Optimum resource allocation in cooperative networks has been studied for diverse system setups and with various optimization metrics. However, there lacks a unifying framework delineating the effects of different factors on resource optimization and its resultant benefit. In this paper, we investigate the relative effects of the optimization metric (error rate vs. outage probability), modulation type (coherent vs. differential) and the relaying protocol (amplify-and-forward (AF) vs. decodeand- forward (DF)). To facilitate such a unifying study, we provide a comprehensive set of system performance and optimization results for four commonly adopted cooperative systems: coherent amplify-and-forward (CAF), coherent decode-and-forward (CDF), differential amplify-and-forward (DAF), and differential decode-and-forward (DDF). Our analyses and simulations suggest that: i) The error rate and outage probability metrics yield similar optimization results; ii) The relaying protocol affects the optimization results more than the modulation type when the number of relays (L) is small; and iii) The CAF system is uniquely different from others when L is large.