This paper reports results from our ongoing research on high-fidelity jet noise simulations. We have previously presented results from the simulations of moderate Reynolds number cold jet flows exhausting from chevron nozzles. This paper presents results from the simulation of a baseline round nozzle without chevrons. The previous chevron nozzles and the current round baseline nozzle were experimentally studied by researchers at the NASA Glenn Research Center. Results from the simulations are compared with the experimental measurements. In this study, the flow inside the nozzle and the free jet flow outside are computed simultaneously by a high-order accurate, multi-block, large eddy simulation (LES) code with overset grid capability. For the round nozzle jet test case, we solve the governing equations on approximately 370 million grid points. The main emphasis of these simulations is to compute the jet flows in as much detail as possible and accurately capture the physical processes that lead to noise generation. The calculations resolve the jet flow field in an unprecedented level of detail by utilizing extreme computational resources and thus, they are the first of their kind. Two separate simulations are performed using turbulent and laminar inflow conditions at the jet nozzle inlet. Results are compared with experimental measurements. Results clearly show that nozzle inflow conditions have an influence on the jet flow field and far field noise.