That's super interesting! What program are you using to simulate that?
Thanks :). This is a custom bit of code for a homework assignment written in Python using an "adaptive" (in time-step) Runge Kutta method. The prof wrote some of it to give us a general direction, and then we had to fill in all of the details. It's only a couple hundred lines of code, nothing too crazy.
For Runge-Kutta, you start off with some initial "state", a set of spatial coordinates and velocity components for each body, and then use a differential equation to evolve the system in time. The solution of the differential equation governs how the bodies move. In other words; you take a (small) step forward in time, take stock of where everything is, calculate the force on each body due to the others, calculate the resulting acceleration, and then repeat the process. It can do as many bodies as you like, but the computational intensity increases explonentially (accidentally typed that word out wrong, but I'm keeping it) with each additional body.