YABRL: SmarThink 0.17a third behind the Ruffians

[Robert Allgeuer]

Geschrieben von: / Posted by: Robert Allgeuer at 29 January 2004 22:01:57:

although very, very close.
After 659 unique games SmarThink 0.17a has scored exactly 2 points more than Ktulu 4.2 and 10 points more than Crafty 19.6, which is nothing. They are all very close.
Interesting that SmarThink 0.17a is so far the only engine that has maintained the upper hand against Ruffian 2.0.0 in the direct comparison, equally interesting is its "weakness" for Leila ...
Time control is 5min + 2sec, for details on platform, tools, conditions etc. please refer to the link below. Next engine is the new Aristarch 4.37.
Robert


    Program                     Elo    +   -   Games   Score   Av.Op.  Draws
 01 Ruffian v2.0.0            : 2678   18  33   700    74.3 %   2494   23.1 %
 02 Ruffian v1.0.1            : 2650   17  27   836    71.3 %   2492   25.6 %
 03 SmarThink v0.17a          : 2590   22  27   659    63.7 %   2492   22.5 %
 04 Ktulu v4.2                : 2588   21  27   674    63.5 %   2492   22.6 %
 05 Crafty v17.14DC           : 2584   20  22   800    63.3 %   2489   31.1 %
 06 Crafty v19.06DCntb        : 2580   21  22   741    61.1 %   2501   30.0 %
 07 Aristarch v4.21           : 2578   19  22   897    61.1 %   2499   23.0 %
 08 Crafty-MPC v18.15DC       : 2562   21  22   804    59.2 %   2497   26.1 %
 09 Delfi v4.3                : 2560   21  22   820    58.8 %   2498   24.0 %
 10 Delfi v4.2                : 2556   25  25   580    58.1 %   2499   27.2 %
 11 SmarThink v0.16b++        : 2554   21  21   836    58.0 %   2498   24.3 %
 12 Little Goliath 2000 v3.9  : 2553   20  20   900    57.6 %   2500   25.7 %
 13 Crafty v18.15DC           : 2551   22  22   741    59.0 %   2488   29.0 %
 14 Pepito v1.59 profile      : 2548   21  20   900    56.9 %   2500   25.7 %
 15 Yace Paderborn            : 2547   21  20   900    56.8 %   2500   25.3 %
 16 SoS 3                     : 2545   21  21   899    56.4 %   2500   21.8 %
 17 SoS 4                     : 2543   24  24   659    57.1 %   2494   22.2 %
 18 Aristarch v4.4            : 2542   36  34   319    54.1 %   2513   20.4 %
 19 Green Light Chess v3.00   : 2533   21  19   900    54.7 %   2500   24.8 %
 20 Yace v0.99.56             : 2533   34  30   360    54.7 %   2500   25.6 %
 21 Little Goliath 2000 v3.5  : 2532   31  25   440    53.6 %   2506   30.9 %
 22 Amyan v1.59               : 2509   24  19   792    51.6 %   2498   25.9 %
 23 Pharaon v2.62             : 2503   23  18   899    50.4 %   2501   23.8 %
 24 Crafty v19.01DC           : 2494   24  19   815    50.4 %   2491   25.5 %
 25 LambChop v10.99           : 2490   19  23   898    48.4 %   2501   23.3 %
 26 Gromit v3.8.2             : 2487   19  23   877    48.1 %   2500   22.7 %
 27 Ktulu v3.9                : 2486   19  24   779    48.6 %   2496   26.1 %
 28 KnightDreamer v3.2        : 2482   20  24   800    47.7 %   2498   25.1 %
 29 SlowChess v2.89b          : 2479   20  24   779    47.0 %   2500   24.4 %
 30 Anmon v5.22               : 2477   19  23   839    46.7 %   2500   26.1 %
 31 Comet B44-2               : 2475   19  23   820    46.5 %   2500   27.3 %
 32 Amy v0.8.3                : 2474   20  22   891    46.1 %   2501   18.9 %
 33 SoS v11-99                : 2472   33  34   359    46.0 %   2500   17.3 %
 34 Tao v5.4                  : 2469   20  21   899    45.3 %   2502   19.7 %
 35 Dragon v4.4.3             : 2457   21  23   766    44.3 %   2497   25.8 %
 36 Comet B62-3               : 2453   21  22   800    43.4 %   2499   26.0 %
 37 PostModernist v1.007      : 2437   22  21   820    41.0 %   2501   25.4 %
 38 Francesca M.0.0.9         : 2432   21  20   899    40.0 %   2502   25.4 %
 39 Comet B60                 : 2429   22  21   780    41.2 %   2491   25.6 %
 40 Leila v0.53h              : 2422   24  20   819    38.8 %   2501   21.4 %
 41 Tcb v0045                 : 2418   23  20   819    38.2 %   2501   24.9 %
 42 Resp v0.19                : 2399   25  19   800    35.9 %   2500   22.2 %
 43 Nejmet v3.07              : 2378   27  19   796    33.0 %   2501   22.0 %
 44 SlowChess v2.78           : 2373   27  19   790    33.5 %   2492   19.6 %
 45 Exchess v4.03             : 2323   31  17   799    26.3 %   2502   22.0 %
 46 Beowulf v2.2              : 2303   34  16   880    23.9 %   2504   18.0 %

