Author: Antonio Dieguez
Date: 11:31:26 05/04/00
Go up one level in this thread
On May 04, 2000 at 14:26:08, Antonio Dieguez wrote: >On May 04, 2000 at 13:25:04, Steve Coladonato wrote: > >>On May 04, 2000 at 12:40:27, Dan Andersson wrote: >> >>>No you are mistaken. You are not looking at a shallow ply depth at all. It is >>>only that by knowing the value of the best move you can discard really bad >>>variations fast, i.e. you only need to prove that the next move cannot be better >>>for you, not its exact value. That was the main point of the 'hole' analogy. >> >>Dan, >> >>But what is it you are looking at to determine the value of the move if it is >>not an evaluation at various ply depths? It seems to be a static evaluation of >>the board similar to an arbiters decision just looking at the board. >> >>Also, given moves A, B, C with B the actual best move, if I start with A and >>spend some time with it, at the end of that time, A is the best move because so >>far it is the only one I've looked at. Now I go to B, and using whatever it is >>that is being used to eval these moves, I see right away that B is better. Do I >>now spend some additional time looking at B or do I just move right on to move >>C? >> >>Steve > >hi > >The alpha beta is suppose that every side will play the best, usually in a given >position a player moves and suppose the move A1 (the first choice) gets 20 >points for that player. >Ok, later you see what happens moving A2 (his second choice) and you suddenly >see that if the other player plays B1 then he eat your horse or got a very cool >position for him or something and he get for example 300 points, so -300 points >for the first player. >Well you will not play A2 because you know it is worse than A1 , and obviously >you don't know its exact value because B2 may be win a very pretty queen for the >second player but that dont care. >Thats on the whole tree, for both sides. > >am sorry 'cause my english. I dont know if it is clear or not (i hope). Just an observation, the "alpha beta" cut the choices, not the profundity.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.