SWCR: Actual results, IvanHoe, Firebird, Gull, Hiarcs

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

Frank Quisinsky
Posts: 7183
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

SWCR: Actual results, IvanHoe, Firebird, Gull, Hiarcs

Post by Frank Quisinsky »

Hi there,

now IPP family engine 3 and 4 of 6 are still running in SWCR.

Actual results (first calculation):

Code: Select all

ELO calculation (1)
After 24.888 SWCR-64 games, 28 different engines, 46 versions included
ELO are fixed to 2.800 for Shredder 12 32bit. All other engines are derived from that fixed number.
Calculated with:  Bayesian 0056

The separate columns give the (mean) Elo performance, the + and - margins of error given with 95 % confidence,
the number of finished games, the relative score given in percentages, the average opponent ELO and finally the
relative number of draws for each program.

01. IPP Houdini 1.03a x64        2.947 22 21  920 79% 2.717 29%
02. Rybka 4 x64                  2.939 18 17 1432 81% 2.692 28%
--. Rybka 3 x64                  2.905 22 22  840 78% 2.687 28%
03. Stockfish 1.8.0 JA x64       2.905 20 19 1072 76% 2.704 31%
--. IPP Fire 1.31 x64            2.900 20 20  920 75% 2.719 36%
--. IPP IvanHoe B52aC x64        2.899 44 42  184 74% 2.732 40% NEW
--. Stockfish 1.7.1 JA x64       2.896 19 18 1120 76% 2.706 34%
--. IPP FireBird 1.1 x64 WD      2.867 42 40  184 71% 2.733 51% NEW
04. Critter 0.80 x64             2.836 18 18 1072 68% 2.706 33%
05. Naum 4.2 x64                 2.832 15 15 1672 68% 2.702 36%
--. Critter 0.70 x64             2.806 19 19  880 65% 2.701 38%
06. Shredder 12 w32              2.800 15 15 1632 63% 2.700 35%
06. Komodo 1.2 JA x64            2.800 16 15 1432 64% 2.696 39%
--. Komodo 1.0 JA x64            2.791 20 20  840 64% 2.693 40%
--. Shredder 12 x64              2.787 18 18 1080 61% 2.706 35%
08. Hiarcs 13.1 w32              2.740 41 41  184 50% 2.739 39% NEW
08. Spark 0.5 x64                2.740 15 15 1432 56% 2.698 37%
10. Thinker 5.4d Inert x64       2.735 14 14 1672 54% 2.704 39%
11. GullChess 1.0a x64           2.733 40 40  184 49% 2.739 44% NEW
12. Zappa Mexico II x64          2.724 14 14 1672 53% 2.704 40%
--. Spark 0.4 x64                2.717 19 19  840 53% 2.696 40%
13. Protector 1.3.4 JA x64       2.713 14 14 1632 51% 2.705 37%
14. Fruit 09_07_05 x64           2.705 14 14 1672 50% 2.705 34%
--. Critter 0.60 x64             2.700 20 20  840 50% 2.697 38%
15. Sjeng WC-2008 x64            2.688 14 14 1672 48% 2.705 36%
16. Hannibal 1.0a x64            2.686 18 18 1072 47% 2.712 36%
--. Protector 1.3.5 x64          2.684 19 19  840 47% 2.706 39%
17. Junior 11.2 x64              2.679 16 16 1432 48% 2.700 30%
18. Onno 1.2.70 x64              2.677 15 15 1432 47% 2.700 38%
--. Onno 1.1.1 x64               2.671 19 19  840 46% 2.698 40%
--. Junior 11.1a x64             2.648 20 20  840 43% 2.700 32%
19. Loop 2007 x64                2.631 15 15 1472 40% 2.705 36%
--. Loop M1-T x64                2.627 19 19  920 36% 2.731 36%
--. Twisted Logic 20100131x x64  2.611 18 18 1120 35% 2.716 32%
20. Umko 1.0 x64                 2.607 18 18 1072 35% 2.715 37% (ponder not possible)
21. SmarThink 1.20 x64           2.602 15 15 1672 36% 2.707 34%
22. Equinox 0.83 x64             2.597 19 19  992 32% 2.731 33%
23. Crafty 23.3 JA x64           2.592 18 18 1072 33% 2.715 34%
--. Cipollino 3.25 x64           2.565 20 20  920 28% 2.734 30%
24. BugChess2 1.7 x64            2.563 21 22  800 29% 2.718 33%
25. Scorpio 2.6 JA x64           2.555 18 18 1120 28% 2.718 32%
--. Crafty 23.2 JA x64           2.554 18 19 1120 28% 2.718 30%
26. Chronos 1.99 x64             2.552 18 18 1120 27% 2.718 33% (ponder not possible)
--. Crafty 23.3 JA x64 NP        2.546 20 21  920 25% 2.735 30% (Test: Without ponder)
27. Daydreamer 1.75 JA x64       2.521 19 19 1120 24% 2.719 30%
28. Tornado 3.6.7 x64            2.479 23 24  800 19% 2.723 24%
Bayesian give us an different ELO compare to ELOstat or the Shredder / Fritz ELO calculations if the remis quote is very high. Absolutley clear that the ELO will get a booster with more games.

Interesting is Hiarcs 13.1 w32. Hiarcs 13.1 w32 reached in SWCR-32 2.735 ELO. After 90 games in SWCR-64 only, now +- the same rating. Very stable rating for Hiarcs.

At the moment only Firebird is playing, I stopped the others. Reason are the crashes I have under Shredder GUI. On weekend the others are playing and Firebird stopped. So I can make the corrections and the tourney ended in around 10 days.

From time to time I set newer results in TalkChess.

