Emphasis on your last sentence. There really isn't much room to advance commercially after Rybka. It is scary how good its evaluation function is compared to ot

Emphasis on your last sentence. There really isn't much room to advance commercially after Rybka. It is scary how good its evaluation function is compared to other engines. Even if some other engine manages to beat its strength, it will not mean much practically to a paying human(even a GM). That said, what interesting developments are you seeing in chess engines? I know that Rybka is aggressively pruning its search tree (its pos/s is an order of magnitude less than say Fritz), but question is how?