Author: James Robertson
Date: 08:11:21 07/14/99
Go up one level in this thread
On July 14, 1999 at 07:26:26, Frank Quisinsky wrote: >On July 14, 1999 at 07:19:26, Frank Quisinsky wrote: > >>On July 14, 1999 at 06:41:42, Steffen Jakob wrote: >> >>>Hi Frank! >>> >>>On July 14, 1999 at 06:20:07, Frank Quisinsky wrote: >>> >>>>Hello Will, >>>> >>>>statistic for longer games. >>>> >>>> 18 x 100 : 3404 = 0.53% (over 200 moves) >>>> 70 x 100 : 3404 = 2.06% (over 150 moves) >>>>386 x 100 : 3404 = 11.34% (over 100 moves) oh 11,34% !!! >>>> >>>>(3404 games, 40 moves in 40 minutes, WinBoard 4.0.0 and 4.0.2, 32 MB >>>>Hash-Tables, Ponder = Off, AMD K6-2 333-416 MHz, AMD K6-3 450 MHz, Celeron 450 >>>>MHz and Pentium III 504 MHz). >>>> >>>>So longest games ... >>>> >>>>Bringer - Stobor 253 moves OK, Nr. 1 :-))) >>>>Bionic - Patzer 245 moves >>>>Phalanx - Patzer 242 moves >>> >>>It would be interesting to see the same statistics only for games which didn't >>>end with a draw. >>> >>>Greetings, >>>Steffen. >> >> >>Hello Steffen, >> >>OK, the new WB database = 3172 games >>(I delate old games with ZChess 0.92-1.0 and Gromit 2.0-2.1, 350 games) >> >>In this database I have 729 remis games = 22,98 %. >> >> 0 - 25 moves = 17 games 2,33 % >> 26 - 50 moves = 101 games 13,85 % >> 51 - 75 moves = 223 games 30,59 % >> 76 - 100 moves = 160 games 21,94 % >>101 - 125 moves = 112 games 15,36 % >>126 - 150 moves = 61 games 8,37 % >>151 - ... moves = 55 games 7,54 % >> >>In 729 games = ~ 83 moves ! >> >>I think that the table-bases very importent for the programs, when I see my >>little statistic. >> >>And ... >>Phalanx, AnMon and Comet have a very good statistic in win under 30 moves. >>Over 400 games with Comet and Phalanx and over 250 games with AnMon ! >> >>Kind regards >>Frank > >Oh, I make a big mistake :-(( >I mean not 400 games under 30 moves, I mean 400 games with Comet and Phalanx and >250 games with AnMon are in the database ! > >Sorry, my english ... > >Kind regards >Frank When people used to send me games in CBV format, I would search for games lost under 30 moves, and almost every one would be my program. Now, however, the opponents still win, but they usually have to wait 40 or 50 moves! :) James
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.