Best
Frank
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: SWCR: Actual results, IvanHoe, Firebird, Gull, Hiarcs

Post by Dr.Wael Deeb »

Thanks Frank for the current results :D
I expect a lot from FireBird 1.1 WD....
Cheers,
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
Frank Quisinsky
Posts: 7183
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

Re: SWCR: Bayesian and ELOstat ... This could be a problem!

Post by Frank Quisinsky »

Hi Wael,

look here (at the moment Firebird 1.1 WD) is playing only.

After 60 games for Firebird 1.1 more (to my first result posting) ...
Have a look on the remis quote. That could be a problem.

Bayesian 0056:

Code: Select all

   1 Houdini 1.03a x64            2947   22   21   920   79%  2717   29% 
   2 Rybka 4 x64                  2938   18   17  1435   80%  2693   29% 
   3 Rybka 3 x64                  2905   23   22   840   78%  2687   28% 
   4 Stockfish 1.8.0 JA x64       2904   19   19  1075   76%  2704   31% 
   5 IvanHoe B52aC x64            2900   43   42   187   74%  2734   41% 
   6 Fire 1.31 x64                2899   20   20   920   75%  2719   36% 
   7 Stockfish 1.7.1 JA x64       2895   19   18  1120   76%  2706   34% 
   8 FireBird 1.1 x64 WD          2884   36   35   252   73%  2733   48% 
   9 Critter 0.80 x64             2836   18   18  1075   67%  2706   33% 
  10 Naum 4.2 x64                 2831   15   15  1675   68%  2702   35% 
  11 Critter 0.70 x64             2806   20   19   880   65%  2701   38% 
  12 Komodo 1.2 JA x64            2800   16   15  1435   64%  2696   39% 
  13 Shredder 12                  2800   15   15  1635   63%  2700   35% 
ELOstat 1.3

Code: Select all

  1 Houdini 1.03a x64              : 2956   21  21   920    79.4 %   2721   29.2 %
  2 Rybka 4 x64                    : 2943   17  17  1435    80.5 %   2697   28.6 %
  3 IvanHoe B52aC x64              : 2917   40  38   187    73.5 %   2740   41.2 %
  4 Rybka 3 x64                    : 2913   22  22   840    78.3 %   2690   27.5 %
  5 FireBird 1.1 x64 WD            : 2912   31  30   252    73.2 %   2738   48.0 %
  6 Fire 1.31 x64                  : 2910   19  19   920    74.6 %   2723   36.1 %
  7 Stockfish 1.7.1 JA x64         : 2906   18  17  1120    75.7 %   2709   33.7 %
  8 Stockfish 1.8.0 JA x64         : 2906   19  18  1075    75.6 %   2709   30.9 %
  9 Critter 0.80 x64               : 2838   18  17  1075    67.5 %   2711   33.0 %
 10 Naum 4.2 x64                   : 2834   14  14  1675    67.7 %   2706   35.5 %
 11 Critter 0.70 x64               : 2809   18  18   880    64.7 %   2704   38.1 %
 12 Komodo 1.2 JA x64              : 2802   14  14  1435    64.3 %   2701   39.3 %
 13 Shredder 12                    : 2800   14  14  1635    63.4 %   2705   34.9 %
You can see, that ELOstat (Fritz and Shredder GUI calculation too) will give a better result for Firebird.

I wrote it before, too many remis games for Bayesian!

Best
Frank
Frank Quisinsky
Posts: 7183
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

Re: SWCR: Bayesian and ELOstat ... This could be a problem!

Post by Frank Quisinsky »

Hi Wael,

no Wael, too many Remis games produced your setting with Fire 1.1 x64. Perhaps a strong setting with fast games vs. the other engines in IPP family or Rybka 3, 4 but its after 300 games clear that Fire 1.1 x64 plays vs. weaker opponents to many games remis.

It seems that IvanHoe have the same problem.

Bayesian 0056:

Code: Select all

   1 Houdini 1.03a x64            2947   22   21   920   79%  2717   29% 
   2 Rybka 4 x64                  2938   18   17  1436   80%  2693   29% 
   3 Rybka 3 x64                  2905   23   22   840   78%  2687   28% 
   4 Stockfish 1.8.0 JA x64       2903   19   19  1077   75%  2704   31% 
   5 Fire 1.31 x64                2899   20   20   920   75%  2719   36% 
   6 IvanHoe B52aC x64            2898   43   41   189   73%  2736   41% 
   7 Stockfish 1.7.1 JA x64       2895   19   18  1120   76%  2706   34% 
   8 FireBird 1.1 x64 WD          2886   34   33   295   74%  2732   47% 
   9 Critter 0.80 x64             2836   18   18  1077   67%  2707   33% 
  10 Naum 4.2 x64                 2831   15   15  1677   68%  2702   36% 
  11 Critter 0.70 x64             2806   20   19   880   65%  2701   38% 
  12 Shredder 12                  2800   15   15  1636   63%  2700   35% 
  13 Komodo 1.2 JA x64            2800   15   15  1437   64%  2697   39% 
  14 Komodo 1.0 JA x64            2790   20   20   840   64%  2692   40% 
  15 Shredder 12 x64              2787   18   18  1080   61%  2706   35% 
  16 Hiarcs 13.1                  2741   40   40   189   49%  2743   40% 
  17 Spark 0.5 x64                2740   15   15  1437   56%  2698   37% 
  18 Thinker 5.4d Inert x64       2735   14   14  1677   54%  2704   39% 
  19 GullChess 1.0a x64           2731   40   40   189   48%  2743   43% 
ELOstat 1.3

