The problems of determining the optimal power allocation, within maximum power bounds, to (i) maximize the minimum Shannon capacity, and (ii) minimize the weighted latency are considered. In the first case, the global optima can be achieved in polynomial time by solving a sequence of linear programs (LP). In the second case, the original non-convex problem is replaced by a convex surrogate (a geometric program), using a functional approximation. Since the approximation error is relatively low, the optima of the surrogate is close to the global optimal point of the original problem. In either cases, there is no assumption on the SINR range. The use of LPs and geometric programming make the proposed algorithms numerically efficient. Computations are provided for corroboration.