FCT1: Rating List, 46.000 45-minutes games on i7 4.3Ghz

[Frank Quisinsky]

Hi there,

and here the actual FCT1 Rating List after 46.000 games:

Code: Select all

 "FCT1 Rating List"
 Will be updated after ~ 1000 new games!
 ---------------------------------------
  
 Date         : December 17th, 2014 (02:00)
 FCT1         : 46000 games
 Version      : 1.16.300 S1
 Hardware     : 2x Intel Core i7-4770K *4.3Ghz, 45-minutes games
 Engines      : Maximum allowed = three versions of same engine!
 
 Calculation  : Ordo 0.97, "Hiarcs 14 WCSC w32" with 2.825 Elo / News 057 & News 062
 Parameter    : Ordo_097.exe -a 2825 -A "Hiarcs 14 WCSC w32" -p 1.pgn -o rating.txt -W -s1000 -D -E -V
  
 
      Program                          Elo    +    -   Games   Score  Av.Op.  Draw	  
  01. Komodo 8 x64                     3107   17   17  1500    81.1%   2822   30.5%
  02. Stockfish 03.08.14 BMI2 x64      3104   17   17  1600    82.7%   2804   30.2%
  --. Stockfish 02.10.14 BMI2 x64 C    3104   19   19  1100    80.0%   2838   34.0%
  --. Komodo 7a x64                    3079   19   19  1250    79.6%   2816   31.9%
  --. Stockfish 5 SSE42 x64            3077   21   21  1000    79.6%   2819   35.4%
  03. Fire 4 x64                       3056   17   17  1600    80.2%   2785   29.2%  NEW, +105 Elo, runs as S1 (Special -01)
  --. Komodo TCECr x64                 3049   20   20  1000    80.3%   2787   32.2%
  04. GullChess 3.0 BMI2 x64           3038   14   14  1800    74.3%   2828   37.3%
  --. GullChess 2.8 Beta BMI2 x64      2986   19   19  1000    74.0%   2790   37.7%
  --. Fire 3.0 AVX x64                 2951   13   13  2050    65.5%   2825   41.2%
  05. Protector 1.7.0 x64              2905   16   16  1150    56.3%   2858   44.0%
  06. Chiron 2.0 x64                   2895   11   11  2400    59.8%   2820   39.2%
  07. Critter 0.90 SSE4 x64            2887   15   15  1350    56.6%   2837   42.7%
  --. Protector 1.6.0 x64              2882   14   14  1600    58.7%   2816   44.4%
  08. Hannibal 1.4b x64                2862   11   11  2400    55.6%   2821   42.2%
  09. Texel 1.04 x64                   2851   11   11  2100    54.6%   2819   39.7%
  --. Protector 1.5.0 JA x64           2845   17   17  1000    56.5%   2797   45.7%
  10. Senpai 1.0 SSE42 x64             2831   11   11  2400    51.8%   2822   39.7%
  11. Hiarcs 14 WCSC w32               2825   10   10  2400    51.0%   2822   42.8%
  12. Shredder 12 x64                  2793   12   12  2100    45.4%   2835   40.0%
  --. Texel 1.03 x64                   2788   18   18  1000    48.7%   2800   42.8%
  13. Junior 12.5.03 x64               2779   15   15  1350    42.9%   2841   38.3%
  --. Junior 13.3.00 x64               2776   14   14  1300    44.5%   2822   39.8%
  --. Junior 13.8.04 Yokohama x64      2775   14   14  1650    43.4%   2833   40.8%
  14. Spike 1.4 Leiden w32             2773   14   14  1550    42.3%   2839   40.6%
  15. DiscoCheck 5.2.1 x64             2761   16   16  1200    39.8%   2848   36.4%
