Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!purdue!lerc.nasa.gov!kira.cc.uakron.edu!malgudi.oar.net!news.sprintlink.net!newsfeed.internetmci.com!uunet!in1.uu.net!hearst.acc.Virginia.EDU!murdoch!viper.cs.Virginia.EDU!ccb8m
From: ccb8m@viper.cs.Virginia.EDU (Charles C. Bundy)
Subject: Re: collision processing
X-Nntp-Posting-Host: viper-fo.cs.virginia.edu
Message-ID: <DCy9Kn.5qF@murdoch.acc.Virginia.EDU>
Sender: usenet@murdoch.acc.Virginia.EDU
Organization: University of Virginia Computer Science Department
References: <Ted.Belding-3107951722410001@pm051-14.dialip.mich.net> <1995Aug1.090158.2966@prim.demon.co.uk> <pnelson.211.005D0480@lagoon.ultranet.com>
Date: Mon, 7 Aug 1995 16:53:11 GMT
Lines: 78

In article <pnelson.211.005D0480@lagoon.ultranet.com> pnelson@lagoon.ultranet.com (Peter Nelson) writes:
>In article <1995Aug1.090158.2966@prim.demon.co.uk> Dave Griffiths <dave@prim.demon.co.uk> writes:
>
>>In article <Ted.Belding-3107951722410001@pm051-14.dialip.mich.net> Ted.Belding@umich.edu (Theodore C. Belding) writes:
>>>In article <pnelson.199.006FC580@lagoon.ultranet.com>,
>>>pnelson@lagoon.ultranet.com (Peter Nelson) wrote:
>>>> The problem I'm having is with "collisions".  A collision
>>>
>>>You may want to take a look at the parallel algorithm literature
>>>in computer science, especially papers on write collisions.
>
>>Not sure how useful that would be. If you consider the billiard ball problem
>>and imagine how it would be solved for multiple processors, each ball would
>>be handled by a separate processor.
>
>  The problem I'm having with this whole discussion is 
>  that people seem to have the attitude that we should
>  be able to discuss these things and solve them off
>  the top of our heads.   I appreciate folks taking the
>  time to do so, but it shouldn't be necessary:

You have a valid point but there IS more literature than one person or
group could ever hope to have access to.  Is it unreasonable to post a
question and bandy about answers/references?  What if no one in the group
has seen an article in question?  Does that mean it doesn't exist? No, but
it certainly doesn't imply that IT DOES.  One could waste enormous time
looking for a paper which describes a solution, rather than rolling up
one's sleeves and concoting a solution.  The phrase "It shouldn't be
necessary" doesn't work.  There is a balance between time spent looking
and time spent solving.  If as you say there is "needless" work spent
re-inventing the wheel, take solace in that fact that there will at least
be variation and NEW papers for you to look up :).

Question: Did Foley and Van Dam JUST serve as editor/compilers of knowledge
          or did they do things as well, and in the course of doing re-invent
          the wheel?

>
>  As I mentioned, my background is in computer
>  graphics, and while individual designers might 
>  redesign or tweak certain standard approaches to
>  suit their particular architectures or needs, the fact
>  remains that you can pick up a Foley and Van Dam
>  or various other texts and read about standard 
>  algorithms that have already been developed and 
>  widely used to solve the basic problems like hidden
>  surface removal, lighting/shading, scanline conversion,
>  transforms, etc.   The strengths and weaknesses of
>  these different algorithms are all discussed or described.

Yes, but collision detection can be an open ended question which doesn't
fit the graphics display modality.  In this case it sounds like the
complete universe is affected, rather than a local space case of two
"space ships" colliding.  However the person doing the simulation
should take heed of your reference and see if it suggests a solution.

>
>  Simulation, complex systems modeling, and ALife
>  have been around for years now, and there has been a 
>  great deal of collaborative work done.   So it shouldn't
>  be necessary to reinvent this stuff from whole cloth by
>  now in 1995.    I just want to know where I can see

2000, 2010, et-al.  It's the way the world works, you think the year 1995
is some intellectual pinnacle?  Hoo Boy....

>  a summary/description of what the major algorithmic
>  approaches are.  
>
>PS  -  BTW, if there *is* no such place then this would
>suggest that maybe this field would advance faster if
>people COULD learn from the work of others.   Most

Duh, what do you suppose forums such as comp.ai.* are!?  Check out ACM,
AAAI and IEEE sigs.

Charles C. Bundy IV
ccb8m@preferred.com
