Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Ruffian!! simply the best

Author: Uri Blass

Date: 07:13:44 12/28/02

Go up one level in this thread


On December 28, 2002 at 09:45:02, Drexel,Michael wrote:

>[D] 8/8/pp6/2pp2p1/3p2Pk/3P1K1P/5PP1/8 b - - 0 1
>
>Ruffian is the best analysis-engine I can think of.
>Ruffian 1.0.1 (96 MB hash)on Athlon@1.6Ghz finds the mate in 13
>in 5 sec (Yes 0:05!!) . This is amazing.

I believe that with the right extensions it should be solved in less than 5
seconds.

This mate is so simple that even humans can see it so it is a shame that chess
engines cannot see it faster.

The plan is simple

Ke2 Kf1 Kg1 Kh2 g3#

It is clear that the only defence of whote is by d1Q and Qxg4 so black needs to
sacrifice the d pawn to prevent mate.

Movei has not the right extensions today but I plan to do it better in tactics
in the future.

I believe that it will be able to see the mate faster than ruffian when I will
have it but it is not going to happen in the near future.

>hey you out there with your 2.8 Ghz Xeons and Deep Fritz 7 engines.
>Try to beat that!
>hope Per-ola will bring out a native Chessbase version of this great program
>soon because there exists some problems with UCI version in Chessbase GUI.
>I would pay 25$ for a native CB engine.
>I dont need chess engines with much positional knowledge for analysis. every
>chessplayer >2200 has much more positional knowledge than all chess-engines
>together.

I believe that you are wrong here.
chess engines may know specific positional knowledge that part of the >2200
players do not know.

I do not think that every >2200 player know everything that is written in books
about endgames.
I believe that at least one of the chess engines may know
things that not every >2200 player knows.

I also believe that the weight of the evaluation of some chess engines may be
better than part of the players with rating that is bigger than 2200.

Humans may do mistakes because they overevaluate some factor in their evaluation
and later they may find that at least one of their program had more realistic
evaluation.

Uri



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.