Path: chuka.playstation.co.uk!news
From: Craig Graham <cgraham@intelligent-group.com>
Newsgroups: scee.yaroze.programming.3d_graphics
Subject: Re: Rotation, ApplyMatrix Query (fairly long)
Date: Mon, 23 Mar 1998 12:30:57 +0000
Organization: Intelligent Research
Lines: 25
Message-ID: <35165600.B98FF682@intelligent-group.com>
References: <34F3EFA2.4CD2@127.0.0.1> <350F949B.93337AA2@immd9.informatik.uni-erlangen.de> <351131B3.FC7030EB@intelligent-group.com> <35125A99.CB324946@immd9.informatik.uni-erlangen.de>
NNTP-Posting-Host: 194.131.235.3
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.03 [en] (Win95; I)

Christoph Luerig wrote:

> Hello Craig!
>
> Especially in this case I would not use a finite series approximation, but
> a newton method based iteration scheme.
>

<SNIP : Deleted the method for doing the iterative method>

> Christoph Luerig
>

Iteration wouldn't be as fast as the FSA though. I use the FSA for
moving point lightsource calculation, and speed is the important thing
rather than accuracy for a moving point source (eg. a plasma ball, or a
static light illuminating a moving object for a brief time).
For static light effects, where they're going to be the same for long
enough to notice any discrepencies, I would pre-calculate with an exact
integer sqrt() function anyway (eg. a lamp on a wall, where the circle of
illumination should be uniformly graduated 'coz you can stand and look at it).

Craig.