Code: Select all

  1 Houdini 1.03a x64              : 2956   21  21   920    79.4 %   2721   29.2 %
  2 Rybka 4 x64                    : 2943   17  17  1436    80.5 %   2697   28.6 %
  3 IvanHoe B52aC x64              : 2915   39  38   189    73.0 %   2742   41.3 %
  4 FireBird 1.1 x64 WD            : 2915   29  28   295    73.6 %   2737   47.5 %
  5 Rybka 3 x64                    : 2912   22  22   840    78.3 %   2690   27.5 %
  6 Fire 1.31 x64                  : 2910   19  19   920    74.6 %   2723   36.1 %
  7 Stockfish 1.7.1 JA x64         : 2906   18  17  1120    75.7 %   2709   33.7 %
  8 Stockfish 1.8.0 JA x64         : 2904   19  18  1077    75.5 %   2709   30.8 %
  9 Critter 0.80 x64               : 2838   18  17  1077    67.4 %   2711   33.1 %
 10 Naum 4.2 x64                   : 2834   14  14  1677    67.7 %   2706   35.5 %
 11 Critter 0.70 x64               : 2809   18  18   880    64.7 %   2704   38.1 %
 12 Komodo 1.2 JA x64              : 2802   14  14  1437    64.2 %   2701   39.2 %
 13 Shredder 12                    : 2800   14  14  1636    63.4 %   2705   34.9 %
 14 Komodo 1.0 JA x64              : 2797   19  18   840    64.3 %   2695   39.5 %
 15 Shredder 12 x64                : 2789   17  17  1080    61.3 %   2710   35.1 %
 16 Hiarcs 13.1                    : 2745   39  39   189    49.5 %   2749   39.7 %
 17 Spark 0.5 x64                  : 2744   14  14  1437    55.9 %   2702   37.3 %
 18 Thinker 5.4d Inert x64         : 2738   13  13  1677    54.2 %   2708   39.3 %
 19 GullChess 1.0a x64             : 2736   37  37   189    48.1 %   2749   43.4 %
But let us look from time to time with more games ... 920 games are to play.

Best
Frank
Frank Quisinsky
Posts: 7183
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

Re: SWCR: Most boring chess I ever saw ...

Post by Frank Quisinsky »

Hi Wael,

again around after 70 games more ...
To many remis games, after 366 FireBird 1.1 x64 WD games too!

Bayesian 0056

Code: Select all

   1 Houdini 1.03a x64            2947   22   21   920   79%  2717   29% 
   2 Rybka 4 x64                  2939   18   17  1439   80%  2693   29% 
   3 Rybka 3 x64                  2904   23   22   840   78%  2687   28% 
   4 Stockfish 1.8.0 JA x64       2903   19   19  1079   75%  2705   31% 
   5 Fire 1.31 x64                2899   20   20   920   75%  2719   36% 
   6 IvanHoe B52aC x64            2896   42   41   192   72%  2739   42% 
   7 Stockfish 1.7.1 JA x64       2895   19   18  1120   76%  2706   34% 
   8 FireBird 1.1 x64 WD          2885   30   30   366   73%  2732   46% 
   9 Critter 0.80 x64             2835   18   18  1080   67%  2707   33% 
  10 Naum 4.2 x64                 2831   15   15  1680   68%  2702   36% 
  11 Critter 0.70 x64             2806   20   19   880   65%  2701   38% 
  12 Shredder 12                  2800   15   15  1639   63%  2701   35% 
  13 Komodo 1.2 JA x64            2800   15   15  1440   64%  2697   39% 
  14 Komodo 1.0 JA x64            2790   20   20   840   64%  2692   40% 
  15 Shredder 12 x64              2787   18   18  1080   61%  2706   35% 
  16 Hiarcs 13.1                  2744   40   40   193   49%  2746   40% 
  17 Spark 0.5 x64                2740   15   15  1440   56%  2699   37% 
  18 Thinker 5.4d Inert x64       2735   14   14  1680   54%  2704   39% 
  19 GullChess 1.0a x64           2732   40   40   192   48%  2745   44% 
  20 Zappa Mexico II x64          2724   14   14  1680   53%  2705   40% 
ELOstat 1.3

Code: Select all

  1 Houdini 1.03a x64              : 2956   21  21   920    79.4 %   2722   29.2 %
  2 Rybka 4 x64                    : 2944   17  17  1439    80.5 %   2698   28.6 %
  3 Rybka 3 x64                    : 2913   22  22   840    78.3 %   2691   27.5 %
  4 IvanHoe B52aC x64              : 2912   39  38   192    72.4 %   2745   41.7 %
  5 Fire 1.31 x64                  : 2911   19  19   920    74.6 %   2724   36.1 %
  6 FireBird 1.1 x64 WD            : 2908   26  26   366    72.8 %   2737   45.6 %
  7 Stockfish 1.7.1 JA x64         : 2906   18  17  1120    75.7 %   2709   33.7 %
  8 Stockfish 1.8.0 JA x64         : 2905   18  18  1079    75.4 %   2710   31.0 %
  9 Critter 0.80 x64               : 2838   18  17  1080    67.3 %   2713   33.1 %
 10 Naum 4.2 x64                   : 2835   14  14  1680    67.6 %   2707   35.6 %
 11 Critter 0.70 x64               : 2810   18  18   880    64.7 %   2705   38.1 %
 12 Komodo 1.2 JA x64              : 2803   14  14  1440    64.1 %   2702   39.3 %
 13 Shredder 12                    : 2800   14  14  1639    63.3 %   2706   35.0 %
 14 Komodo 1.0 JA x64              : 2798   19  18   840    64.3 %   2696   39.5 %
 15 Shredder 12 x64                : 2790   17  17  1080    61.3 %   2710   35.1 %
 16 Hiarcs 13.1                    : 2749   38  38   193    49.5 %   2752   39.9 %
 17 Spark 0.5 x64                  : 2745   14  14  1440    55.9 %   2703   37.3 %
 18 Thinker 5.4d Inert x64         : 2738   13  13  1680    54.2 %   2709   39.3 %
 19 GullChess 1.0a x64             : 2737   37  37   192    47.9 %   2752   43.8 %
 20 Zappa Mexico II x64            : 2728   13  13  1680    52.6 %   2710   40.0 %
Look here:

