SCCT Swiss (Own Books) - Breaking News!

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

SCCT Swiss (Own Books) - Breaking News!

Post by Sedat Canbaz »

Dear Chess Friends,

For right now Houdini 4 takes the lead, my congratulations to Robert Houdart!

As I mentioned before,
All participants are playing with their own opening books, exception:
Gull
Houdini
Protector
Sjeng

I set Perfect 13.ctg (learning function disabled, optimized 3/35) for the above mentioned top chess engines

Note also that I disabled some openings in Perfect 13 book
*Due to the below openings are critical, not recommended for serious tournaments:
-For Whites: B22 Sicilian opening is disabled
-For Blacks: C42 Petroff defence is disabled
-For Blacks: B12/B19 Caro-Kann openings are disabled
-For Blacks: French opening is disabled

Honestly, (with this current small book optimization),
I did not expect to see good results, because is too old dated, plus I did not import any single game...
I've spent no more than 5 minutes during optimization over the mentioned openings
But as we see Houdini 4 + Perfect 13.ctg is doing not bad so far...

Let's see what will the final standings



Image
Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

Re: SCCT Swiss (Own Books) - Breaking News!

Post by Sedat Canbaz »

New Update:

Currently we have a new leader: Stockfish (it is not a surprise :))

Btw, I noticed Hiarcs's performance is very good so far
It seems Hiarcs's own opening book is very strong!

Latest standings:

Image
Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

Re: SCCT Swiss (Own Books) - Breaking News!

Post by Sedat Canbaz »

Hour by hour the competition is going much more exiting and interesting!

For right now we have two leaders (equal points and same average):Stockfish and Houdini !

Still is not clear which participant will be the winner of this tour...

Frankly,
I am quite satisfied by the performance of the engines which are using Perfect 13 book so far, currently all of them are in Top 10!

And here is the current standings after 78 games (games yet to play:12)

Image
Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

Re: SCCT Swiss (Own Books):Champion - Houdini 4

Post by Sedat Canbaz »

Flash Flash Flash: Houdini 4 is the New Champion of Own Books!!

Well done to Robert Houdart!

Note:Houdini 4 played with Perfect 13 book

And here is current standings (last games are playing, soon all games will be available)
Image

Feel the difference in SCCT and in using Perfect books!

Best,
Sedat
Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

Re: SCCT Swiss (Own Books):Champion - Houdini 4

Post by Sedat Canbaz »

Dear Friends,

SCCT Swiss Own books is over too!

For Full Standings:
http://www.sedatcanbaz.com/chess/?page_id=1166

Games:
http://www.mediafire.com/download/074vc ... s_2014.rar

Btw, here is the overall performance (based on games by Neutral/Without/Own books)


So, SCCT Champion and POLL winner is Stockfish :), actually mostly of the Voters were for Stockfish too!!

Code: Select all

   # PLAYER                    : RATING    POINTS  PLAYED    (%)
   1 Stockfish 15/260214       : 3469.5      20.5      27   75.9%
   2 Houdini 4 x64 6c          : 3426.5      18.5      27   68.5%
   3 Gull 2.8b x64 6c          : 3404.0      11.5      18   63.9%
   4 Komodo TCECr x64 6c       : 3389.4      17.5      27   64.8%
   5 Rybka 4.1 x64 6c          : 3297.3      15.0      27   55.6%
   6 Protector 1.6.0 x64 6c    : 3296.3      16.0      27   59.3%
   7 Spike 1.4 x64 6c          : 3238.6      13.5      27   50.0%
   8 Hiarcs 14 WCSC w32 6c     : 3237.6      14.0      27   51.9%
   9 Chiron 2 x64 6c           : 3232.7      14.5      27   53.7%
  10 Hannibal 1.4b x64 6c      : 3218.3      13.5      27   50.0%
  11 Junior 13.3 x64 6c        : 3186.1      13.5      27   50.0%
  12 Zappa Mexico II x64 6c    : 3181.4      13.5      27   50.0%
  13 Texel 1.03 x64 6c         : 3172.3      13.5      27   50.0%
  14 Sjeng WC2008 x64 6c       : 3152.2      12.5      27   46.3%
  15 Spark 1.0 x64 6c          : 3146.6      13.0      27   48.1%
  16 Shredder 12 x64 6c        : 3117.1      11.5      27   42.6%
  17 Cheng 4.0.36a x64 6c      : 3066.1       3.5       9   38.9%
  18 Tornado 4.88 x64 6c       : 3029.8       9.5      27   35.2%
  19 Gaviota 0.86 x64 6c       : 3009.4       9.0      27   33.3%
  20 Fritz 10 w32 6c           : 3000.0       9.5      27   35.2%
  21 Crafty 23.8 x64 6c        : 2914.2       6.5      27   24.1%

