Hmm, AFAIK, I've been one of the most vocal around here about the polar patterns of the models with the smaller horn/higher XO point and where these have audible shortcomings compared to the early larger horn/lower XO point models, the PE project OTOH apparently does an excellent job of balancing the design trade-offs/minimizing the transition anomalies from omni to beaming of a multiple point source system, so to my way of thinking you're comparing 'apples n' oranges'.

This doesn't make the HOSS's collapsing directivity 'good' per se though, just that when choosing to use multiple point source drivers to cover a wide BW it's automatically considered an acceptable trade-off, so the performance criteria is reduced to how smooth (transparent) it is. Most of the 604 models OTOH don't do so well, so in this regard I imagine the HOSS is superior, though without polar and impedance plots combined with BW limited listening tests there's no way to know for sure.

GM