Games        :  17790 (finished)
White Wins   :   7317 (41.1 %)
Black Wins   :   6160 (34.6 %)
Draws        :   4313 (24.2 %)
Unfinished   :      0
White Perf.  : 53.3 %
Black Perf.  : 46.7 %

(3) SmarThink v0.17a          : 659 (+346,=148,-165), 63.7 %
Ruffian v2.0.0                :  20 (+  7,=  7,-  6), 52.5 %
Ktulu v4.2                    :  20 (+  8,=  7,-  5), 57.5 %
Crafty v19.06DCntb            :  20 (+  8,=  8,-  4), 60.0 %
Aristarch v4.21               :  20 (+ 11,=  3,-  6), 62.5 %
Crafty-MPC v18.15DC           :  20 (+  9,=  4,-  7), 55.0 %
Delfi v4.3                   :  20 (+  7,=  6,-  7), 50.0 %
SmarThink v0.16b++            :  20 (+  6,= 10,-  4), 55.0 %
Little Goliath 2000 v3.9      :  20 (+  7,=  4,-  9), 45.0 %
Pepito v1.59 profile          :  20 (+ 10,=  4,-  6), 60.0 %
Yace Paderborn                :  20 (+  9,=  3,-  8), 52.5 %
SoS 3                         :  20 (+ 11,=  4,-  5), 65.0 %
SoS 4                         :  20 (+  7,=  5,-  8), 47.5 %
Green Light Chess v3.00       :  20 (+  9,=  2,-  9), 50.0 %
Amyan v1.59                   :  20 (+  9,=  8,-  3), 65.0 %
Pharaon v2.62                 :  20 (+  7,=  4,-  9), 45.0 %
LambChop v10.99               :  20 (+ 12,=  5,-  3), 72.5 %
Gromit v3.8.2                 :  20 (+ 15,=  2,-  3), 80.0 %
SlowChess v2.89b              :  20 (+ 14,=  2,-  4), 75.0 %
KnightDreamer v3.2            :  20 (+  9,=  4,-  7), 55.0 %
Anmon v5.22                   :  20 (+  9,=  7,-  4), 62.5 %
Amy v0.8.3                    :  20 (+ 12,=  5,-  3), 72.5 %
Comet B44-2                   :  20 (+ 10,=  4,-  6), 60.0 %
Tao v5.4                      :  20 (+ 14,=  2,-  4), 75.0 %
Dragon v4.4.3                 :  19 (+ 11,=  3,-  5), 65.8 %
Comet B62-3                   :  20 (+ 14,=  4,-  2), 80.0 %
PostModernist v1.007          :  20 (+ 12,=  5,-  3), 72.5 %
Francesca M.0.0.9             :  20 (+ 15,=  3,-  2), 82.5 %
Leila v0.53h                  :  20 (+  7,=  5,-  8), 47.5 %
Tcb v0045                     :  20 (+  8,=  7,-  5), 57.5 %
Resp v0.19                    :  20 (+ 11,=  4,-  5), 65.0 %
Nejmet v3.07                  :  20 (+ 15,=  3,-  2), 82.5 %
Exchess v4.03                 :  20 (+ 18,=  1,-  1), 92.5 %
Beowulf v2.2                  :  20 (+ 15,=  3,-  2), 82.5 %





YABRL (Yet Another Blitz Rating List)

[Heinz van Kempen]

Geschrieben von: / Posted by: Heinz van Kempen at 30 January 2004 13:51:27:
Als Antwort auf: / In reply to: YABRL: SmarThink 0.17a third behind the Ruffians geschrieben von: / posted by: Robert Allgeuer at 29 January 2004 22:01:57:

although very, very close.
After 659 unique games SmarThink 0.17a has scored exactly 2 points more than Ktulu 4.2 and 10 points more than Crafty 19.6, which is nothing. They are all very close.
Interesting that SmarThink 0.17a is so far the only engine that has maintained the upper hand against Ruffian 2.0.0 in the direct comparison, equally interesting is its "weakness" for Leila ...
Time control is 5min + 2sec, for details on platform, tools, conditions etc. please refer to the link below. Next engine is the new Aristarch 4.37.
Robert


