User blog comment:Wythagoras/Graphs/@comment-1605058-20140911185546/@comment-10429372-20140912183350

Sorry for being so unclear, but I ran out of time while writing comment.

LittlePeng, I got the strength of PSSCG indeed form Friedman.

Deedlit, I meant that it uses the same principle. For example look to how Hyp cos goes form theta(Omega^omega) to theta(Omega^omega,1). Replacing the theta(Omega^omega+1)-graph with K_3,3 with one vertex to connect, we can get in a similiar way form theta(Omega_omega) to theta(Omega_omega,1). Also, a triangle with two K_3,3s connected would have level epsilon(theta(Omega_omega)+1), beacause double edge is not allowed. I forgot.