POV-Ray : Newsgroups : povray.advanced-users : Spline length/constant speed along spline. : Re: Spline length/constant speed along spline. Server Time
30 Jul 2024 08:17:31 EDT (-0400)
  Re: Spline length/constant speed along spline.  
From: Josh English
Date: 26 May 2000 14:43:27
Message: <392EC5D9.865E72D4@spiritone.com>
Actually, I have succeeded.

I couldn't follow everything you had, and the system of equations was a bit beyond
what I'm capable
of. Once I figured out what you were doing and I followed your logic, I applied it to
what I already
know, bezier curves.

I took an array of position vectors, and came up with a method of generating the
control points for
the bezier curves, and then I create an array of bezier curves, and voila, I have a
series of
predefined points being connected by a single path.

I couldn't have done it without that document, tho. When it comes down to it, a bezier
curve is a
third degree polynomial, so I can easily calculate the first and second derivative. I
know I have
first derivative continuity, but I am not convinced that my method guarentees second
derivative
continuity yet, so if I tried to animate a point through this series of curves it
would have a speed
different that what I'd expect. That's assuming that it crosses every point at equal
intervals. I am
now working on calculating "timestops" along the points, which should allow for a
continuous speed
along the spline.

Thanks for the document. I am keeping the goal to understand it enough on my list, so
someday it will
make sence for me.

Josh

Serge LAROCQUE wrote:

> I was just thinking, since this cubic spline algorithm requires solving a set of
linear
> equations, I don't know if you could use it directly in a POV-Ray scene file, you
might need to
> get a spreadsheet to calculate and export the points...... not the greatest if
you're going to
> change it a lot, or animate. I suppose you could use the VB macros in an Excel
spreadsheet to
> export a set of points that would change over time, but again you would have to
ensure you export
> one for each clock value................... :-/ Depends on what you want to do with
the algorithm
> I guess :-) BTW did you look at the spreadsheet I included? You can change the
y-values, not the
> x-values though.

--
Josh English
eng### [at] spiritonecom
"May your hopes, dreams, and plans not be destroyed by a few zeros."


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.