Consider, from someone who knows speedcubing, that it’s not 3,000+ random algorithms, more like a decision tree of algorithms. Many are similar and related; man

Consider, from someone who knows speedcubing, that it’s not 3,000+ random algorithms, more like a decision tree of algorithms. Many are similar and related; many are rotated or mirrored. Long practice and pattern recognition make most algorithm selection second-nature. Yes, there’s a ton of memorization involved, but the limited problem domain, and therefore easy mental modeling of it, make this exactly the kind of task at which humans excel. Professional chess and go players can replay many, sometimes all, of their past games from memory. Surely that’s a greater feat.