Comic Discussion > QUESTIONABLE CONTENT
WCDT: 2281-85 (24-28 September 2012) Weekly Comics Discussion Thread
Barmymoo:
Freudian or Oedipal? Or both?
jwhouk:
Yes.
Skewbrow:
--- Quote from: Akima on 01 Oct 2012, 00:49 ---Chess is so Freudian. The Queen (the mother) is the most powerful piece on the board, and the object of the game is to kill the King (the father). Fathers typically teach their sons to play, and every son's ambition is to defeat his father (the King). :-D
--- End quote ---
Possibly :psyduck: But chess players (designers of chess problems in particular) have a penchant for puzzles, where the solution is to sacrifice your own mother in order to kill the opposing father. Doesn't fit too well, I think?
And are you saying that the young go/weiqi players do not aspire to beat their father/elder brother/sensei/whatever at the game? With or without handicap stones? Also, following Sun Tzu's advice "To a surrounded enemy, you must leave a way of escape" is a non-starter in go.
jmucchiello:
--- Quote from: Near Lurker on 30 Sep 2012, 14:38 ---Whoever said that to you is wrong. We don't know how to make a computer play go (I am not apologizing for calling it the name just about every English speaker knows it under) well only because we don't yet know how humans do. The methods we use to teach a computer chess and backgammon are clearly inadequate, but that's just a matter of applying the wrong tactic. Consider Kasparov's insistence that Deep Blue was cheating when it saw through a tactic that he though required "creative thinking" - he was making the same mistake as your friend, thinking that a computer cannot duplicate that which we can't immediately see its method of duplicating. No, the fact that humans can be trained to play go well is sufficient evidence that a computer can be taught to.
--- End quote ---
We don't teach computers to play Chess. We teach computers to evaluate all possible future Chess moves from a given board state and ask it to pick the one that leads to the best outcome. 90% of Chess programming is a breadth first search through a tree of board states. The only trick to Deep Blue was IBM made actual computer cores where "the state of a chess board" was a native "register" to facilitate faster and deeper decision trees. The way they "cheated" was in tuning the evaluation algorithm based on Kasparov's style of play. Kasparov's complaint was that he could not in turn study Deep Blue's style of play.
Is it cold in here?:
--- Quote from: jmucchiello on 01 Oct 2012, 09:22 ---all possible future Chess moves from a given board state
--- End quote ---
I'm pretty sure that's not what you meant?
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version