State preparation is needed to perform a variety of quantum algorithms. In this thesis, the focus lies on a version that aims to program amplitudes and phases of a superposition consisting of a polynomial number of states. The quantum random access memory (QRAM) a pioneering ansatz, which solves this problem needs an exponential amount of resources. A less general, but not as hardware-intense protocol has been proposed. In this thesis phase development during the adiabatic-diabatic segments of the protocol is studied and the ansatz is extended to include phase programming using only local fields. The necessary field strengths can be calculated analytically and the phases can be controlled very precisely and in more situations compared to the amplitudes. The extended protocol is then benchmarked, where the overlap between the target state and the prepared state is used as a measure of performance. It was shown that the protocol struggles to prepare superposition states, which have eigenstates that contribute to the superposition with a low amplitude, but shows better performance if every eigenstate that contributes to the superposition contributes with similar amplitudes.