  16. iCE 2.0 v2240 POP x64            2757   11   11  2600    50.5%   2758   42.9%
  17. Quazar 0.4 x64                   2753   12   12  2100    40.4%   2836   40.9%
  18. SmarThink 1.70 SSE3 x64          2751   12   12  1950    41.7%   2822   35.5%
  19. Spark 1.0 x64                    2749    9    9  3400    47.9%   2772   39.9%
  20. Deuterium 14.3.34.130 POP x64    2748   28   28   350    45.9%   2786   43.1%  NEW, + 29 Elo
  21. Zappa Mexico II x64              2745   14   14  1650    38.2%   2846   40.8%
  --. SmarThink 1.60 x64               2741   17   17  1100    38.0%   2841   36.2%
  22. Fizbo 1.3.1 x64                  2737   28   28   350    44.4%   2787   38.0%  NEW, + 58 Elo
  23. Vajolet2 1.28 POP x64            2723   15   15  1450    36.4%   2837   38.7%
  --. Vajolet2 1.45 POP x64            2722   10   10  2900    45.4%   2764   41.1%
  24. Gaviota 1.0 AVX x64              2720    9    9  3400    44.1%   2773   36.5%
  --. Deuterium 14.2.33.276 x64        2719   10   10  3050    44.3%   2771   39.9%
  25. Andscacs 0.70 POP x64            2719   28   28   350    42.0%   2788   43.4%  NEW, +124 Elo
  26. Tornado 5.0 SSE4 x64             2714   11   11  2200    44.0%   2768   37.4%
  --. SmarThink 1.50 SSE3 x64          2700   17   17  1000    36.9%   2804   37.1%
  27. Nirvanachess 1.7 x64             2692   11   11  2450    42.1%   2761   38.8%
  --. Nirvanachess 1.8 x64             2682   30   30   300    41.7%   2745   38.0%  NEW, - 10 Elo
  --. Fizbo 1.2 x64                    2679   12   12  2100    41.8%   2749   39.0%
  --. Nirvanachess 1.6 x64             2674   16   16  1350    29.6%   2848   33.6%
  28. Arasan 17.4 POP x64              2672   14   14  1350    48.6%   2687   41.4%
  29. Rodent 1.6 Build 6 POP x64       2665   30   30   350    35.1%   2790   36.6%  NEW, + 37 Elo (version 1.5 not tested)
  --. Tornado 6.0 SSE x64              2664   31   31   300    39.2%   2746   40.3%  NEW, - 50 Elo
  30. Cheng4 0.36c x64                 2664   15   15  1350    47.3%   2687   41.5%
  31. Crafty 24.1 SSE42 x64            2645   13   13  1350    44.7%   2688   35.7%
  32. EXchess 7.31b x64                2641   14   14  1350    44.2%   2688   39.3%
  --. Crafty 24.0 SSE42 x64            2632   19   19  1000    28.4%   2808   34.4%
  33. Glaurung 2.2 JA x64              2632   17   17  1000    47.1%   2652   40.9%
  --. Rodent 1.4 POP Build 2 x64       2628   16   16  1000    46.6%   2653   40.4%
  34. Atlas 3.70em x64                 2627   17   17  1050    44.6%   2672   38.7%
  35. OctoChess r5190 SSE4 x64         2623   16   16  1000    45.9%   2653   45.5%
  36. Rhetoric 1.4.1 x64               2598   13   13  2000    31.9%   2751   35.8%
  --. Andscacs 0.64 POP x64            2595   17   17  1000    41.9%   2654   39.8%
  37. Godel 3.4.9 x64                  2584   17   17  1000    40.4%   2655   39.3%
  38. Djinn 1.021 POP x64              2514   18   18  1000    31.2%   2658   36.6%
  39. ProDeo 1.87 w32                  2508   18   18  1000    30.4%   2659   31.3%
  
