu>
References: <MPG DOT dcddb78f470c89c98968a AT news DOT demon DOT co DOT uk> <Pine DOT SV4 DOT 3 DOT 94 DOT 970428125700 DOT 28102C-100000 AT aludra DOT usc DOT edu> <5k75ea$j15 AT news DOT ox DOT ac DOT uk>
Reply-To: rellwood <rellwood AT aludra DOT usc DOT edu>
NNTP-Posting-Host: aludra.usc.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
In-Reply-To: <5k75ea$j15@news.ox.ac.uk>
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

On 30 Apr 1997, George Foot wrote:

> rellwood (rellwood AT aludra DOT usc DOT edu) wrote:
> 
> : The best method is the slowest, and that is to AND each nonzero pixel in
> : sprite A with each nonzero pixel in sprite B.  If any of the ANDs return
> 
> I think you mean OR.

No I meant AND.  If OR was used, the function would return a true if it
encountered a pixel from sprite A *OR* sprite B in a given location, which
is not necesarially a collision.  If AND was used it would only return
true if it encountered a pixel from sprite A *AND* sprite B in a given
location, which always signifies a collision between the two sprites.

> The technique with circles is not necessarily more accurate; if the sprite

You're right here.  I tend to think of sprites as being generally more
circular in design then rectangular, but of course this isn't always the
case!

-Richard Ellwood



--