Happy New Year! Small Ratings List Updated...

[Erik Roggenburg]

Latest Update: Rybka 2.3.2a mp with enough games to have error bars +/-20. No surprise, but Rybka 2.3.2a mp is the leader of the list. I had to get 2.3.2a tested in anticipation of a comparison with Rybka 3.x.

Fritz 11 continues to score well. It is only 90 Elo behind Rybka 2.3.2a single CPU. This makes me wonder about what Deep Fritz 11 will bring. 50 Elo? 75 Elo? 25 Elo? In all honesty, I hope that Deep Fritz 11 will bring some updates to the GUI to make it work a little better. The last update caused havok with my UCI engines.

Code: Select all

    Program                          Elo    +   -   Games   Score   Av.Op.  Draws

  1 Rybka 2.3.2a mp 32-bit 2T      : 2835   20  20   780    70.8 %   2681   36.9 %
  2 Rybka 2.3.1 mp 32-bit 2T       : 2818   13  13  2230    72.8 %   2647   29.8 %
  3 Rybka 2.3.2a 32-bit            : 2817   18  18   900    70.9 %   2662   38.2 %
  4 Rybka 2.2 mp 32-bit 2T         : 2811   15  15  1580    72.7 %   2641   30.4 %
  5 Rybka 2.3 LK mp 32-bit 2T      : 2806   18  18  1140    73.1 %   2632   28.8 %
  6 Rybka 2.3 mp 32-bit 2T         : 2805   18  18  1140    73.0 %   2632   30.4 %
  7 Deep Shredder 11 UCI 2T        : 2729   17  17  1140    61.2 %   2650   30.6 %
  8 Fritz 11                       : 2727   16  16  1140    60.9 %   2650   37.5 %
  9 Zappa Mexico 2T                : 2725   16  16  1170    60.7 %   2649   36.9 %
 10 HIARCS 11.1 MP UCI 2T          : 2722   11  11  2330    59.8 %   2653   36.7 %
 11 Deep Fritz 10 2T               : 2720   12  12  2390    58.6 %   2659   31.2 %
 12 Loop 13.5.32 2T                : 2717   12  12  2220    59.7 %   2649   37.3 %
 13 LoopMP 12.32 2T                : 2698   13  13  1680    56.9 %   2650   36.5 %
 14 Naum 2.2 2T                    : 2698   16  16  1110    56.7 %   2651   39.7 %
 15 HIARCS 11 MP UCI 2T            : 2691   14  14  1480    58.1 %   2634   37.2 %
 16 Deep Shredder 10 UCI 2T        : 2668   12  12  2190    52.0 %   2654   30.7 %
 17 Naum 2.1 MP 2T                 : 2660   12  12  1980    49.8 %   2661   39.1 %
 18 HIARCS 11 UCI                  : 2651   14  14  1470    52.0 %   2637   35.4 %
 19 Fruit 2.3.1                    : 2645   16  16  1170    49.6 %   2648   35.2 %
 20 Toga II 1.3x4                  : 2638   17  17  1080    49.8 %   2639   36.4 %
 21 Toga II 1.2.1a                 : 2637   12  12  2160    47.4 %   2656   34.6 %
 22 Deep Junior 10.1 2T            : 2633   12  12  2390    45.9 %   2661   30.3 %
 23 Spike 1.2 Turin                : 2628   11  11  2310    45.7 %   2658   36.7 %
 24 Hiarcs X54 UCI                 : 2624   16  16  1140    51.2 %   2616   35.7 %
 25 Hiarcs X50 UCI                 : 2622   16  16  1140    51.0 %   2616   36.7 %
 26 Deep Sjeng 2.7 2T              : 2620   17  17  1140    44.9 %   2656   33.4 %
 27 Fritz 9                        : 2618   15  15  1440    48.4 %   2629   29.8 %
 28 Fruit 2.2.1                    : 2612   12  12  2160    43.7 %   2656   33.4 %
 29 Glaurung 2.0.1 2T              : 2608   16  16  1170    41.3 %   2669   36.1 %
 30 Shredder 7.04 2T               : 2601   26  26   500    44.7 %   2638   29.4 %
 31 Glaurung 1.2 SMP 2T            : 2581   12  12  2280    39.6 %   2654   29.0 %
 32 Scorpio 2.0 2T                 : 2570   17  18  1080    38.2 %   2653   31.0 %
 33 Fritz 8 Bilbao                 : 2565   26  27   510    37.5 %   2654   25.9 %
 34 Chess Tiger 2007.1 UCI         : 2560   12  12  2330    36.2 %   2659   32.9 %
 35 Naum 2.0                       : 2543   16  16  1200    38.9 %   2622   36.2 %
 36 Scorpio 1.91 2T                : 2538   13  13  1920    32.1 %   2668   31.8 %
 37 Deep Pharaon 3.5.1 2T          : 2520   12  12  2280    31.5 %   2655   29.5 %
 38 Chess Tiger 15.0               : 2520   17  17  1140    35.8 %   2621   34.5 %
 39 Scorpio 1.8 2T                 : 2498   16  16  1380    31.1 %   2636   30.0 %
 40 Deep Frenzee 3.0 2T            : 2495   13  13  2270    28.3 %   2656   24.8 %

[Yarget]

Thanks for your great ratinglist Erik! I am looking forward to see Rybka 3.0, DF 11 and the upcoming Naum in your ratinglist.

Regards
Per

[Graham Banks]

Thanks for your great ratinglist Erik! I am looking forward to see Rybka 3.0, DF 11 and the upcoming Naum in your ratinglist.

Regards
Per

Yes - Erik has been sensible in organising a systematic approach. It's hard to maintain a useful and meaningful rating list when one's cpu time is restricted.

Regards, Graham.

[Erik Roggenburg]

I'm definitely going to include the latest Rybka, Naum, Deep Fritz, and HIARCS when they are released.

Right now, I'm testing Alaric 707 and I've made a small change to the methodology. Instead of sticking with a test set, I've decided to use Sedat's Perfect13.ctg. I wanted to see more of a variety in the openings, because to be honest, I use the games as a little bit of a reference for my own studies.

I have wanted to use a neutral book for a while, and when Sedat released his latest, it just seemed natural. If, for some reason, my Fritz 11 GUI goes bonkers, I'll fall back on Arena and use one of Harry Schnapp's books. His books have a wide, WIDE variety of openings, and Harry's books also seem to have lines that arrive at particular openings through a variety of move orders. I don't know if that is coincidental, by design, or my imagination!