Developer Forum

This is a problem that really has nothing to do with the C++ language, which
is what is discussed here. You might try a *** or simulation newsgroup
(although I have no idea what those might be), or do some searches on
groups.google.com, using terms like "simulation", "modeling", and
"collision".

That said, making some very strict assumptions (as you have), it is possible
to make the model totally deterministic, such that the same initial
conditions would result in the same "final" state, but any formula to
compute that final state would be horrendously complex (assuming it's
possible at all). Most likely, you'd have to rely on simulation techniques.

One approach might be to find the "next" collision (in time) from any given
set of conditions, then move everything to the state they'd be in at that
time, make your changes to the state of the balls due to the collision, and
repeat the process. At the point where the "next" collision comes later in
time than the desired final state time, step forward just to the final state
time and report your results.

How to calculate the "next" collision, and how to make the transformation(s)
at that point in time, are, as they say, exercises left to the reader. :-)

-Howard

> groups.google.com, using terms like "simulation", "modeling", and

I have done those searches. Everyone seems to be using iterative
aproximations.


This is what I am interested."Is it possible?" "If so, how" "if not, why
not"

Also apologies. I know this is the wrong group for this question. I tried in
alt.math and alt.sci.physics but most people didn't understand the question.
You did, so I answered here :-p

For a good reason.


Theoretically it is possible. The billiard players do it all the time.
But as said: taking everything into account it will get tremendously
complex.

You might start with simplifying your billiard model. Personally I would
drop the problem of accounting for rotating balls first. They induce a drag
and the balls don't wander on straight lines any more. But straight lines
simplify the model, since their intersections can easily be computed.

The main problem in simulation is always the same: Reality is much to
complex to do a complete simulation. You need to accept some simplifications
all the time.

--
Karl Heinz Buchegger
XXXX@XXXXX.COM