	  
  
 Move average                : 174.64 / 87.32
 White advantage             : 39.39
 Draw rate (equal opponents) : 46.13%
 
 White Wins   :  16299 (35.4%)
 Black Wins   :  11869 (25.8%)
 Draws        :  17832 (38.8%)
 Unfinished   :      0

 White Perf.  :   54.8%
 Black Perf.  :   45.2%

 ECO A        =   8332 Games (18.1%)
 ECO B        =  10833 Games (23.6%)
 ECO C        =   9722 Games (21.1%)
 ECO D        =   9021 Games (19.6%)
 ECO E        =   8092 Games (17.6%)

[Vinvin]

Hi Frank, why did you write +105 Elo for "Fire 4 x64 " ? 3056-2951=+65 Elo

Hi there,

and here the actual FCT1 Rating List after 46.000 games:

Code: Select all

 "FCT1 Rating List"
 Will be updated after ~ 1000 new games!
 ---------------------------------------
  
 Date         : December 17th, 2014 (02:00)
 FCT1         : 46000 games
 Version      : 1.16.300 S1
 Hardware     : 2x Intel Core i7-4770K *4.3Ghz, 45-minutes games
 Engines      : Maximum allowed = three versions of same engine!
 
 Calculation  : Ordo 0.97, "Hiarcs 14 WCSC w32" with 2.825 Elo / News 057 & News 062
 Parameter    : Ordo_097.exe -a 2825 -A "Hiarcs 14 WCSC w32" -p 1.pgn -o rating.txt -W -s1000 -D -E -V
  
 
      Program                          Elo    +    -   Games   Score  Av.Op.  Draw	  
...
  03. Fire 4 x64                       3056   17   17  1600    80.2%   2785   29.2%  NEW, +105 Elo, runs as S1 (Special -01)
...
  --. Fire 3.0 AVX x64                 2951   13   13  2050    65.5%   2825   41.2%
 ...

[Frank Quisinsky]

3056 - 2951 = 105Elo
Very little mistake by yourself Vincent!

But I have an other problem ...
Have a look in the round robin results from Shrerdder GUI.
http://www.amateurschach.de/ftptrigger/ ... -01__.html
Compare 3036 GullChess 3 ELO (edit before round robin started) with 3027 Fire 4 ELO (after round robin).

And now have a look in the Rating list calculated with Ordo (message before).

In my Round-Robin table (calculated with Shredder GUI)
Gullchess 3 is 9 ELO better!

With Ordo 0.97 calculation Fire 4 is 18 Elo better!

Not easy to understand for the vistiors of my LIVE test!
FACT is ... Fire 4 is around 20 ELO stronger as Gull 3 and to if I am looking on available Fire 4 results in ultra fast games vs. a hand full opponents not to see. I believe that Fire 4 is clearly stronger with longer time controls. CCRL is testing Fire 4 too ... I am sure the results will be the same as my results. CCRL is to compare with FCT1 (time control).

Have a look here ...
Here is my Rating List calculated with EloStat 1.3 ... all is fine!
Different here +19 (calculated with Hiarcs = 2.825).
Different here between Fire 4 and Fire 3 = 95 Elo.

