The growth of a crystal is usually determined by its surface. Many factors influence the growth dynamics. Energy barriers associated with the presence of steps most often decide the emerging pattern. The height and type of Ehrlich−Schwoebel step barriers lead to the growth of nanocolumns, nanowires (NWs), pyramids, and bunches or meanders in the same system. Surface diffusion is another factor that determines the nature of growth. We used the (2 + 1)D cellular automaton model to investigate the additional effect of diffusion along with step barriers. We show that when we change only the diffusion rate, the length of the meanders or the height of the bunches increases, while the cracked structure of the nanopillars changes into very long, tall NWs. We show that the length of the step−step correlation is a good characterization of the resulting patterns.