Author: Dann Corbit
Date: 13:04:32 07/28/04
Go up one level in this thread
On July 28, 2004 at 07:34:24, Uri Blass wrote:
>The infinite loop suggest no significant improvement
>Many versions of Crafty19.15 played in the same tournament with Crafty19.14
>
>see http://www.digichess.gr/infiniteloop/home.php
>
>Crafty19.14 scored equal or better relative to many versions of 19.15
>
>If the probability of Every 19.15 to score more points than 19.14 in a
>tournament of 28 games is 60%(I expect it from a significant improvement) then
>the probability of all versions not to score better than 19.14 is 0.4^4<0.05
>
>
>Tried to promote to the infinite loop and failed
>Crafty 19.15 SE Ramat-Gan [Polgar] 17.5(equal to 19.14 DC)
>Crafty 19.15 PS 16.5
>Crafty 19.15 BH 15.5
>Crafty 19.15 DC 9.5
>
There is just enough data at this stage to confidently state:
"We have no idea which one is strongest."
Note the size of the error bars:
Program Elo + - Games Score Av.Op. Draws
1 Thinker 4.6b : 2721 114 111 28 67.9 % 2592 35.7 %
2 Crafty 19.14 DC : 2683 121 99 28 62.5 % 2594 39.3 %
3 Crafty 19.15 SERG [Polgar]: 2683 121 135 28 62.5 % 2594 17.9 %
4 Thinker 4.6c : 2683 121 111 28 62.5 % 2594 32.1 %
5 Crafty 19.15 PS : 2659 126 107 28 58.9 % 2596 32.1 %
6 Gothmog 1.0 beta 7 : 2647 128 110 28 57.1 % 2597 28.6 %
7 Crafty 19.15 BH : 2635 131 75 28 55.4 % 2598 53.6 %
8 Gothmog 1.0 beta 10 : 2612 138 100 28 51.8 % 2599 32.1 %
9 Comet B.68 : 2600 117 117 28 50.0 % 2600 28.6 %
10 Green Light Chess 3.00.3.4: 2577 115 134 28 46.4 % 2602 21.4 %
11 Green Light Chess 2.18 : 2517 111 121 28 37.5 % 2606 32.1 %
12 King Of Kings 2.56 : 2505 119 118 28 35.7 % 2607 28.6 %
13 King of Kings 2.52 : 2492 102 116 28 33.9 % 2608 39.3 %
14 Crafty 19.15 DC : 2492 102 116 28 33.9 % 2608 39.3 %
15 Comet B.48 : 2492 102 116 28 33.9 % 2608 39.3 %
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.