    Program                     Elo    +   -   Games   Score   Av.Op.  Draws
01 Ruffian v2.0.0            : 2678   18  33   700    74.3 %   2494   23.1 %
02 Ruffian v1.0.1            : 2650   17  27   836    71.3 %   2492   25.6 %
03 SmarThink v0.17a          : 2590   22  27   659    63.7 %   2492   22.5 %
04 Ktulu v4.2                : 2588   21  27   674    63.5 %   2492   22.6 %
05 Crafty v17.14DC           : 2584   20  22   800    63.3 %   2489   31.1 %
06 Crafty v19.06DCntb        : 2580   21  22   741    61.1 %   2501   30.0 %
07 Aristarch v4.21           : 2578   19  22   897    61.1 %   2499   23.0 %
08 Crafty-MPC v18.15DC       : 2562   21  22   804    59.2 %   2497   26.1 %
09 Delfi v4.3                : 2560   21  22   820    58.8 %   2498   24.0 %
10 Delfi v4.2                : 2556   25  25   580    58.1 %   2499   27.2 %
11 SmarThink v0.16b++        : 2554   21  21   836    58.0 %   2498   24.3 %
12 Little Goliath 2000 v3.9  : 2553   20  20   900    57.6 %   2500   25.7 %
13 Crafty v18.15DC           : 2551   22  22   741    59.0 %   2488   29.0 %
14 Pepito v1.59 profile      : 2548   21  20   900    56.9 %   2500   25.7 %
15 Yace Paderborn            : 2547   21  20   900    56.8 %   2500   25.3 %
16 SoS 3                     : 2545   21  21   899    56.4 %   2500   21.8 %
17 SoS 4                     : 2543   24  24   659    57.1 %   2494   22.2 %
18 Aristarch v4.4            : 2542   36  34   319    54.1 %   2513   20.4 %
19 Green Light Chess v3.00   : 2533   21  19   900    54.7 %   2500   24.8 %
20 Yace v0.99.56             : 2533   34  30   360    54.7 %   2500   25.6 %
21 Little Goliath 2000 v3.5  : 2532   31  25   440    53.6 %   2506   30.9 %
22 Amyan v1.59               : 2509   24  19   792    51.6 %   2498   25.9 %
23 Pharaon v2.62             : 2503   23  18   899    50.4 %   2501   23.8 %
24 Crafty v19.01DC           : 2494   24  19   815    50.4 %   2491   25.5 %
25 LambChop v10.99           : 2490   19  23   898    48.4 %   2501   23.3 %
26 Gromit v3.8.2             : 2487   19  23   877    48.1 %   2500   22.7 %
27 Ktulu v3.9                : 2486   19  24   779    48.6 %   2496   26.1 %
28 KnightDreamer v3.2        : 2482   20  24   800    47.7 %   2498   25.1 %
29 SlowChess v2.89b          : 2479   20  24   779    47.0 %   2500   24.4 %
30 Anmon v5.22               : 2477   19  23   839    46.7 %   2500   26.1 %
31 Comet B44-2               : 2475   19  23   820    46.5 %   2500   27.3 %
32 Amy v0.8.3                : 2474   20  22   891    46.1 %   2501   18.9 %
33 SoS v11-99                : 2472   33  34   359    46.0 %   2500   17.3 %
34 Tao v5.4                  : 2469   20  21   899    45.3 %   2502   19.7 %
35 Dragon v4.4.3             : 2457   21  23   766    44.3 %   2497   25.8 %
36 Comet B62-3               : 2453   21  22   800    43.4 %   2499   26.0 %
37 PostModernist v1.007      : 2437   22  21   820    41.0 %   2501   25.4 %
38 Francesca M.0.0.9         : 2432   21  20   899    40.0 %   2502   25.4 %
39 Comet B60                 : 2429   22  21   780    41.2 %   2491   25.6 %
40 Leila v0.53h              : 2422   24  20   819    38.8 %   2501   21.4 %
41 Tcb v0045                 : 2418   23  20   819    38.2 %   2501   24.9 %
42 Resp v0.19                : 2399   25  19   800    35.9 %   2500   22.2 %
43 Nejmet v3.07              : 2378   27  19   796    33.0 %   2501   22.0 %
44 SlowChess v2.78           : 2373   27  19   790    33.5 %   2492   19.6 %
45 Exchess v4.03             : 2323   31  17   799    26.3 %   2502   22.0 %
46 Beowulf v2.2              : 2303   34  16   880    23.9 %   2504   18.0 %

Hello Robert,
your rating list is very similar to mine. With so many games it tends to become realistic and reliable.
Best Regards
Heinz

[Robert Allgeuer]

Geschrieben von: / Posted by: Robert Allgeuer at 30 January 2004 14:08:01:
Als Antwort auf: / In reply to: Re: YABRL: SmarThink 0.17a third behind the Ruffians geschrieben von: / posted by: Heinz van Kempen at 30 January 2004 13:51:27:

Hello Robert,
your rating list is very similar to mine. With so many games it tends to become realistic and reliable.
Best Regards
Heinz

Yes, I agree, when comparing different rating lists I often notice a strong correlation between the results, which is a good sign ....
I am interested how the correlation will work out with your Nunn rating list, because there the influence of the book is taken out, and one can estimate, how big this influence is.
Robert