Four results on out-of-equilibrium 2PI simulations in 3+1 dimensions

We perform an analysis of a number of approximations and methods used in numerical simulations of real-time Kadanoff-Baym equations based on truncations of the 2PI effective action. We compare the loop expansion to the 1/N expansion and compare their classical limit to classical-statistical simulations. We also compare implementations based on a space-time lattice discretization at the level of the action to an ad hoc momentum discretization at the level of the equations of motions. We extract some rules of thumb for performing 2PI-simulations of out-of-equilibrium systems.