SWCR: Houdini 2.0c w32 and Houdini 2.0c x64 LIVE

[Frank Quisinsky]

Code: Select all

Rank Name                            Elo    +    - games score oppo. draws 
   1 Houdini 2.0c x64               3004   35   33   407   81%  2754   25% 
   2 Houdini 1.5 x64                2998   14   14  2320   78%  2771   29% 
   3 Houdini 1.5 w32                2982   19   19  1197   79%  2754   29% 
   4 Rybka 4 x64 Exp. 42            2967   20   19  1200   80%  2729   26% 
   5 Rybka 4.1 x64 Exp. 79TD v.1    2962   20   19  1200   79%  2735   26% 
   6 Komodo 3.0 x64                 2960   17   17  1454   77%  2745   31% 
   7 Rybka 4 x64 Exp. 61            2959   21   21  1000   78%  2741   27% 
   8 Rybka 4.1 x64                  2955   15   15  1734   74%  2769   35% 
   9 Critter 1.2 x64                2952   16   16  1654   76%  2745   33% 
  10 Stockfish 2.1.1 JA x64 PHQ     2947   18   18  1280   76%  2745   32% 
  11 Houdini 1.03a x64              2943   21   21  1000   80%  2709   30% 
  12 Rybka 4 x64                    2939   18   17  1480   80%  2696   28% 
  13 IvanHoe B47cB x64              2939   15   15  1734   71%  2780   38% 
  14 Fire 2.2 xTreme x64            2938   16   16  1614   75%  2741   33% 
  15 Fire 1.5 xTreme x64            2935   17   16  1400   73%  2761   36% 
  16 Critter 1.2 w32                2932   18   18  1156   75%  2751   36% 
  17 IvanHoe B49jA x64              2931   19   18  1160   76%  2737   34% 
  18 Stockfish 2.1.1 JA x64         2928   15   15  1774   73%  2749   35% 
  19 Komodo 2.03 JA x64             2926   17   17  1480   73%  2748   31% 
  20 IvanHoe B52aC x64              2917   20   20  1000   76%  2725   34% 
  21 Robbolito 0.09 x64             2917   17   17  1480   74%  2729   33% 

= +87 to the first predecessor Robbolito 0.09 x64.
But I think the rating go higher with more games. Only in middle game tactic Houdini 2.0c seems to be a bit weeker as 1.5. In all other things after my statistics version 2.0c seems to be better. Need more games ...

[MM]

Code: Select all

Rank Name                            Elo    +    - games score oppo. draws 
   1 Houdini 2.0c x64               3004   35   33   407   81%  2754   25% 
   2 Houdini 1.5 x64                2998   14   14  2320   78%  2771   29% 
   3 Houdini 1.5 w32                2982   19   19  1197   79%  2754   29% 
   4 Rybka 4 x64 Exp. 42            2967   20   19  1200   80%  2729   26% 
   5 Rybka 4.1 x64 Exp. 79TD v.1    2962   20   19  1200   79%  2735   26% 
   6 Komodo 3.0 x64                 2960   17   17  1454   77%  2745   31% 
   7 Rybka 4 x64 Exp. 61            2959   21   21  1000   78%  2741   27% 
   8 Rybka 4.1 x64                  2955   15   15  1734   74%  2769   35% 
   9 Critter 1.2 x64                2952   16   16  1654   76%  2745   33% 
  10 Stockfish 2.1.1 JA x64 PHQ     2947   18   18  1280   76%  2745   32% 
  11 Houdini 1.03a x64              2943   21   21  1000   80%  2709   30% 
  12 Rybka 4 x64                    2939   18   17  1480   80%  2696   28% 
  13 IvanHoe B47cB x64              2939   15   15  1734   71%  2780   38% 
  14 Fire 2.2 xTreme x64            2938   16   16  1614   75%  2741   33% 
  15 Fire 1.5 xTreme x64            2935   17   16  1400   73%  2761   36% 
  16 Critter 1.2 w32                2932   18   18  1156   75%  2751   36% 
  17 IvanHoe B49jA x64              2931   19   18  1160   76%  2737   34% 
  18 Stockfish 2.1.1 JA x64         2928   15   15  1774   73%  2749   35% 
  19 Komodo 2.03 JA x64             2926   17   17  1480   73%  2748   31% 
  20 IvanHoe B52aC x64              2917   20   20  1000   76%  2725   34% 
  21 Robbolito 0.09 x64             2917   17   17  1480   74%  2729   33% 

= +87 to the first predecessor Robbolito 0.09 x64.
But I think the rating go higher with more games. Only in middle game tactic Houdini 2.0c seems to be a bit weeker as 1.5. In all other things after my statistics version 2.0c seems to be better. Need more games ...

Hi Frank, did you ever try with contempt zero? I think you could find some differences. I found Houdini more positional and solid (and stronger).

Regards

[Frank Quisinsky]

Hi Maurizio,

no I never try out contemp settings. Each engine should play with the standard settings. From time to time I test settings too.

Yes, I think Houdini 2.0c is stronger as version 1.5 and more solid. Middlegame tactics could be a small problem if I am looking in the lost games, but all this is hard to say. All other technical points are good. Time management with ponder = on, stability ... all is wonderful.

