This is another pretty inefficient way to determine PI. This implementation is motivated by two Youtube video (Video 1, Video 2) by the channel 3Blue1Brown. This is a simple implementation that calculates the trajectory and the number of collisions. There is also a paper about this: Galperin, Gregory. "Playing pool with π (the number π from a billiard point of view)." Regular and chaotic dynamics 8.4 (2003): 375-394.. While I tried to parallelize the velocity updating, I realized that it triples the calculation time.

TODO:

• ISSUE For mass ratios above 1:10**8 the program give a wrong results.
• Export data in two csv files.
• Adapt trajectory.py to the csv input.