We have performed ab initio neutrino radiation hydrodynamics simulations in three and two spatial dimensions (3D and 2D) of core-collapse supernovae from the same 15 M progenitor through 440 ms after core bounce. Both 3D and 2D models achieve explosions, however, the onset of explosion (shock revival) is delayed by ∼100 ms in 3D relative to the 2D counterpart and the growth of the diagnostic explosion energy is slower. This is consistent with previously reported 3D simulations utilizing iron-core progenitors with dense mantles. In the ∼100 ms before the onset of explosion, diagnostics of neutrino heating and turbulent kinetic energy favor earlier explosion in 2D. During the delay, the angular scale of convective plumes reaching the shock surface grows and explosion in 3D is ultimately lead by a single, large-angle plume, giving the expanding shock a directional orientation not dissimilar from those imposed by axial symmetry in 2D simulations. We posit that shock revival and explosion in the 3D simulation may be delayed until sufficiently large plumes form, whereas such plumes form more rapidly in 2D, permitting earlier explosions.

We present four ab initio axisymmetric core-collapse supernova simulations initiated from 12, 15, 20, and 25 M  zero-age main sequence progenitors. All of the simulations yield explosions and havebeen evolved for at least 1.2 s after core bounce and 1 s after material first becomes unbound. These simulations were computed with our CHIMERA code employing RbR spectral neutrino transport, special and general relativistic transport effects, and state-of-the-art neutrino interactions. Continuing the evolution beyond 1 s after core bounce allows the explosions to develop more fully and the processes involved in powering the explosions to become more clearly evident. We compute explosion energy estimates, including the negative gravitational binding energy of the stellar envelope outside the expanding shock, of 0.34, 0.88, 0.38, and 0.70 Bethe (B≡10 51 erg) and increasing at 0.03, 0.15, 0.19, and 0.52 B s 1 -, respectively, for the 12, 15, 20, and 25 M  models at the endpoint of this report. We examine the growth of the explosion energy in our models through detailed analyses of the energy sources and flows. We discuss how the explosion energies may be subject to stochastic variations as exemplfied by the effect of the explosion geometry of the 20 M  model in reducing its explosion energy. We compute the proto-neutron star masses and kick velocities. We compare our results for the explosion energies and ejected Ni 56 masses against some observational standards despite the large error bars in both models and observations.