Code: Select all

Zappa Mexico II x64           :  16 (+  7,=  9,-  0), 71.9 %
Protector 1.3.4 JA x64        :  16 (+  7,=  8,-  1), 68.8 %
Naum 4.2 x64                  :  16 (+  5,= 11,-  0), 65.6 %
Loop 2007 x64                 :  16 (+ 12,=  4,-  0), 87.5 %
SmarThink 1.20 x64            :  15 (+ 14,=  1,-  0), 96.7 %
Thinker 5.4d Inert x64        :  16 (+  8,=  8,-  0), 75.0 %
Sjeng WC-2008 x64             :  16 (+ 12,=  3,-  1), 84.4 %
Komodo 1.2 JA x64             :  16 (+  5,= 11,-  0), 65.6 %
Junior 11.2 x64               :  16 (+ 10,=  4,-  2), 75.0 %
Fruit 09_07_05 x64            :  16 (+  9,=  7,-  0), 78.1 %
Rybka 4 x64                   :  15 (+  0,= 11,-  4), 36.7 %
Onno 1.2.70 x64               :  16 (+ 11,=  4,-  1), 81.2 %
Spark 0.5 x64                 :  16 (+  9,=  5,-  2), 71.9 %
Shredder 12                   :  15 (+  7,=  7,-  1), 70.0 %
Stockfish 1.8.0 JA x64        :  15 (+  8,=  6,-  1), 73.3 %
Hannibal 1.0a x64             :  17 (+  4,= 12,-  1), 58.8 %
Critter 0.80 x64              :  16 (+  7,=  9,-  0), 71.9 %
Crafty 23.3 JA x64            :  16 (+ 11,=  5,-  0), 84.4 %
Umko 1.0 x64                  :  16 (+ 10,=  6,-  0), 81.2 %
Equinox 0.83 x64              :  16 (+ 11,=  5,-  0), 84.4 %
GullChess 1.0a x64            :  16 (+  8,=  8,-  0), 75.0 %
IvanHoe B52aC x64             :  16 (+  2,= 13,-  1), 53.1 %
Hiarcs 13.1                   :  17 (+  6,= 10,-  1), 64.7 %
Or ...
6 FireBird 1.1 x64 WD : 2908 366 (+183,=167,- 16), 72.8 %

Very boring games, no fast win games ... more or less to many games goes in endgame. Firebird 1.1 win the very late middle games or played remis.

Sorry but your setting isn't good with longer time controls and ponder = on.

920 games are to play. But I am absolutley sure no bigger changes if all games played.

Shredder Classic 4.0 GUI Crash quota = around 6%, 25 more as Houdini has. Could be cohere with the remis quota.

Wael, I respect your work but what I saw (from IvanHoe too) is the most boring chess in SWCR. Each other engine plays more interesting aggressive chess as Fire and Ivanhoe. Can be see in each statistic you can make in own work with the games if available.

Ipp Family are the new remis king :-)
But let us wait of all results. After IvanHoe B0.52aC x64 and Firebird 1.1 WD x64 two other IvanHoe versions I will test.

Best
Frank
Frank Quisinsky
Posts: 7183
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

Re: SWCR: FireBird 1.1 WD = Saitek Simultano ...

Post by Frank Quisinsky »

Hi Wael,

most boring chess I ever saw (for me as kibitzer) played the chess computer Saitek Simultano. For over 20 years I made a Match vs. Mephisto Polgar 5Mhz and tournemant rules (40 moves in 120 minutes).

20 games!
Each game goes over 100 moves and Polgar win with 17.5 : 2.5!

Must thinking on it ...
After this matches I played the next 3 months football in my free time. No computer chess in this time, lost my interest :-)

Honest:
FireBird don't have important endgame knowledge, Rybka have. Rybka played also very boring chess. I like aggressive computer chess. But Rybka made some points more in endgame as FireBird.

I hope that one of the IvenHoe's (works fine without Shredder GUI crashes, time management is better too) plays more interesting computer chess.

If I look in Spark games ...
What for a great engine!

Code: Select all

[Event "SWCR, 40/10, ponder = on, Q9550 "]
[Site "Trier"]
[Date "2010.09.23"]
[Round "34.4"]
[White "Spark 0.5 x64"]
[Black "Fire 1.31 x64"]
[Result "1-0"]
[ECO "B10"]
[PlyCount "69"]
[EventDate "2010.??.??"]
[EventType "rapid"]
[EventCountry "GER"]
[EventCategory "19"]
[Source "Frank Quisinsky"]
[SourceDate "2010.09.26"]

1. e4 c6 2. d3 d5 3. Nd2 g6 4. f4 Bg7 5. Ngf3 Nh6 6. e5 d4 7. Nb3 c5 8. c3 dxc3
9. bxc3 Qc7 10. Ba3 b6 11. d4 O-O 12. Bb5 a6 13. Be2 Rd8 14. O-O Nf5 15. Qd3 a5
16. d5 Na6 17. Ng5 Qa7 18. Rf2 h6 19. Ne4 Bb7 20. c4 e6 21. g4 exd5 22. Nf6+
Bxf6 23. exf6 Bc8 24. Qh3 Kh7 25. Nd2 Nd6 26. f5 Nb4 27. g5 h5 28. Bxh5 gxh5
29. Qxh5+ Kg8 30. Qh6 Ne8 31. Rf4 Qe7 32. fxe7 Rd6 33. Qf8+ Kh7 34. Rh4+ Rh6
35. Rxh6# 1-0
Perhaps in 5.000 IPP family games I can see 1x such a game. I think never ever ...

