User blog comment:LittlePeng9/Random Turing machines/@comment-10429372-20150506145612/@comment-1605058-20150508130149

I'm honestly not interested enough in confirming that this machine works fine for n=3 case that I'd be willing to spend few days writing a TM simulator in C++ for that. Also, to do the simulation in reasonable time (few hours), I would need speed of few hundred million transitions per second, which I don't think is doable without any speed-up tricks.