Author: A. Cozzie
Date: 07:21:03 08/28/05
Go up one level in this thread
On August 28, 2005 at 09:23:15, Peter Berger wrote: >Thanks for your very nice tournament report. It's nice to hear that you go >commercial although it will be hard to get used to this gay teddy ;) . > >Joking aside. > >--- > >Concerning "sceptical programs" and my alleged quote ( that I don't remember >just like that ;) ) . I don't know how useful that term is ( and I am sceptical >about it :) ) , but there is something serious about it. The major contender in >this group is certainly Fruit right now. Professionals have become very >speculative about material in recent years, and I think they might have overdone >it at times. Zappa is not very sceptical either. Ah, if I didn't get the quote right you can post the correct one and I'll fix it. I think the commercials are simply tuned to beat amatuers. To win an 11 round swiss tournament usually takes 8.5/11 or more, which means scoring _really well_ against the bottom half. And winning with Black especially is getting really hard. You have to be speculative to mix it up a bit and give them a position where they can make mistakes. Shredder, Junior, and Diep are all _very_ aggressive. Fruit is actually pretty aggressive, but not on kingsafety. Crafty I feel is usually a bit too passive. Zappa tries to walk the fine line of 'correctness' between Junior and Crafty/Fruit. Sometimes it works, sometimes it doesn't. >I am currently playing a big series of games between Fruit and Junior for fun >who play at nearly the exact same level, but reach their aims in very different >ways. Speculative vs sceptical is a term that fits quite well IMHO. Very >fascinating games btw, I'll probably post some later. >-- > >The "Strategy of Attack" ( trademark Futé and Jean-Louis Boussin) . Yes, this >one was a bit funny, but maybe there should be at least a little explanation on >what it is about. I'll try. > >From what I understood Futé does some complicated static evaluation at the root >position (pre-processing) to guide the search and trigger some search extensions >(his strategies). So far there were only two implemented in the program, the >"strategy of attack" and the "strategy of defense" ( that was not really working >well yet according to the author). > >Usually Futé's search lacked some depth so that it was outsearched quite often >in games. Now when the "Strategy of Attack" kicked in ( as far as I understood >triggered by the number of pieces directed against the opponent's king mainly) , >Futé managed to search some lines a lot deeper. This is what happened in the >Zappa game for all I know. >The vagueness is not due to lack of explanations by Futé's author but more by my >ignorance. *No one* could complain about Jean-Louis Boussin not talking and >explaining enough ;) ) . > >In practice that led to the Futé's author looking at his screen and always >hoping that at some point the "strategy of attack" would kick in . E.g. it >happened once in the game Crafty-Futé - unfortunately in a position where there >was already a mate in 3 against Futé . > >Peter
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.