This paper discusses the status, advances, open problems, and perspectives in the area of large-scale microscopic nuclear mass calculations. This field of research is past phenomenological approaches that gave us a very good understanding of general features and trends, but lacked fundamental derivations and had limited predictive power. At present, the focus is on microscopic descriptions of nuclei whereupon they are treated as finite quantum objects built of (quasi)nucleons. Nuclear ground states and masses are in this approach determined by basic fields, which are the particle and spin densities along with their derivatives and gradients up to the second order in relative momenta. These fields interact in such a way that the total energy of a given system is a functional of densities, defined and understood in the general framework of the Kohn-Sham theory. The determination of such a universal functional, along with all the dynamic corrections required by data, is the main purpose of current investigations. In this endeavor, we strive not only to have a reliable theoretical tool to calculate properties of very exotic systems that will not be soon accessible in experiment, but also wish to have a spectroscopic quality description of well-known systems, in which very precise data do exist now, and can be used as a rich source for determination of theoretical parameters. Such a program of research should npt only be rooted in the fundamentals of low-energy QCD methods and ideas, but also, by definition, must rely on experiment for elements that cannot be derived from first principles. It is a vast and ambitious program presently under way.