Best
Frank
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: SWCR: Most boring chess I ever saw ...

Post by Dr.Wael Deeb »

Frank Quisinsky wrote:Hi Wael,

again around after 70 games more ...
To many remis games, after 366 FireBird 1.1 x64 WD games too!

Bayesian 0056

Code: Select all

   1 Houdini 1.03a x64            2947   22   21   920   79%  2717   29% 
   2 Rybka 4 x64                  2939   18   17  1439   80%  2693   29% 
   3 Rybka 3 x64                  2904   23   22   840   78%  2687   28% 
   4 Stockfish 1.8.0 JA x64       2903   19   19  1079   75%  2705   31% 
   5 Fire 1.31 x64                2899   20   20   920   75%  2719   36% 
   6 IvanHoe B52aC x64            2896   42   41   192   72%  2739   42% 
   7 Stockfish 1.7.1 JA x64       2895   19   18  1120   76%  2706   34% 
   8 FireBird 1.1 x64 WD          2885   30   30   366   73%  2732   46% 
   9 Critter 0.80 x64             2835   18   18  1080   67%  2707   33% 
  10 Naum 4.2 x64                 2831   15   15  1680   68%  2702   36% 
  11 Critter 0.70 x64             2806   20   19   880   65%  2701   38% 
  12 Shredder 12                  2800   15   15  1639   63%  2701   35% 
  13 Komodo 1.2 JA x64            2800   15   15  1440   64%  2697   39% 
  14 Komodo 1.0 JA x64            2790   20   20   840   64%  2692   40% 
  15 Shredder 12 x64              2787   18   18  1080   61%  2706   35% 
  16 Hiarcs 13.1                  2744   40   40   193   49%  2746   40% 
  17 Spark 0.5 x64                2740   15   15  1440   56%  2699   37% 
  18 Thinker 5.4d Inert x64       2735   14   14  1680   54%  2704   39% 
  19 GullChess 1.0a x64           2732   40   40   192   48%  2745   44% 
  20 Zappa Mexico II x64          2724   14   14  1680   53%  2705   40% 
ELOstat 1.3

Code: Select all

  1 Houdini 1.03a x64              : 2956   21  21   920    79.4 %   2722   29.2 %
  2 Rybka 4 x64                    : 2944   17  17  1439    80.5 %   2698   28.6 %
  3 Rybka 3 x64                    : 2913   22  22   840    78.3 %   2691   27.5 %
  4 IvanHoe B52aC x64              : 2912   39  38   192    72.4 %   2745   41.7 %
  5 Fire 1.31 x64                  : 2911   19  19   920    74.6 %   2724   36.1 %
  6 FireBird 1.1 x64 WD            : 2908   26  26   366    72.8 %   2737   45.6 %
  7 Stockfish 1.7.1 JA x64         : 2906   18  17  1120    75.7 %   2709   33.7 %
  8 Stockfish 1.8.0 JA x64         : 2905   18  18  1079    75.4 %   2710   31.0 %
  9 Critter 0.80 x64               : 2838   18  17  1080    67.3 %   2713   33.1 %
 10 Naum 4.2 x64                   : 2835   14  14  1680    67.6 %   2707   35.6 %
 11 Critter 0.70 x64               : 2810   18  18   880    64.7 %   2705   38.1 %
 12 Komodo 1.2 JA x64              : 2803   14  14  1440    64.1 %   2702   39.3 %
 13 Shredder 12                    : 2800   14  14  1639    63.3 %   2706   35.0 %
 14 Komodo 1.0 JA x64              : 2798   19  18   840    64.3 %   2696   39.5 %
 15 Shredder 12 x64                : 2790   17  17  1080    61.3 %   2710   35.1 %
 16 Hiarcs 13.1                    : 2749   38  38   193    49.5 %   2752   39.9 %
 17 Spark 0.5 x64                  : 2745   14  14  1440    55.9 %   2703   37.3 %
 18 Thinker 5.4d Inert x64         : 2738   13  13  1680    54.2 %   2709   39.3 %
 19 GullChess 1.0a x64             : 2737   37  37   192    47.9 %   2752   43.8 %
 20 Zappa Mexico II x64            : 2728   13  13  1680    52.6 %   2710   40.0 %
Look here:

