I think I was too focused on the specifics of the analogy to realize it was a general statement. Thank you for clearing that up for me. I did not mean to dele

I think I was too focused on the specifics of the analogy to realize it was a general statement. Thank you for clearing that up for me. I did not mean to delete the comment you replied to. I have included it below: It is terribly confusing analogy for me. Initially I thought the computer engine analogy was what was intended. But then I started thinking about someone rated 2100, down two bishops playing stockfish on a decent box with 3-4-5 EGTBs and a two bishop advantage. That seemed like a no brainer so I moved on to crafty, and then GNU chess and then what? I realized that I could not think of an engine that would consider a win with two bishops an accomplishment. As far as "maybe 2600 is a better target" goes you have to go all the way down to #88 on the CCRL in order to get to 2600.[^1] That is with even material and definitely not using 64 cores (the CCRL uses a AMD X2 4600+ at 2.4GHz benchmark). I am sorry but thinking about a chess engine running on 64 cores with a two bishop advantage against a 2100 rated human is not reasonable. If I knew nothing about computers and Go I could reasonably read the 2-bishops/2100-ELO analogy and conclude that CrazyStone was laughable. I do not know much about computers and Go but I do know enough that I do not consider CrazyStone's performance laughable. [^1]: http://www.computerchess.org.uk/ccrl/4040/ http://www.computerchess.org.uk/ccrl/4040/