I am curious whether anyone has done large-scale simulations of point masses in 3-space with random initial positions and velocities, under Newtonian mechanics & gravity.

I'm especially curious whether that's enough for galaxy-shaped objects to appear.

--Dan