Code: Select all

Zappa Mexico II x64           :  16 (+  7,=  9,-  0), 71.9 %
Protector 1.3.4 JA x64        :  16 (+  7,=  8,-  1), 68.8 %
Naum 4.2 x64                  :  16 (+  5,= 11,-  0), 65.6 %
Loop 2007 x64                 :  16 (+ 12,=  4,-  0), 87.5 %
SmarThink 1.20 x64            :  15 (+ 14,=  1,-  0), 96.7 %
Thinker 5.4d Inert x64        :  16 (+  8,=  8,-  0), 75.0 %
Sjeng WC-2008 x64             :  16 (+ 12,=  3,-  1), 84.4 %
Komodo 1.2 JA x64             :  16 (+  5,= 11,-  0), 65.6 %
Junior 11.2 x64               :  16 (+ 10,=  4,-  2), 75.0 %
Fruit 09_07_05 x64            :  16 (+  9,=  7,-  0), 78.1 %
Rybka 4 x64                   :  15 (+  0,= 11,-  4), 36.7 %
Onno 1.2.70 x64               :  16 (+ 11,=  4,-  1), 81.2 %
Spark 0.5 x64                 :  16 (+  9,=  5,-  2), 71.9 %
Shredder 12                   :  15 (+  7,=  7,-  1), 70.0 %
Stockfish 1.8.0 JA x64        :  15 (+  8,=  6,-  1), 73.3 %
Hannibal 1.0a x64             :  17 (+  4,= 12,-  1), 58.8 %
Critter 0.80 x64              :  16 (+  7,=  9,-  0), 71.9 %
Crafty 23.3 JA x64            :  16 (+ 11,=  5,-  0), 84.4 %
Umko 1.0 x64                  :  16 (+ 10,=  6,-  0), 81.2 %
Equinox 0.83 x64              :  16 (+ 11,=  5,-  0), 84.4 %
GullChess 1.0a x64            :  16 (+  8,=  8,-  0), 75.0 %
IvanHoe B52aC x64             :  16 (+  2,= 13,-  1), 53.1 %
Hiarcs 13.1                   :  17 (+  6,= 10,-  1), 64.7 %
Or ...
6 FireBird 1.1 x64 WD : 2908 366 (+183,=167,- 16), 72.8 %

Very boring games, no fast win games ... more or less to many games goes in endgame. Firebird 1.1 win the very late middle games or played remis.

Sorry but your setting isn't good with longer time controls and ponder = on.

920 games are to play. But I am absolutley sure no bigger changes if all games played.

Shredder Classic 4.0 GUI Crash quota = around 6%, 25 more as Houdini has. Could be cohere with the remis quota.

Wael, I respect your work but what I saw (from IvanHoe too) is the most boring chess in SWCR. Each other engine plays more interesting aggressive chess as Fire and Ivanhoe. Can be see in each statistic you can make in own work with the games if available.

Ipp Family are the new remis king :-)
But let us wait of all results. After IvanHoe B0.52aC x64 and Firebird 1.1 WD x64 two other IvanHoe versions I will test.

Best
Frank
Hi Frank,
Maybe the playing style is boring,but the performance of 72% against these other chess monsters is impressive....
Strange enough,running on my overclocked quad,FireBird 1.1 wd x64 plays extremely strong and it's leading my rating list thus far :D
I hope my settings don't return you to the football playing again :wink:
Cheers,
Dr.D

P.S.note that there are 16 loses only out of 366....don't know about you but I am haapy with that....
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
Frank Quisinsky
Posts: 7183
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

Re: SWCR FireBird 1.1 WD x64: After 488 of 920 ...

Post by Frank Quisinsky »

Hi Weal,

Bayesian 0056:

Code: Select all

   1 Houdini 1.03a x64            2947   22   21   920   79%  2717   29% 
   2 Rybka 4 x64                  2939   17   17  1445   80%  2694   29% 
   3 Rybka 3 x64                  2905   23   22   840   78%  2687   28% 
   4 Stockfish 1.8.0 JA x64       2903   19   19  1085   75%  2706   31% 
   5 Fire 1.31 x64                2899   20   20   920   75%  2719   36% 
   6 IvanHoe B52aC x64            2896   41   40   198   72%  2743   42% 
   7 Stockfish 1.7.1 JA x64       2895   19   18  1120   76%  2706   34% 
   8 FireBird 1.1 x64 WD          2886   26   26   488   73%  2732   44% 
   9 Critter 0.80 x64             2835   18   18  1085   67%  2708   33% 
  10 Naum 4.2 x64                 2832   15   14  1685   68%  2703   36% 
  11 Critter 0.70 x64             2806   20   19   880   65%  2701   38% 
  12 Shredder 12                  2800   15   15  1645   63%  2701   35% 
  13 Komodo 1.2 JA x64            2799   15   15  1444   64%  2698   39% 
  14 Komodo 1.0 JA x64            2790   20   20   840   64%  2692   40% 
  15 Shredder 12 x64              2787   18   18  1080   61%  2706   35% 
  16 Hiarcs 13.1                  2741   40   40   198   48%  2749   39% 
  17 Spark 0.5 x64                2740   15   15  1445   56%  2699   37% 
  18 Thinker 5.4d Inert x64       2735   14   14  1685   54%  2705   39% 
  19 GullChess 1.0a x64           2733   39   39   198   47%  2750   43% 
  20 Zappa Mexico II x64          2724   14   14  1685   53%  2705   40% 
ELOstat:

Code: Select all

  1 Houdini 1.03a x64              : 2956   21  21   920    79.4 %   2721   29.2 %
  2 Rybka 4 x64                    : 2943   17  17  1445    80.4 %   2698   28.9 %
  3 Rybka 3 x64                    : 2913   22  22   840    78.3 %   2690   27.5 %
  4 IvanHoe B52aC x64              : 2910   38  37   198    71.7 %   2749   42.4 %
  5 Fire 1.31 x64                  : 2910   19  19   920    74.6 %   2723   36.1 %
  6 Stockfish 1.7.1 JA x64         : 2906   18  17  1120    75.7 %   2709   33.7 %
  7 FireBird 1.1 x64 WD            : 2906   23  23   488    72.5 %   2737   43.9 %
  8 Stockfish 1.8.0 JA x64         : 2905   18  18  1085    75.3 %   2711   31.2 %
  9 Critter 0.80 x64               : 2837   17  17  1085    67.1 %   2713   33.2 %
 10 Naum 4.2 x64                   : 2835   14  14  1685    67.6 %   2707   35.7 %
 11 Critter 0.70 x64               : 2810   18  18   880    64.7 %   2704   38.1 %
 12 Komodo 1.2 JA x64              : 2802   14  14  1444    64.0 %   2702   39.3 %
 13 Shredder 12                    : 2800   14  14  1645    63.2 %   2706   35.0 %
 14 Komodo 1.0 JA x64              : 2798   19  18   840    64.3 %   2696   39.5 %
 15 Shredder 12 x64                : 2789   17  17  1080    61.3 %   2710   35.1 %
 16 Hiarcs 13.1                    : 2745   38  38   198    48.5 %   2755   39.4 %
 17 Spark 0.5 x64                  : 2745   14  14  1445    55.9 %   2704   37.3 %
 18 GullChess 1.0a x64             : 2738   37  37   198    47.5 %   2756   43.4 %
 19 Thinker 5.4d Inert x64         : 2738   13  13  1685    54.1 %   2709   39.3 %
 20 Zappa Mexico II x64            : 2728   13  13  1685    52.6 %   2710   40.1 %
Really interesting:
20 ELO difference between Bayesian 0056 and ELOstat 1.3.

