Author: Ricardo Gibert
Date: 22:06:03 01/06/02
Go up one level in this thread
On January 06, 2002 at 16:04:09, Uri Blass wrote: >On January 06, 2002 at 15:01:35, Severi Salminen wrote: > >> >>>Well...to begin with, "a movegenerator that allways generated the best move" is >>>clearly an unrealizable pipe dream. With such a generator, there is no reason to >>>search. You would generate the best move and then play it straight away. >> >>Yep, that's what I meant. >> >>[snip...] >> >>>The moral of the story is: The move you want to *play* should be the *best*, but >>>the moves you want to *analyze* are the ones that cost you the least amount of >>>work overall most of the time. >>> >>>Don't confuse the properties you desire in the moves you *play* with the >>>properties you desire in the moves you *analyze*. They are not the same. >> >>The best move (as seen from engine's point of view) allways results the least >>amount of work. Assuming that I also search the best reply to it and so on. > >No > >suppose that beta<0 so your engine is happy with a draw. >move A is the best and give material advantage but you need to search a lot of >nodes to get a cutoff. > >suppose that the line that your engine searched end with the moves >Rc1-b1 Rd8-e8 Rb1-c1 > >if you search the move Re8xe1 first you get a cutoff because Re8xe1 is winning >the game but you need to search a lot of nodes to get the cutoff. > >If you search Re8-d8 first you get immediate reptition(Rc1-b1 Re8-d8 Rb1-c1 >Rd8-e8 lead to repetition of the position) so you get the cut off by only one >node. > >Uri ETC (Enhanced Transposition Cutoffs) is an example of technique that aims to minimize work performed on average and takes no interest in finding the best move.
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.