Have fun,
Sedat
Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

Re: SCCT Swiss (Own Books):Champion - Houdini 4

Post by Sedat Canbaz »

And once more I'd like to congratulate Top 3 standings:
Stockfish team
Robert Houdart
Vadim Demichev

Best Regards,
Sedat
Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

Re: SCCT Swiss (Own Books):Champion - Houdini 4

Post by Sedat Canbaz »

Btw,
As I mentioned before: I strongly believe that 'Stockfish' is the strongest engine in the world, plus free!

And here is the overall comparing (almost equal in points between Commercial vs Freeware engines):
Image
jdart
Posts: 4410
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

Re: SCCT Swiss (Own Books):Champion - Houdini 4

Post by jdart »

Surprisingly quite a few of these games have sub-optimal play in the opening. For example:

[pgn]
[Event "SCCT"]
[Site "?"]
[Date "2014.03.04"]
[Round "7"]
[White "Stockfish 260214 x64 6c"]
[Black "Rybka 4.1 x64 6c"]
[Result "1-0"]
[ECO "B82"]
[Annotator "0.31;0.39"]
[PlyCount "165"]
[EventDate "2014.03.03"]
[Source "Doe"]
[TimeControl "2700+15"]

{Intel(R) Core(TM) i7 CPU 980 @ 3.33GHz 3337 MHz W=36.7 plies; 10.394kN/s;
OwnBook.ctg B=21.5 plies; 480kN/s; 1.147 TBAs; Rybka4.ctg} 1. e4 {0.01/0 0} c5
{0.00/0 0} 2. Nf3 {0.01/0 0} d6 {0.00/0 0} 3. d4 {0.01/0 0} cxd4 {0.00/0 0} 4.
Nxd4 {0.01/0 0} Nf6 {0.00/0 0} 5. Nc3 {0.01/0 0} Nc6 {0.00/0 0} 6. Be3 {0.01/0
0} e6 {0.00/0 0} 7. f4 {0.01/0 0} e5 {0.00/0 0} 8. Nf3 {0.01/0 0} Be7 {0.00/0 0
} 9. Qd2 {0.31/30 142} O-O {0.00/0 0} 10. O-O-O {0.36/27 52} exf4 {0.39/18 134}
11. Bxf4 {0.28/31 147} Be6 {0.37/19 65} 12. Ng5 {0.28/31 38 (a3)} Bg4 {0.26/19
118} 13. Be2 {0.34/30 98} Re8 {0.25/19 203 (h6)} 14. Bxg4 {0.53/29 65 (Kb1)}
Nxg4 {0.29/17 23} 15. Kb1 {0.38/29 75} Nce5 {0.26/19 100} 16. h3 {0.46/31 39
(Nd5)} Nf6 {0.31/21 134} 17. Qf2 {0.43/29 57 (Nf3)} Qa5 {0.04/20 94 (Qc7)} 18.
Nd5 {0.05/28 80 (Qe2)} Nc4 {0.00/20 130 (Bd8)} 19. Bc1 {0.31/30 112} h6 {0.00/
20 27} 20. h4 {0.15/31 142} Bd8 {0.00/19 40} 21. Qf3 {0.24/32 330 (Qf5)} Rc8 {
0.00/18 129} 22. Rhf1 {0.08/32 58} Qb5 {0.00/19 110 (hxg5)} 23. b3 {0.37/31 56
(Nxf7)} Ne5 {0.26/16 34} 24. Qf5 {0.63/34 63} Qc6 {0.42/17 54 (Qc5)} 25. Nxf6+
{0.04/31 68} Bxf6 {0.42/18 18} 26. Qh7+ {0.51/32 34} Kf8 {0.42/8 0} 27. c4 {0.
04/33 52} Nxc4 {-300.02/19 56 (hxg5)} 28. Rxf6 {1.20/35 53} Na3+ {0.37/19 9}