I believe ELO will go high with more games.
Perhaps 20 ELO more as 1.5 in SWCR, I don't know but this should be the result I await after the games I saw and analyzes I do with the games I have.

Houdini 2.0c have problems vs. BugChess2 1.9 in much of the games. Most Houdini won vs. BugChess late but lost 1 or 2 games vs. BugChess2. An engine with is over 400 ELO weaker. BugChess2 must have some good ideas included. Very nice engine!

Let us wait of the complete final result.

Best
Frank

[Frank Quisinsky]

Again, I think I can go higher ...

Code: Select all

Rank Name                            Elo    +    - games score oppo. draws 
   1 Houdini 2.0c x64               3018   30   29   570   82%  2750   23% 
   2 Houdini 1.5 x64                2998   14   14  2320   78%  2771   29% 
   3 Houdini 1.5 w32                2982   19   19  1197   79%  2754   29% 
   4 Rybka 4 x64 Exp. 42            2967   20   19  1200   80%  2729   26% 
   5 Rybka 4.1 x64 Exp. 79TD v.1    2962   20   19  1200   79%  2735   26% 
   6 Komodo 3.0 x64                 2960   17   17  1459   77%  2746   32% 
   7 Rybka 4 x64 Exp. 61            2959   21   21  1000   78%  2741   27% 
   8 Rybka 4.1 x64                  2955   15   15  1739   74%  2770   35% 
   9 Critter 1.2 x64                2952   16   16  1659   76%  2746   33% 
  10 Stockfish 2.1.1 JA x64 PHQ     2947   18   18  1280   76%  2745   32% 
  11 Houdini 1.03a x64              2943   21   21  1000   80%  2709   30% 
  12 Rybka 4 x64                    2939   18   17  1480   80%  2695   28% 
  13 IvanHoe B47cB x64              2939   15   15  1739   71%  2780   38% 
  14 Fire 2.2 xTreme x64            2938   16   16  1619   75%  2742   33% 
  15 Fire 1.5 xTreme x64            2935   17   16  1400   73%  2761   36% 
  16 Critter 1.2 w32                2932   18   18  1156   75%  2751   36% 
  17 IvanHoe B49jA x64              2931   19   18  1160   76%  2737   34% 
  18 Stockfish 2.1.1 JA x64         2929   15   15  1779   73%  2750   35% 
  19 Komodo 2.03 JA x64             2926   17   17  1480   73%  2748   31% 
  20 IvanHoe B52aC x64              2917   20   20  1000   76%  2725   34% 
  21 Robbolito 0.09 x64             2917   17   17  1480   74%  2729   33% 

[Frank Quisinsky]

Code: Select all

Rank Name                            Elo    +    - games score oppo. draws 
   1 Houdini 2.0c x64               3011   24   23   870   81%  2750   25%, 330 games to play
   2 Houdini 1.5 x64                2998   14   14  2320   78%  2771   29% 
  50 Fritz 13                       2817   17   17  1221   63%  2725   43%, 59 games to play 
  63 Deep Fritz 12 w32              2779   14   14  1760   62%  2698   41% 
  75 Fritz 12 w32                   2745   17   17  1160   59%  2687   44% 

Since today 00:00 the latest 163 games from the Fritz 13 round robin are running too. I think the final results from Houdini 2.0c and Fritz 13 are 1 1/2 days available.

Best
Frank

[MM]

Code: Select all

Rank Name                            Elo    +    - games score oppo. draws 
   1 Houdini 2.0c x64               3011   24   23   870   81%  2750   25%, 330 games to play
   2 Houdini 1.5 x64                2998   14   14  2320   78%  2771   29% 
  50 Fritz 13                       2817   17   17  1221   63%  2725   43%, 59 games to play 
  63 Deep Fritz 12 w32              2779   14   14  1760   62%  2698   41% 
  75 Fritz 12 w32                   2745   17   17  1160   59%  2687   44% 

Since today 00:00 the latest 163 games from the Fritz 13 round robin are running too. I think the final results from Houdini 2.0c and Fritz 13 are 1 1/2 days available.

Best
Frank

Thanks Frank

[Frank Quisinsky]

Hi there,

here are the final results from the round robins:
1. Fritz 13
2. Houdini 2.0c x64

Results can be found in the afternoon on my webpage too. Latest game vs. ChessTiger is still running ... need a bit but Houdini will win the latest game too.

Best
Frank

Code: Select all

   1 Houdini 2.0c x64               3022   21   20  1199   82%  2751   24% 
   2 Houdini 1.5 x64                2998   14   14  2320   78%  2771   29% 
   3 Houdini 1.5 w32                2983   19   19  1200   79%  2755   29% 
  11 Houdini 1.03a x64              2943   21   21  1000   80%  2710   30% 
  50 Fritz 13                       2819   16   16  1320   63%  2727   42% 
  63 Deep Fritz 12 w32              2780   14   14  1760   62%  2699   41% 
  76 Fritz 12 w32                   2746   17   17  1160   59%  2688   44% 

[Frank Quisinsky]

Hi Loonie,

I made now a database update.
150.248 SWCR 40-minute games with ponder = on can be found in my download area. Included are all Houdini 2.0c x64 games.

Have fun

Best
Frank