Purpose

---

  This function calculates the potential energy of a collection of
  point masses. It is a wrapper to poten_slow and poten_tree, and
  attempts to call the most efficient program.

Inputs

---

  pos: A [3, n] array of 3D positions
  mass: An n-element vector of masses

Outputs

---

  The gavitational potential energy, given by
  PE = sum_i (sum j > i (m_i * m_j / r_ij) )

Description

---

  The poten_tree program scales as N log N, but has more overhead
  than the N^2 poten_slow algorithm. Tests on a uniform grid of
  particles suggest that the algorithms run in similar times for
  ~10^4 particles. For larger systems, poten_tree is faster. This
  algorithm calls poten_tree, with a theta of 1.5, when nobj > 10^3

Modification History

---

  July 2010: Written by Chris Beaumont