Path: chuka.playstation.co.uk!news
From: "Steve Dunn" <steve@backroom.tm>
Newsgroups: scee.yaroze.programming.3d_graphics
Subject: Re: Collision detection
Date: Thu, 2 Apr 1998 21:02:18 +0100
Organization: PlayStation Net Yaroze (SCEE)
Lines: 167
Message-ID: <6g0qkh$6ne45@chuka.playstation.co.uk>
References: <6fqnjp$sqg1@emeka.playstation.co.uk> <6fr6mi$6ne31@chuka.playstation.co.uk> <3522121E.5527A54E@immd9.informatik.uni-erlangen.de>
NNTP-Posting-Host: userm899.uk.uudial.com
X-Newsreader: Microsoft Outlook Express 4.72.2106.4
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.2106.4

Thanks Christoph,
    Any ideas where I can access these
without becoming a member of IEEE (I
only got CSE's at school (suprising
eh?:)  )  )

Steve 'university of life' Dunn.

Christoph Luerig wrote in message
<3522121E.5527A54E@immd9.informatik.uni-
erlangen.de>...
>Hello!
>
>Here are some references, that might be
helpful:
>
>Webb R., Gigante M.: "Using Dynamic
Bounding Volume Hierarchies to
>Improve Efficiency of Rigid Body
Simulations", Visual
>Computing, Springer, pp. 825-842, 1992
>
>Pentland, A. P.: "Computational
Complexity Versus Virtual Worlds",
>Computer Graphics, 24(2): 185-192,
March 1990
>
>Moore M., Wilhelms J.: "Collision
Detection and Response from Computer
>Animation", Computer Graphics, 22(4):
289-298, August 1988
>
>Personally I would start with the last
reference.,as it intorduces to
>the problem comples of collision
detection and reaction. The second
>reference discusses the complexity
problem mentioned by Sony dev.
>support and the first one introduces to
a advanced technique, where a
>hierarchy of bounding volumes is used.
This approach may be interesting
>if you are using a very complex scene,
where little preassumptions can
>be made.
>
>
>
>
>> Hi Lewis,
>>     Thanks for the reply.
>> Couldn't find tmd.h & c in the car
demo
>> (couldn't download the car demo!  Had
to
>> get it from SCEA).
>> Found it elsewhere in other demos.
Will
>> probably have some Q's when perusing
>> through them.
>>
>> >For N objects, and potentially
>> (N.N-1)/2 pairs of objects, collision
>> >detection
>> >can be expensive;
>> Sorry, I didn't understand that
>> sentence.  Perhaps I will understand
it
>> more when I delve into things a bit
>> deeper.
>>
>> >
>> >Other than that, I suggest looking
at
>> the variety of books and
>> >websites ....
>> Any suggestions on good (and
relevant)
>> books and sites ?  I've been looking
for
>> the game programmers guru book or
>> whatever they call it.  I remember
>> seeing a dusty copy in my local PC
>> World, but some other sad gimp must
have
>> got there before me !
>>
>> Cheers
>>
>> Steve
>>
>> >          via smtpd (for
>> [194.203.13.2]) with SMTP; 31 Mar
1998
>> 13:33:39 UT
>> > by camsg001.camb.scee.sony.co.uk
>> (8.8.6/8.8.6) with SMTP id NAA24103
>> > for <news@playstation.co.uk>; Tue,
31
>> Mar 1998 13:33:24 +0100 (BST)
>> >From:
Lewis_Evans@Playstation.sony.com
>> >To: news@playstation.co.uk
>> >
>> >Generic / easily applicable 3d
>> collision detection
>> >and handling: cube, sphere,
cylinder.
>> >The first two are especially
>> straightforward to detect collisions
>> with;
>> >the only time-consuming part may be
>> using tmd-access routines
>> >to find the dimensions of your TMD
>> models
>> >(but see tmd.c, tmd.h in SCEE demo
>> <car>).
>> >
>> >For N objects, and potentially
>> (N.N-1)/2 pairs of objects, collision
>> >detection
>> >can be expensive; people usually use
>> assumptions about game world
>> >to reduce the number of tests, eg
>> having the world split into areas,
etc.
>> >For your gaem it sounds like you can
>> use 2d methods, if everything's
>> >on the plane, and you may be able to
>> use grid-square partitioning
>> >of the plane, which should help.
>> >
>> >Other than that, I suggest looking
at
>> the variety of books and
>> >websites ....
>> >
>> >
>> >          Lewis
>> >
>> >
>> >
>> >
>> >If anyone can help on collision
>> >detection (3D), I'd be very
grateful.
>> >I have a XZ world and want to check
>> >collision with any object.  The
objects
>> >can be located anywhere in the
world,
>> >and can be any size, and be at any
>> >rotation(constantly changing).
>> >Any help appreciated.
>> >
>> >Steve
>> >
>> >
>> >
>> >
>> >
>
>
>Christoph Luerig
>