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Stockfish 03.08.14 BMI2 x64    : 3074   16  15  1600    82.7 %   2803   30.2 %
  2 Stockfish 02.10.14 BMI2 x64 C  : 3073   18  17  1100    80.0 %   2832   34.0 %
  3 Komodo 8 x64                   : 3072   16  16  1500    81.1 %   2819   30.5 %
  4 Stockfish 5 SSE42 x64          : 3053   18  18  1000    79.6 %   2816   35.4 %
  5 Komodo 7a x64                  : 3051   17  17  1250    79.6 %   2814   31.9 %
  6 Komodo TCECr x64               : 3032   19  19  1000    80.3 %   2788   32.2 %
  7 Fire 4 x64                     : 3028   16  16  1600    80.2 %   2785   29.2 %
  8 GullChess 3.0 BMI2 x64         : 3009   13  13  1800    74.3 %   2824   37.3 %
  9 GullChess 2.8 Beta BMI2 x64    : 2972   18  17  1000    74.0 %   2791   37.7 %
 10 Fire 3.0 AVX x64               : 2933   12  12  2050    65.5 %   2822   41.2 %
 11 Protector 1.7.0 x64            : 2895   15  15  1150    56.3 %   2850   44.0 %
 12 Chiron 2.0 x64                 : 2885   11  11  2400    59.8 %   2817   39.2 %
 13 Critter 0.90 SSE4 x64          : 2878   14  14  1350    56.6 %   2832   42.7 %
 14 Protector 1.6.0 x64            : 2875   13  13  1600    58.7 %   2814   44.4 %
 15 Hannibal 1.4b x64              : 2857   11  11  2400    55.6 %   2817   42.2 %
 16 Texel 1.04 x64                 : 2848   12  12  2100    54.6 %   2816   39.7 %
 17 Protector 1.5.0 JA x64         : 2843   16  16  1000    56.5 %   2797   45.7 %
 18 Senpai 1.0 SSE42 x64           : 2830   11  11  2400    51.8 %   2818   39.7 %
 19 Hiarcs 14 WCSC w32             : 2825   11  11  2400    51.0 %   2818   42.8 %
 20 Shredder 12 x64                : 2798   12  12  2100    45.4 %   2830   40.0 %
 21 Texel 1.03 x64                 : 2791   16  16  1000    48.7 %   2800   42.8 %
 22 Junior 12.5.03 x64             : 2785   15  15  1350    42.9 %   2835   38.3 %
 23 Junior 13.8.04 Yokohama x64    : 2783   13  13  1650    43.4 %   2829   40.8 %
 24 Spike 1.4 Leiden w32           : 2780   13  13  1550    42.3 %   2834   40.6 %
 25 Junior 13.3.00 x64             : 2780   15  15  1300    44.5 %   2818   39.8 %
 26 DiscoCheck 5.2.1 x64           : 2770   16  16  1200    39.8 %   2842   36.4 %
 27 Quazar 0.4 x64                 : 2763   11  12  2100    40.4 %   2831   40.9 %
 28 iCE 2.0 v2240 POP x64          : 2762   10  10  2600    50.5 %   2759   42.9 %
 29 SmarThink 1.70 SSE3 x64        : 2760   12  12  1950    41.7 %   2818   35.5 %
 30 Spark 1.0 x64                  : 2757    9   9  3400    47.9 %   2772   39.9 %
 31 Zappa Mexico II x64            : 2756   13  13  1650    38.2 %   2840   40.8 %
 32 Deuterium 14.3.34.130 POP x64  : 2756   27  28   350    45.9 %   2784   43.1 %
 33 SmarThink 1.60 x64             : 2751   17  17  1100    38.0 %   2836   36.2 %
 34 Fizbo 1.3.1 x64                : 2746   29  29   350    44.4 %   2785   38.0 %
 35 Vajolet2 1.28 POP x64          : 2736   14  14  1450    36.4 %   2833   38.7 %
 36 Vajolet2 1.45 POP x64          : 2732   10  10  2900    45.4 %   2764   41.1 %
 37 Gaviota 1.0 AVX x64            : 2732    9   9  3400    44.1 %   2773   36.5 %
 38 Deuterium 14.2.33.276 x64      : 2731   10  10  3050    44.3 %   2771   39.9 %
 39 Andscacs 0.70 POP x64          : 2729   27  28   350    42.0 %   2786   43.4 %
 40 Tornado 5.0 SSE4 x64           : 2725   12  12  2200    44.0 %   2767   37.4 %
 41 SmarThink 1.50 SSE3 x64        : 2710   17  17  1000    36.9 %   2804   37.1 %
 42 Nirvanachess 1.7 x64           : 2706   11  11  2450    42.1 %   2762   38.8 %
 43 Fizbo 1.2 x64                  : 2693   12  12  2100    41.8 %   2750   39.0 %
 44 Nirvanachess 1.6 x64           : 2691   16  16  1350    29.6 %   2842   33.6 %
 45 Nirvanachess 1.8 x64           : 2689   31  31   300    41.7 %   2748   38.0 %
 46 Arasan 17.4 POP x64            : 2683   14  14  1350    48.6 %   2693   41.4 %
 47 Rodent 1.6 Build 6 POP x64     : 2681   29  30   350    35.1 %   2788   36.6 %
 48 Cheng4 0.36c x64               : 2674   14  14  1350    47.3 %   2693   41.5 %
 49 Tornado 6.0 SSE x64            : 2672   31  31   300    39.2 %   2749   40.3 %
 50 Crafty 24.1 SSE42 x64          : 2657   15  15  1350    44.7 %   2694   35.7 %
 51 EXchess 7.31b x64              : 2654   14  15  1350    44.2 %   2694   39.3 %
 52 Crafty 24.0 SSE42 x64          : 2646   18  18  1000    28.4 %   2807   34.4 %
 53 Glaurung 2.2 JA x64            : 2642   17  17  1000    47.1 %   2661   40.9 %
 54 Atlas 3.70em x64               : 2641   16  17  1050    44.6 %   2679   38.7 %
 55 Rodent 1.4 POP Build 2 x64     : 2638   17  17  1000    46.6 %   2662   40.4 %
 56 OctoChess r5190 SSE4 x64       : 2633   16  16  1000    45.9 %   2662   45.5 %
 57 Rhetoric 1.4.1 x64             : 2620   13  13  2000    31.9 %   2752   35.8 %
 58 Andscacs 0.64 POP x64          : 2606   17  17  1000    41.9 %   2663   39.8 %
 59 Godel 3.4.9 x64                : 2596   17  17  1000    40.4 %   2664   39.3 %
 60 Djinn 1.021 POP x64            : 2530   18  18  1000    31.2 %   2667   36.6 %
 61 ProDeo 1.87 w32                : 2523   19  19  1000    30.3 %   2667   31.3 %

[Ozymandias]

Regardless of the exact difference, this is a big step forward for Fire. I wonder what the similarity is, with Houdini, right now.