29. Kb2 {1.12/36 47} Qc3+ {0.37/19 18} 30. Kxa3 {1.12/1 0} Qxf6 {0.37/19 15}
31. Nf3 {1.12/37 35} Qg6 {0.48/20 46} 32. Qxg6 {1.40/34 58} fxg6 {0.48/21 9}
33. Re1 {1.30/34 64} Rc2 {0.49/20 15} 34. Bf4 {1.25/36 35 (Bd2)} Ke7 {0.62/22
226} 35. g3 {1.33/35 35 (Rd1)} Rc3 {0.54/23 317} 36. Nd2 {1.37/35 72 (Nd4)}
Rec8 {0.51/23 85 (Rd3)} 37. Kb2 {1.44/36 34} Rd3 {0.52/23 70} 38. Nc4 {1.33/35
54} b5 {0.52/24 41} 39. Na3 {1.41/36 33} b4 {0.66/23 36} 40. Nc4 {1.50/34 33}
Ke6 {0.64/22 53} 41. Re2 {1.47/33 66 (Kc2)} Rf8 {0.69/23 68} 42. Ne3 {1.56/35
32 (Kc2)} Rc8 {0.72/23 122 (a6)} 43. Kb1 {1.56/34 43} Kd7 {0.77/23 146 (a6)}
44. Nc4 {1.46/33 42} Re8 {0.77/24 58} 45. Kc2 {1.26/34 48} Rc3+ {0.77/25 55}
46. Kb2 {1.34/34 30 (Kd2)} Rd3 {0.70/22 29} 47. e5 {1.52/33 70 (Kc2)} Re6 {0.
70/22 36} 48. Kc2 {1.54/34 88} Rc3+ {0.70/23 44} 49. Kd1 {1.56/36 120 (Kb1)}
dxe5 {0.71/21 39} 50. Rxe5 {1.56/35 34} Rd3+ {0.69/22 28} 51. Ke2 {1.50/35 84}
Rd5 {0.74/23 24} 52. Kf3 {1.47/35 34} Rdxe5 {0.80/23 29} 53. Nxe5+ {1.51/35 32}
Ke7 {0.83/24 65 (Kd6)} 54. Nd3 {2.20/33 32} a5 {0.91/24 21} 55. Bc7 {2.06/35
38 (Ne5)} Ra6 {1.09/24 77} 56. Ke4 {2.22/34 21} h5 {1.00/24 24 (Ra8)} 57. Nc5 {
2.11/38 31 (Nb2)} Ra7 {1.17/26 32 (Ra8)} 58. Bf4 {2.39/37 21 (Bb8)} Ra8 {1.32/
26 38 (Kf7)} 59. Ke5 {2.95/38 55 (Kd5)} Kf7 {1.58/24 49 (Rf8)} 60. Ne4 {3.29/
37 48 (Nb7)} Kf8 {1.58/25 44 (Ke7)} 61. Nd6 {3.66/36 39} Ke7 {1.83/24 27 (a4)}
62. Kd5 {3.92/37 47} Kd7 {1.94/23 33 (a4)} 63. Nc4 {4.16/37 24} Rf8 {2.20/23 29
} 64. Kc5 {4.52/38 87} Re8 {2.24/24 23 (Rf5+)} 65. Kb5 {4.92/38 16} Re2 {2.29/
24 15} 66. Nd2 {5.17/38 35 (Kxa5)} g5 {2.35/23 47} 67. Bxg5 {5.17/39 15} Re5+ {
2.83/23 26 (Kd6)} 68. Kb6 {5.45/37 21} Rf5 {2.83/25 36 (Rd5)} 69. Bf4 {5.92/37
24 (Nc4)} Ke6 {3.30/22 39} 70. Nc4 {6.37/38 15} g6 {3.30/22 19} 71. Nxa5 {6.77/
39 15} g5 {3.81/21 16} 72. Bxg5 {9.61/44 24} Rf2 {3.90/25 17} 73. Kc5 {11.59/
44 15 (Kb5)} Rxa2 {4.04/22 10} 74. Kxb4 {14.95/45 15 (Nc4)} Kd6 {4.04/22 18
(Kd5)} 75. Kb5 {51.56/45 31} Kd5 {5.52/23 16 (Ra1)} 76. b4 {75.35/54 34 (Bf4)}
Rb2 {5.52/19 5 (Rg2)} 77. Bf4 {82.93/60 28} Rb1 {5.52/21 6 (Re2)} 78. Nc4 {105.
23/57 34} Rb3 {5.52/21 13 (Kd4)} 79. Ne3+ {136.29/57 89} Kd4 {5.52/20 14 (Ke6)}
80. Kc6 {139.15/55 51 (Nf5+)} Rc3+ {6.56/20 27} 81. Kb6 {139.35/51 28} Rd3 {13.
57/22 21} 82. b5 {136.29/46 17} Rb3 {14.42/22 13 (Kc3)} 83. Nf5+ {69.21/44 11}
1-0
[/pgn]

