TITLE: Balls
NAME: Martin Vilcans
COUNTRY: Sweden
EMAIL: marvil@algonet.se
WEBPAGE: http://www.algonet.se/~marvil/
TOPIC: Physics and math
COPYRIGHT: I SUBMIT TO THE STANDARD RAYTRACING COMPETITION COPYRIGHT.
JPGFILE: balls.jpg
ZIPFILE: balls.zip
RENDERER USED: 
    Povray for Windows 3.0

TOOLS USED: 
    paper and pencil

RENDER TIME: 
    21 hours, 18 minutes

HARDWARE USED: 
    Pentium-166, 32 MB


IMAGE DESCRIPTION: 


A pool player tries to take an intellectual approach to the game. It's all
about physics and math.


DESCRIPTION OF HOW THIS IMAGE WAS CREATED: 


Since billiard balls is one of the standard raytracing objects (because
they're so easy to make), I had to think of some way to make the image a
bit more original. For this reason, and because I wanted the connection
to the subject to be clearer, I added the lines drawn with white chalk
on the table (don't try this at home), and used mathematical signs on
the balls (note that the sign on the black ball is an infinity sign, not
an eight). I also put in a metallic ruler and a plastic protractor.

To get the correct measurements on the table and the balls, I searched
the Web for data, and found some information from the Billiard Congress
of America, which wasn't very easy to understand (hey, I don't even know
the difference between pocket billiard, carom billiard and american
snooker!). I think I got it right anyway.

The chalk lines are made with a pigment map with a gradient across the
width of the line. Closest to the centre of the line, the pigment uses a
high frequency bozo pattern with lots of turbulence, where half of the
color map is white and the rest is clear. In the outer edges of the
line, I use the same bozo battern, though 900f the pigment is clear
and the rest is white. This way, there is more white in the centre of
the line than in the edges.

The ruler and the protractor uses a lot of polygons for the lines. On
the ruler, it would have been better to make them with a pigment map,
which would not slow down the rendering as much. The figures on the
ruler and the protractor are rendered with the text statement.

To make the scene more realistic, I added a whole room around the table.
There are two windows on the right, which are visible only as reflections
on the balls. That's a shame, because I put in quite a lot of work on them
(each window consists of over 30 primitives, which is a bit overkill for
this scene). But composition rules, so the windows are not visible.
Neither is the roof or ceiling of the room. It's possible to change the
camera angle by changing "#declare cam=1" to "#declare cam=5" in Balls.pov
and rerendering the scene to see more of the room. I'll put that image on
my web page.

Above the table, I put a big metallic reflector containing six light
bulbs. Each of these light bulb is a light source, and there is a
big area light to simulate the light from the reflector. There is one
more area light behind and to the right of the camera, then there is
a slightly more bluish light (sunlight) coming in through the window.
All in all, there are lots of light sources to slow down the rendering,
but it looks good with soft shadows and shadows across the table. The
table would look boring without them.

If you have any comments on this image, I'd love to hear them! Use the
comment field in the voting form or send me a mail at marvil@algonet.se.

Thank you.


Martin Vilcans