[Frank Quisinsky]

Hi there,

and here the calculation with Bayesian 0056. Hiarcs in all calculations with 2825 Elo.

Different Fire 3 to Fire 4:
Ordo 0.97 = 105 Elo
EloStat 1.3 = 95 Elo
Bayesian 0056 = 93 Elo

Different Fire 4 to GullChess 3
Ordo 0.97 = 18 Elo
EloStat 1.3 = 19 Elo
Bayesian 0056 = 19 Elo

Best
Frank

Code: Select all

Rank Name                            Elo    +    - games score oppo. draws 
   1 Komodo 8 x64                   3063   17   17  1500   81%  2822   31% 
   2 Stockfish 03.08.14 BMI2 x64    3058   17   17  1600   83%  2806   30% 
   3 Stockfish 02.10.14 BMI2 x64 C  3058   20   19  1100   80%  2835   34% 
   4 Komodo 7a x64                  3040   18   18  1250   80%  2816   32% 
   5 Stockfish 5 SSE42 x64          3034   20   20  1000   80%  2818   35% 
   6 Fire 4 x64                     3024   17   16  1600   80%  2790   29% 
   7 Komodo TCECr x64               3013   20   20  1000   80%  2791   32% 
   8 GullChess 3.0 BMI2 x64         3005   15   15  1800   74%  2826   37% 
   9 GullChess 2.8 Beta BMI2 x64    2961   19   19  1000   74%  2794   38% 
  10 Fire 3.0 AVX x64               2931   13   13  2050   66%  2824   41% 
  11 Protector 1.7.0 x64            2895   17   17  1150   56%  2852   44% 
  12 Chiron 2.0 x64                 2885   12   12  2400   60%  2819   39% 
  13 Critter 0.90 SSE4 x64          2879   16   16  1350   57%  2834   43% 
  14 Protector 1.6.0 x64            2873   14   14  1600   59%  2816   44% 
  15 Hannibal 1.4b x64              2856   12   12  2400   56%  2820   42% 
  16 Texel 1.04 x64                 2847   13   13  2100   55%  2819   40% 
  17 Protector 1.5.0 JA x64         2841   18   17  1000   57%  2799   46% 
  18 Senpai 1.0 SSE42 x64           2829   12   12  2400   52%  2821   40% 
  19 Hiarcs 14 WCSC w32             2825   12   12  2400   51%  2821   43% 
  20 Shredder 12 x64                2797   12   12  2100   45%  2831   40% 
  21 Texel 1.03 x64                 2792   18   18  1000   49%  2802   43% 
  22 Junior 12.5.03 x64             2785   16   16  1350   43%  2837   38% 
  23 Junior 13.3.00 x64             2783   16   16  1300   45%  2821   40% 
  24 Junior 13.8.04 Yokohama x64    2781   14   14  1650   43%  2830   41% 
  25 Spike 1.4 Leiden w32           2777   15   14  1550   42%  2835   41% 
  26 DiscoCheck 5.2.1 x64           2768   17   17  1200   40%  2844   36% 
  27 iCE 2.0 v2240 POP x64          2765   11   11  2600   50%  2766   43% 
  28 Quazar 0.4 x64                 2761   12   13  2100   40%  2832   41% 
  29 Deuterium 14.3.34.130 POP x64  2760   30   30   350   46%  2791   43% 
  30 SmarThink 1.70 SSE3 x64        2760   13   13  1950   42%  2821   36% 
  31 Spark 1.0 x64                  2759   10   10  3400   48%  2778   40% 
  32 Zappa Mexico II x64            2756   14   14  1650   38%  2841   41% 
  33 SmarThink 1.60 x64             2751   17   18  1100   38%  2837   36% 
  34 Fizbo 1.3.1 x64                2750   30   30   350   44%  2792   38% 
  35 Vajolet2 1.28 POP x64          2737   15   15  1450   36%  2834   39% 
  36 Andscacs 0.70 POP x64          2736   30   30   350   42%  2792   43% 
  37 Vajolet2 1.45 POP x64          2736   11   11  2900   45%  2771   41% 
  38 Gaviota 1.0 AVX x64            2734   10   10  3400   44%  2778   37% 
  39 Deuterium 14.2.33.276 x64      2733   11   11  3050   44%  2777   40% 
  40 Tornado 5.0 SSE4 x64           2728   12   12  2200   44%  2774   37% 
  41 SmarThink 1.50 SSE3 x64        2715   18   18  1000   37%  2806   37% 
  42 Nirvanachess 1.7 x64           2711   12   12  2450   42%  2769   39% 
  43 Nirvanachess 1.8 x64           2700   32   32   300   42%  2756   38% 
  44 Fizbo 1.2 x64                  2699   13   13  2100   42%  2758   39% 
  45 Arasan 17.4 POP x64            2693   15   15  1350   49%  2705   41% 
  46 Nirvanachess 1.6 x64           2692   16   17  1350   30%  2842   34% 
  47 Rodent 1.6 Build 6 POP x64     2688   31   31   350   35%  2794   37% 
  48 Tornado 6.0 SSE x64            2687   32   32   300   39%  2756   40% 
  49 Cheng4 0.36c x64               2685   15   15  1350   47%  2705   41% 
  50 Crafty 24.1 SSE42 x64          2669   16   16  1350   45%  2706   36% 
  51 EXchess 7.31b x64              2666   15   16  1350   44%  2706   39% 
  52 Crafty 24.0 SSE42 x64          2657   19   19  1000   28%  2809   34% 
  53 Glaurung 2.2 JA x64            2656   17   18  1000   47%  2675   41% 
  54 Rodent 1.4 POP Build 2 x64     2654   18   18  1000   47%  2675   40% 
  55 Atlas 3.70em x64               2651   17   18  1050   45%  2692   39% 
  56 OctoChess r5190 SSE4 x64       2650   17   17  1000   46%  2675   46% 
  57 Rhetoric 1.4.1 x64             2628   13   14  2000   32%  2760   36% 
  58 Andscacs 0.64 POP x64          2625   18   18  1000   42%  2677   40% 
  59 Godel 3.4.9 x64                2615   18   18  1000   40%  2677   39% 
  60 Djinn 1.021 POP x64            2554   18   18  1000   31%  2680   37% 
  61 ProDeo 1.87 w32                2543   19   19  1000   30%  2681   31% 

