The performance of four recent semi-empirical interatomic potentials for iron, developed or used within the FP6 Perfect Project, is evaluated by comparing them between themselves and with available experimental or, more often, density functional theory data. The quantities chosen for the comparison are of specific interest for radiation damage studies, i.e. they concern mainly properties of point-defects and their clusters, as well as dislocations. For completeness, an earlier, widely used (also within the Project) iron potential is included in the comparison exercise as well. This exercise allows conclusions to be drawn about the reliability of the available potentials, while providing a snapshot of the state-of-the-art concerning fundamental properties of iron, thereby being also useful as a kind of handbook and as a framework for the validation of future semi-empirical interatomic potentials for iron. It is found that Mendelev-type potentials are currently the best choice in order to "extend density functional theory" to larger scales and this justifies their widespread use, also for the development of iron alloy potentials. However, a fully reliable description of self-interstitial atom clusters and dislocations with interatomic potentials remains largely an elusive objective, that calls for further effort within the concerned scientific community.