(WPMaxSAT). The objective of this paper is to analyze the instances of the 2010 MaxSAT Evaluation in order to gain new insights into their computational hardness, answer some questions that have been asked to us as organizers, and evaluate how appropriate are the current settings of parameters such as timeout and available RAM memory. To this end, we conducted a number of experiments, which were performed on a cluster with 160 2 GHz AMD Opteron 248 Processors with 1 GB of RAM memory. In the experiments, we considered the 2,675 instances of the 2010 MaxSAT Evaluation: 544 MaxSAT instances, 349 WMaxSAT instances, 1,122 PMaxSAT instances, and 660 WPMaxSAT instances. Instances were assigned to one of the following three categories: random, crafted and industrial. We used the 17 solvers that participated in MaxSAT-2010. They can be classified into three main types: branch and bound (B&B) solvers, satisfiability-based (sat-based) and unsatisfiability-based (unsat-based) solvers. In the first type, we find 10 solvers: akmaxsat, akmaxsat ls, IncMaxSatz, IncWMaxSatz, Maxsat Power, LS Power, WMaxsat Power, LSW Power, WMaxSatz-2009, and WMaxSatz+. In the second type, we find 2 solvers: SAT4J-Maxsat,and QMaxSAT. In the third type, we find 5 solvers: WPM1, PM2, WPM2, wbo 1.4a, and wbo 1.4b.