[Vinvin]

Oops, it seems that I need a calculator to do simple subtraction
But my calculation was biased because I remembered to have seen closer ratings on the live tournament page ...

3056 - 2951 = 105Elo
Very little mistake by yourself Vincent!

But I have an other problem ...
Have a look in the round robin results from Shrerdder GUI.
http://www.amateurschach.de/ftptrigger/ ... -01__.html
Compare 3036 GullChess 3 ELO (edit before round robin started) with 3027 Fire 4 ELO (after round robin).

And now have a look in the Rating list calculated with Ordo (message before).

In my Round-Robin table (calculated with Shredder GUI)
Gullchess 3 is 9 ELO better!

With Ordo 0.97 calculation Fire 4 is 18 Elo better!

Not easy to understand for the vistiors of my LIVE test!
FACT is ... Fire 4 is around 20 ELO stronger as Gull 3 and to if I am looking on available Fire 4 results in ultra fast games vs. a hand full opponents not to see. I believe that Fire 4 is clearly stronger with longer time controls. CCRL is testing Fire 4 too ... I am sure the results will be the same as my results. CCRL is to compare with FCT1 (time control).

Have a look here ...
Here is my Rating List calculated with EloStat 1.3 ... all is fine!
Different here +19 (calculated with Hiarcs = 2.825).
Different here between Fire 4 and Fire 3 = 95 Elo.

