Computer Chess Club Archives


Search

Terms

Messages

Subject: Yace PB - vs - Aristarch 4.50 (Stats)

Author: pavel

Date: 20:53:32 04/20/05


First 1000 Games:
--------------------------------------------------------------------------------
      Program                     Elo    +   -   Games   Score   Av.Op.  Draws

  1 Aristarch 4.50              : 2519   19  19  1000    55.4 %   2481   23.3 %
  2 Yace Paderborn              : 2481   19  19  1000    44.6 %   2519   23.3 %


Individual statistics:

1 Aristarch 4.50            : 2519  1000 (+437,=233,-330), 55.4 %

Yace Paderborn                : 1000 (+437,=233,-330), 55.4 %

2 Yace Paderborn            : 2481  1000 (+330,=233,-437), 44.6 %

Aristarch 4.50                : 1000 (+330,=233,-437), 44.6 %
--------------------------------------------------------------------------------




Second 1000 Games:
--------------------------------------------------------------------------------
       Program                   Elo    +   -   Games   Score   Av.Op.  Draws

  1 Aristarch 4.50             : 2524   19  19  1000    56.8 %   2476   24.8 %
  2 Yace Paderborn             : 2476   19  19  1000    43.2 %   2524   24.8 %


Individual statistics:

1 Aristarch 4.50            : 2524  1000 (+444,=248,-308), 56.8 %

Yace Paderborn                : 1000 (+444,=248,-308), 56.8 %

2 Yace Paderborn            : 2476  1000 (+308,=248,-444), 43.2 %

Aristarch 4.50                : 1000 (+308,=248,-444), 43.2 %
--------------------------------------------------------------------------------




Third 1000 games:
--------------------------------------------------------------------------------
     Program                     Elo    +   -   Games   Score   Av.Op.  Draws

  1 Aristarch 4.50             : 2509   19  19  1000    52.5 %   2491   20.9 %
  2 Yace Paderborn             : 2491   19  19  1000    47.4 %   2509   20.9 %


Individual statistics:

1 Aristarch 4.50            : 2509  1000 (+421,=209,-370), 52.5 %

Yace Paderborn                : 1000 (+421,=209,-370), 52.5 %

2 Yace Paderborn            : 2491  1000 (+370,=209,-421), 47.5 %

Aristarch 4.50                : 1000 (+370,=209,-421), 47.4 %
--------------------------------------------------------------------------------



Overall (Total 3158 games played so far):
--------------------------------------------------------------------------------
     Program                     Elo    +   -   Games   Score   Av.Op.  Draws

  1 Aristarch 4.50             : 2517   11  11  3158    54.9 %   2483   22.9 %
  2 Yace Paderborn             : 2483   11  11  3158    45.1 %   2517   22.9 %


Individual statistics:

1 Aristarch 4.50            : 2517  3158 (+1372,=723,-1063), 54.9 %

Yace Paderborn                : 3158 (+1372,=723,-1063), 54.9 %

2 Yace Paderborn            : 2483  3158 (+1063,=723,-1372), 45.1 %

Aristarch 4.50                : 3158 (+1063,=723,-1372), 45.1 %
--------------------------------------------------------------------------------

Number of games played per opening:
http://www.pavs.org/eco3.txt


Comments:
It seems that the positional learning effect started kicking in after 2000
games. Or it could be just random results?
IMO the next 2 thousands games will interesting.

Some info for people who have missed:-

Game setup:

Chessbase GUI
Own Book
Learn on (Yace Book and Positional Learn, Aristarch only supports book learn)
TB ON
Time Control: 2min/game
Ponder off
Hash: 50+

Hardware: PIV 2.66/760mb Ram


Objective:
Primary - To observe the effects of positional learning for YACE.
Secondary - Trying to minimize the error margin in games score.

Help needed to clean the PGN database. Chessbase format put annotations in PGN.
Attempt to clean it with PGNextract failed.
Also help needed to form some kind of Graph to show "progress" after games for
YACE.

PGN available only by e-mail.

Nutty Experiment Regards,
pavs



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.