6. .. e6, while it transposes into a common Sicilian line, is not as good as alternatives such as .. e5 (Sicilian B56) or .. Ng4, IMO.

--Jon
Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 11:58 am
Location: Antalya/Turkey

Re: SCCT Swiss (Own Books):Champion - Houdini 4

Post by Sedat Canbaz »

Frankly (this morning, before to see your posting)
I was wondering why you still did not make any comment regarding the tour with own books :)
Because as far as I know you are BIG fun of such own book tournaments and now as we see... my estimation is right :)

Actually as main tournament we should concentrate to Neutral book competition
Due to the performance of some engines suffer with own/without books

For example,
Houdini won, but I think Perfect 13.ctg helped a lot...

About your published game,
Personally I would not prefer as Whites to be played to 6.Be3
Because in case of playing by Blacks to 6.Ng4, then Whites are falling in danger...

And here is a screenshot where Perfect 15.ctg (as Blacks) suggest to 6.Ng4
Image


Btw, any news about Arasan next release ?
Note also that a few days ago,I started a new blitz testing with top 20 mp engines
And I have no patience to test your next new Arasan release



Best Wishes,
Sedat
jdart
Posts: 4410
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

Re: SCCT Swiss (Own Books):Champion - Houdini 4

Post by jdart »

6. .. Ng4 is preferred in the latest Hiarcs book. The main line is probably 7. Bb5 Nxe3 8. fxe3 Bd7 9. O-O e6 which is fairly equal.
Btw, any news about Arasan next release ?
I am thinking about a new release soon. The next version will be 17.0 since I have made some fairly major code changes. There is some measurable improvement over version 16.3 in playing strength but nothing very dramatic. However, it will play differently from 16.3 overall. There is also a new and improved opening book.

I need to will do some testing but can probably release something in the next week or so if there is interest.

--Jon