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Stockfish 03.08.14 BMI2 x64    : 3074   16  15  1600    82.7 %   2803   30.2 %
  2 Stockfish 02.10.14 BMI2 x64 C  : 3073   18  17  1100    80.0 %   2832   34.0 %
  3 Komodo 8 x64                   : 3072   16  16  1500    81.1 %   2819   30.5 %
  4 Stockfish 5 SSE42 x64          : 3053   18  18  1000    79.6 %   2816   35.4 %
  5 Komodo 7a x64                  : 3051   17  17  1250    79.6 %   2814   31.9 %
  6 Komodo TCECr x64               : 3032   19  19  1000    80.3 %   2788   32.2 %
  7 Fire 4 x64                     : 3028   16  16  1600    80.2 %   2785   29.2 %
  8 GullChess 3.0 BMI2 x64         : 3009   13  13  1800    74.3 %   2824   37.3 %
  9 GullChess 2.8 Beta BMI2 x64    : 2972   18  17  1000    74.0 %   2791   37.7 %
 10 Fire 3.0 AVX x64               : 2933   12  12  2050    65.5 %   2822   41.2 %
 11 Protector 1.7.0 x64            : 2895   15  15  1150    56.3 %   2850   44.0 %
 12 Chiron 2.0 x64                 : 2885   11  11  2400    59.8 %   2817   39.2 %
 13 Critter 0.90 SSE4 x64          : 2878   14  14  1350    56.6 %   2832   42.7 %
 14 Protector 1.6.0 x64            : 2875   13  13  1600    58.7 %   2814   44.4 %
 15 Hannibal 1.4b x64              : 2857   11  11  2400    55.6 %   2817   42.2 %
 16 Texel 1.04 x64                 : 2848   12  12  2100    54.6 %   2816   39.7 %
 17 Protector 1.5.0 JA x64         : 2843   16  16  1000    56.5 %   2797   45.7 %
 18 Senpai 1.0 SSE42 x64           : 2830   11  11  2400    51.8 %   2818   39.7 %
 19 Hiarcs 14 WCSC w32             : 2825   11  11  2400    51.0 %   2818   42.8 %
 20 Shredder 12 x64                : 2798   12  12  2100    45.4 %   2830   40.0 %
 21 Texel 1.03 x64                 : 2791   16  16  1000    48.7 %   2800   42.8 %
 22 Junior 12.5.03 x64             : 2785   15  15  1350    42.9 %   2835   38.3 %
 23 Junior 13.8.04 Yokohama x64    : 2783   13  13  1650    43.4 %   2829   40.8 %
 24 Spike 1.4 Leiden w32           : 2780   13  13  1550    42.3 %   2834   40.6 %
 25 Junior 13.3.00 x64             : 2780   15  15  1300    44.5 %   2818   39.8 %
 26 DiscoCheck 5.2.1 x64           : 2770   16  16  1200    39.8 %   2842   36.4 %
 27 Quazar 0.4 x64                 : 2763   11  12  2100    40.4 %   2831   40.9 %
 28 iCE 2.0 v2240 POP x64          : 2762   10  10  2600    50.5 %   2759   42.9 %
 29 SmarThink 1.70 SSE3 x64        : 2760   12  12  1950    41.7 %   2818   35.5 %
 30 Spark 1.0 x64                  : 2757    9   9  3400    47.9 %   2772   39.9 %
 31 Zappa Mexico II x64            : 2756   13  13  1650    38.2 %   2840   40.8 %
 32 Deuterium 14.3.34.130 POP x64  : 2756   27  28   350    45.9 %   2784   43.1 %
 33 SmarThink 1.60 x64             : 2751   17  17  1100    38.0 %   2836   36.2 %
 34 Fizbo 1.3.1 x64                : 2746   29  29   350    44.4 %   2785   38.0 %
 35 Vajolet2 1.28 POP x64          : 2736   14  14  1450    36.4 %   2833   38.7 %
 36 Vajolet2 1.45 POP x64          : 2732   10  10  2900    45.4 %   2764   41.1 %
 37 Gaviota 1.0 AVX x64            : 2732    9   9  3400    44.1 %   2773   36.5 %
 38 Deuterium 14.2.33.276 x64      : 2731   10  10  3050    44.3 %   2771   39.9 %
 39 Andscacs 0.70 POP x64          : 2729   27  28   350    42.0 %   2786   43.4 %
 40 Tornado 5.0 SSE4 x64           : 2725   12  12  2200    44.0 %   2767   37.4 %
 41 SmarThink 1.50 SSE3 x64        : 2710   17  17  1000    36.9 %   2804   37.1 %
 42 Nirvanachess 1.7 x64           : 2706   11  11  2450    42.1 %   2762   38.8 %
 43 Fizbo 1.2 x64                  : 2693   12  12  2100    41.8 %   2750   39.0 %
 44 Nirvanachess 1.6 x64           : 2691   16  16  1350    29.6 %   2842   33.6 %
 45 Nirvanachess 1.8 x64           : 2689   31  31   300    41.7 %   2748   38.0 %
 46 Arasan 17.4 POP x64            : 2683   14  14  1350    48.6 %   2693   41.4 %
 47 Rodent 1.6 Build 6 POP x64     : 2681   29  30   350    35.1 %   2788   36.6 %
 48 Cheng4 0.36c x64               : 2674   14  14  1350    47.3 %   2693   41.5 %
 49 Tornado 6.0 SSE x64            : 2672   31  31   300    39.2 %   2749   40.3 %
 50 Crafty 24.1 SSE42 x64          : 2657   15  15  1350    44.7 %   2694   35.7 %
 51 EXchess 7.31b x64              : 2654   14  15  1350    44.2 %   2694   39.3 %
 52 Crafty 24.0 SSE42 x64          : 2646   18  18  1000    28.4 %   2807   34.4 %
 53 Glaurung 2.2 JA x64            : 2642   17  17  1000    47.1 %   2661   40.9 %
 54 Atlas 3.70em x64               : 2641   16  17  1050    44.6 %   2679   38.7 %
 55 Rodent 1.4 POP Build 2 x64     : 2638   17  17  1000    46.6 %   2662   40.4 %
 56 OctoChess r5190 SSE4 x64       : 2633   16  16  1000    45.9 %   2662   45.5 %
 57 Rhetoric 1.4.1 x64             : 2620   13  13  2000    31.9 %   2752   35.8 %
 58 Andscacs 0.64 POP x64          : 2606   17  17  1000    41.9 %   2663   39.8 %
 59 Godel 3.4.9 x64                : 2596   17  17  1000    40.4 %   2664   39.3 %
 60 Djinn 1.021 POP x64            : 2530   18  18  1000    31.2 %   2667   36.6 %
 61 ProDeo 1.87 w32                : 2523   19  19  1000    30.3 %   2667   31.3 %

