For a number of years, Sigma Energy has been developing a novel type of a wave energy converter. It is a point absorber which transforms, by original mechanical PTO system, the wave-induced vertical motion of the circular float due to waves into electrical energy. The performance of the device has been thoroughly analyzed by original software developed by Sigma Energy as well as by several model tests. All this resulted in a good proof of concept and validation of the technology used. The last model tests (supported by the MARINET initiative) were performed with two main objectives: to verify the numerical prediction tools and to check the new electronic system for the generator control. Several other innovations applied on the device were carefully observed and analyzed as well. The tests, in general, showed a very favorable operating behavior of the device. They also clearly presented potentials and limitations of both the numerical and the experimental approach. As one of the important outcomes, they indicated some additional resistance of an idling generator that was not predicted by the numerical modeling. If the additional electrical damping is measured and incorporated into the numerical codes, a very good correlation between the numerical prediction and experimental results was found. But, in spite of the general agreement, some discrepancies were found. Some dilemmas which could not be solved by a numerical prediction were clarified, but some (unexpected) problems did occur as well. In the belief that the lessons learned could have a more general interest, the paper presents the main results of the experiments, their analysis and a comparison to numerical prediction without avoiding some of the remaining dilemmas.