The material point method (MPM) represents an alternative discretization method for numerical simulations. It aims to combine the benefits of a Lagrangian representation of bodies and an Eulerian numerical solution approach. Therefore, especially at high material deformations the method is not prone to mesh distortions such as the finite element method (FEM). For this reason, the MPM is used to a great extent for modeling granular materials as in geo-mechanics. However, high deformations occur in many industrial processes on metallic materials. The Split-Hopkinson-Pressure-Bar (SHPB) experiment is used to characterize material properties at high deformation rates. Although widely used, this experiment is not yet standardized and shows a variety of sensitivities, e.g. to friction. Inter alia for this reason, simulations are conducted with the experiment to allow for a better evaluation of the measured data. The purpose of this work from an engineering point of view is to analyze the performance of the MPM on an SHPB experiment. In order to validate the experimental results for the material characterization under dynamic loading conditions we introduce frictional contact. We use arbitrary tri-linear brick domains in a 3D CPDI1 scheme, instead of originally used parallelepipeds. This allows for a more flexible geometry approximation using standard meshes. The results of the method are analyzed with respect to discretization sensitivity and discussed in the context of the experimental results for a 42CrMo4 steel. We were able to show that the method is capable to reproduce the SHPB experiment. Additionally the method shows convergency in the results with finer discretizations. Thus, the MPM has underlined its importance as an alternative simulation technique for problems with high deformation.