[Frank Quisinsky]

Hi Vincent,

yes ... that is what I mean!
To follow the ratings in Shredder GUI output isn't a good way. So I created an extra page for my special test-runs ... and here I calculated with Ordo from time to time (better to follow the ratings here).

Have fun with FCT1.

Best
Frank

[Frank Quisinsky]

Hi Juan,

the first Houdini version is to 99% the same as one of the first Robbolite versions by Norman. Different People find it out (example: the programmer of Thinker and others). As non programmer I have to respect it, to many proofs are available about it, interesting are also information the Critter programmer gave us. Such an engine isn't interesting for myself. If I remember on the first messages by Robert Houdart a big lie. Such programmer I don't like ...

I can't understand that commercials used such a work. That is a bitchslap to the group of chess programmers and computer chess fans in my view.

I like it to test the work by programmers with can do really a good work in programming. I will not know which of others codes the Houdini programmer is using if the first version of him used 99% of the code Norman released.

Thats my opinion to that topic!

Best
Frank

[Ozymandias]

I wasn't talking about the state of things at the beginning. I'm interested in knowing what are the differences, similarities, between Fire 4 and Houdini 4. Being aware of the common origin, I wonder whether they are close both in ELO as well as style. A similarity test would be a quick way of starting to find out. If they're similar, it's unlikely they'll play differently. If they aren't, then you have to actually look at the games.

[Frank Quisinsky]

Hi Juan,

a lot of difference in the style ...

1. Houdini is stronger in tactics after the opening book moves. Stronger in tactics in the late middlegame but often with more holes ...

2. Fire is stronger in endgames after all what I saw if I am looking in the games and stats I made. Very dynamic style in the very late middlegame / start endgame. Good to see here that Fire often find after 10 seconds better moves and switched often to the better move.

For me absolutely clear, that Fire will get a boost with more time. Houdini lost power with more time (better for fast time controls).

3. Fire is improved in more dynamic. The new Version search a way to win the games. Often I am thinking that the older Fire 3 is happy with a fast draw in the middlegame. Draw Quote is clearly lesser with Fire 4. After the openings not to beat but is playing more safty as risky.

We are speaking about two complete other engine in playing style. I believe that Fire will be stronger as Houdini with longer time controls or have here the same level. It should be clear that this Version of Fire will be stronger and stronger with more time.

If Norman can make the late middlegame better (here Fire 4 have different holes, good to see in the lost games I have) this Version will be on the same level as Stockfish and Komodo.

Games are available. Copy the lost games in a database and have a look in the late middlegame of Fire. Often I am thinking all is perfect and often ... what do Fire 4 here.

Again, after the opening and in the endgames all is perfect ... really very very strong.

Nice work by Norman ...
He is alone on the way to place 1 and around 100 Elo more is a fantastic work. But this was clear with all his knowledge that he is able to do that.

Best
Frank

First impressions.
I will make more stats about it.
At the Moment I have not so many stats, only my impressions in visiting much of the 1.600 games I have.