Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Intelligent software, please (this example is only tactics if you

Author: Uri Blass

Date: 00:54:10 11/27/01

Go up one level in this thread


On November 27, 2001 at 03:40:41, Johan Melin wrote:

>On November 26, 2001 at 06:09:57, Uri Blass wrote:
>
>>On November 25, 2001 at 15:14:16, Victor Fernandez wrote:
>>
>>>On November 25, 2001 at 13:37:09, Mike S. wrote:
>>>
>>>>The number of types of positions and ideas chess programs don't handle well, is
>>>>shrinking rapidly IMO. Do you have examples?
>>>
>>>Hi Mike !
>>>
>>>Yes, an example: an ending with few pieces .
>>>
>>>Kotov-Botvinnik
>>>
>>>8/8/4b1p1/2Bp3p/5P1P/1pK1Pk2/8/8 b - - 0 1
>>>
>>>The correct move is ...g5 !!! Few programs play this.
>>>
>>>Regards,
>>>
>>>  Victor
>>
>>search the right lines)
>>
>>only material yace version that I got from the programmer could prove that black
>>can win a bishop for a pawn.
>>
>>It did not use the KB KPP tablabases but it used the KB vs KP tablebases
>>so I guess that the evaluation is not because of KB vs KP endgame
>>It used the following values(bishop=3.50,pawn=0.80)
>>
>>New position
>>[D]8/8/4b1p1/2Bp3p/5P1P/1pK1Pk2/8/8 b - - 0 1
>>
>>Analysis by Yace 0.99.56:
>>
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 1   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 2   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 3   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 4   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 5   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 6   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 7   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 8   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 9   00:00:00
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 14   00:00:00  5kN
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 14   00:00:00  5kN
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 14   00:00:00  5kN
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 14   00:00:00  5kN
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 14   00:00:00  5kN
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 15   00:00:02  1571kN
>>1...g5 2.fxg5 d4+ 3.exd4 Kg3 4.g6 Kxh4
>>  -+  (-2.70)   Depth: 16   00:00:06  4318kN
>>
>>(Blass, Tel-aviv 26.11.2001)
>>
>>Uri
>
>Did Yace do any position learning before this test? Solving it at depth 1 is a
>bit to good ...
>
>/Johan Melin

Yes
It was also the only material version because I wanted to know if it is possible
to prove that g5 wins material.

I also posted analysis of the original yace with no learning in another post.

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.