Code: Select all

7 FireBird 1.1 x64 WD       : 2906  488 (+247,=214,- 27), 72.5 %

Zappa Mexico II x64           :  21 (+  8,= 13,-  0), 69.0 %
Protector 1.3.4 JA x64        :  21 (+ 12,=  8,-  1), 76.2 %
Naum 4.2 x64                  :  21 (+  6,= 14,-  1), 61.9 %
Loop 2007 x64                 :  20 (+ 16,=  4,-  0), 90.0 %
SmarThink 1.20 x64            :  21 (+ 17,=  4,-  0), 90.5 %
Thinker 5.4d Inert x64        :  21 (+ 12,=  9,-  0), 78.6 %
Sjeng WC-2008 x64             :  21 (+ 14,=  6,-  1), 81.0 %
Komodo 1.2 JA x64             :  20 (+  7,= 13,-  0), 67.5 %
Junior 11.2 x64               :  21 (+ 13,=  5,-  3), 73.8 %
Fruit 09_07_05 x64            :  22 (+ 12,=  7,-  3), 70.5 %
Rybka 4 x64                   :  21 (+  0,= 16,-  5), 38.1 %
Onno 1.2.70 x64               :  21 (+ 14,=  6,-  1), 81.0 %
Spark 0.5 x64                 :  21 (+ 11,=  7,-  3), 69.0 %
Shredder 12                   :  21 (+ 10,= 10,-  1), 71.4 %
Stockfish 1.8.0 JA x64        :  21 (+  8,= 11,-  2), 64.3 %
Hannibal 1.0a x64             :  22 (+  8,= 13,-  1), 65.9 %
Critter 0.80 x64              :  21 (+  9,= 12,-  0), 71.4 %
Crafty 23.3 JA x64            :  21 (+ 15,=  6,-  0), 85.7 %
Umko 1.0 x64                  :  22 (+ 16,=  6,-  0), 86.4 %
Equinox 0.83 x64              :  22 (+ 15,=  6,-  1), 81.8 %
GullChess 1.0a x64            :  22 (+ 11,= 10,-  1), 72.7 %
IvanHoe B52aC x64             :  22 (+  3,= 17,-  2), 52.3 %
Hiarcs 13.1                   :  22 (+ 10,= 11,-  1), 70.5 %
29 Shredder Classic 4.0 GUI crashes after 488 games = 5,94%.
Houdini 1.03a x64 made 36 om 920 = 3,91%

For the other readers ...
This crashes comes only under Shredder Classic 4.0 GUI with ponder = on. More often if resign factor = off.

Remis quote now 43,9%.
Perhaps after 920 games the same performance as Fire 1.31 x64?
Let us wait :-)

Best
Frank
User avatar
Dr.Wael Deeb
Posts: 9773
Joined: Wed Mar 08, 2006 8:44 pm
Location: Amman,Jordan

Re: SWCR FireBird 1.1 WD x64: After 488 of 920 ...

Post by Dr.Wael Deeb »

Frank Quisinsky wrote:Hi Weal,

Bayesian 0056:

Code: Select all

   1 Houdini 1.03a x64            2947   22   21   920   79%  2717   29% 
   2 Rybka 4 x64                  2939   17   17  1445   80%  2694   29% 
   3 Rybka 3 x64                  2905   23   22   840   78%  2687   28% 
   4 Stockfish 1.8.0 JA x64       2903   19   19  1085   75%  2706   31% 
   5 Fire 1.31 x64                2899   20   20   920   75%  2719   36% 
   6 IvanHoe B52aC x64            2896   41   40   198   72%  2743   42% 
   7 Stockfish 1.7.1 JA x64       2895   19   18  1120   76%  2706   34% 
   8 FireBird 1.1 x64 WD          2886   26   26   488   73%  2732   44% 
   9 Critter 0.80 x64             2835   18   18  1085   67%  2708   33% 
  10 Naum 4.2 x64                 2832   15   14  1685   68%  2703   36% 
  11 Critter 0.70 x64             2806   20   19   880   65%  2701   38% 
  12 Shredder 12                  2800   15   15  1645   63%  2701   35% 
  13 Komodo 1.2 JA x64            2799   15   15  1444   64%  2698   39% 
  14 Komodo 1.0 JA x64            2790   20   20   840   64%  2692   40% 
  15 Shredder 12 x64              2787   18   18  1080   61%  2706   35% 
  16 Hiarcs 13.1                  2741   40   40   198   48%  2749   39% 
  17 Spark 0.5 x64                2740   15   15  1445   56%  2699   37% 
  18 Thinker 5.4d Inert x64       2735   14   14  1685   54%  2705   39% 
  19 GullChess 1.0a x64           2733   39   39   198   47%  2750   43% 
  20 Zappa Mexico II x64          2724   14   14  1685   53%  2705   40% 
ELOstat:

Code: Select all

  1 Houdini 1.03a x64              : 2956   21  21   920    79.4 %   2721   29.2 %
  2 Rybka 4 x64                    : 2943   17  17  1445    80.4 %   2698   28.9 %
  3 Rybka 3 x64                    : 2913   22  22   840    78.3 %   2690   27.5 %
  4 IvanHoe B52aC x64              : 2910   38  37   198    71.7 %   2749   42.4 %
  5 Fire 1.31 x64                  : 2910   19  19   920    74.6 %   2723   36.1 %
  6 Stockfish 1.7.1 JA x64         : 2906   18  17  1120    75.7 %   2709   33.7 %
  7 FireBird 1.1 x64 WD            : 2906   23  23   488    72.5 %   2737   43.9 %
  8 Stockfish 1.8.0 JA x64         : 2905   18  18  1085    75.3 %   2711   31.2 %
  9 Critter 0.80 x64               : 2837   17  17  1085    67.1 %   2713   33.2 %
 10 Naum 4.2 x64                   : 2835   14  14  1685    67.6 %   2707   35.7 %
 11 Critter 0.70 x64               : 2810   18  18   880    64.7 %   2704   38.1 %
 12 Komodo 1.2 JA x64              : 2802   14  14  1444    64.0 %   2702   39.3 %
 13 Shredder 12                    : 2800   14  14  1645    63.2 %   2706   35.0 %
 14 Komodo 1.0 JA x64              : 2798   19  18   840    64.3 %   2696   39.5 %
 15 Shredder 12 x64                : 2789   17  17  1080    61.3 %   2710   35.1 %
 16 Hiarcs 13.1                    : 2745   38  38   198    48.5 %   2755   39.4 %
 17 Spark 0.5 x64                  : 2745   14  14  1445    55.9 %   2704   37.3 %
 18 GullChess 1.0a x64             : 2738   37  37   198    47.5 %   2756   43.4 %
 19 Thinker 5.4d Inert x64         : 2738   13  13  1685    54.1 %   2709   39.3 %
 20 Zappa Mexico II x64            : 2728   13  13  1685    52.6 %   2710   40.1 %
Really interesting:
20 ELO difference between Bayesian 0056 and ELOstat 1.3.

Code: Select all

7 FireBird 1.1 x64 WD       : 2906  488 (+247,=214,- 27), 72.5 %

Zappa Mexico II x64           :  21 (+  8,= 13,-  0), 69.0 %
Protector 1.3.4 JA x64        :  21 (+ 12,=  8,-  1), 76.2 %
Naum 4.2 x64                  :  21 (+  6,= 14,-  1), 61.9 %
Loop 2007 x64                 :  20 (+ 16,=  4,-  0), 90.0 %
SmarThink 1.20 x64            :  21 (+ 17,=  4,-  0), 90.5 %
Thinker 5.4d Inert x64        :  21 (+ 12,=  9,-  0), 78.6 %
Sjeng WC-2008 x64             :  21 (+ 14,=  6,-  1), 81.0 %
Komodo 1.2 JA x64             :  20 (+  7,= 13,-  0), 67.5 %
Junior 11.2 x64               :  21 (+ 13,=  5,-  3), 73.8 %
Fruit 09_07_05 x64            :  22 (+ 12,=  7,-  3), 70.5 %
Rybka 4 x64                   :  21 (+  0,= 16,-  5), 38.1 %
Onno 1.2.70 x64               :  21 (+ 14,=  6,-  1), 81.0 %
Spark 0.5 x64                 :  21 (+ 11,=  7,-  3), 69.0 %
Shredder 12                   :  21 (+ 10,= 10,-  1), 71.4 %
Stockfish 1.8.0 JA x64        :  21 (+  8,= 11,-  2), 64.3 %
Hannibal 1.0a x64             :  22 (+  8,= 13,-  1), 65.9 %
Critter 0.80 x64              :  21 (+  9,= 12,-  0), 71.4 %
Crafty 23.3 JA x64            :  21 (+ 15,=  6,-  0), 85.7 %
Umko 1.0 x64                  :  22 (+ 16,=  6,-  0), 86.4 %
Equinox 0.83 x64              :  22 (+ 15,=  6,-  1), 81.8 %
GullChess 1.0a x64            :  22 (+ 11,= 10,-  1), 72.7 %
IvanHoe B52aC x64             :  22 (+  3,= 17,-  2), 52.3 %
Hiarcs 13.1                   :  22 (+ 10,= 11,-  1), 70.5 %
29 Shredder Classic 4.0 GUI crashes after 488 games = 5,94%.
Houdini 1.03a x64 made 36 om 920 = 3,91%

For the other readers ...
This crashes comes only under Shredder Classic 4.0 GUI with ponder = on. More often if resign factor = off.

Remis quote now 43,9%.
Perhaps after 920 games the same performance as Fire 1.31 x64?Let us wait :-)

Best
Frank
Nope,Fire 1.31 will drop behind FireBird 1.1 wd....it's about 30-40 Elo weaker in my rating list :D
Cheers,
Dr.D
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
Frank Quisinsky
Posts: 7183
Joined: Wed Nov 18, 2009 7:16 pm
Location: Gutweiler, Germany
Full name: Frank Quisinsky

Re: SWCR FireBird 1.1 WD x64: After 488 of 920 ...

Post by Frank Quisinsky »

Hi Wael,

have a look in the comparsion CEGT to SWCR.
No engine have more as 30 ELO difference.

I think the my SWCR ratings are very exact to the conditions I used.

I think you test with very fast time controls. In this case is the difference to your rating list is clear ... think so!

Best
Frank