We study the capacity of cellular systems with interference-adaptation dynamic channel allocation (DCA) through a set of heuristics that evaluate the required number of channels for some mobile traffic pattern. In particular, we evaluate the improvement in the reuse factor given the knowledge of the mobiles' locations. Assuming that the mobiles' locations are sampled from the uniform random distribution or are fixed on a uniform grid, we show the effect of a number of parameters, such as the number of mobiles per cell, the minimum allowable signal-to-interference ratio, and the limited knowledge of mobiles' locations. We also investigate the effect of shadow fading and signal-based power control. Although previous papers have proposed various heuristics with varying performance, we present heuristics that are shown to give the maximum packing results based on our assumptions. In particular, the single-interferer assumption, used throughout our work, is justified and the optimality of the square-root signal-based power control is proven.