+GGT+ Deep Rybka 3 wins Sokolsky Gambit + Firebird 1B leads+

[rainhaus]

+GGT+ Deep Rybka 3 wins Sokolsky Gambit + Firebird 1b leads +

AOO Sokolsky gambit

1.b4 e5 2.Bb2 f6 3.e4
The 3rd and last A00-gambit in this tournament

Wikipedia(cut citation):
Alexey Pavlovich Sokolsky (1908-1969) was a Ukrainian-Belarusian chess player of International Master strength and an opening theoretician. He was the first Soviet Correspondence Chess Champion (1948-51).
The name of Sokolsky is known now mostly due to his opening research and development of the chess opening 1.b4 which became known as Sokolsky Opening. It is also known as the Polish Opening, or the Orangutan Opening, the name Savielly Tartakower gave it in 1924.
Sokolsky wrote over a dozen books (The Modern Openings in Theory and Practice, Debyut 1 b2-b4, a book about his eponymous opening.

Contrary to Amar and Ware you can find about one thousand games with the Sokolsky gambit in the internet databases. Even world class players like Richard Réti, Boris Spassky, Vasily Smyslov, Tartakower and Géza Maroczy had played it. Today, negative report. However, except for the Queen's Gambit only a few top-class players trust in a gambit at all.


Win%

Code: Select all

Games        :     42 (finished)

White Wins   :      6 (14.3 %)
Black Wins   :     21 (50.0 %)
Draws        :     15 (35.7 %)
Unfinished   :      0

White Perf.  : 32.1 %
Black Perf.  : 67.9 % 

Like Amar and Ware a distinct negative balance for the third and last A00-gambit.

Sokolsky Gambit, Human Chess
Internet Database

Code: Select all

                   Games     1-0     =-=     0-1    Score
-----------------------------------------------------------
 All report games    963     448     137     378    53.6% 

-----------------------------------------------------------
Off course, the human balance is much better. Who plays it, is mostly well prepared and knows the theory.

A rare event: in this round one engine (Zappa) couldn't win one white game.


Starting evaluation

After 3.e4: 0,03 to 0,45 cp

..is again a bad indicator for that problematic opening. The long-term effects lay outside the computable scope.

Opening Table Sokolsky Gambit(SCID)

Code: Select all

---------------------------------------------------------------------
  +6 =15 -21 (13.5/42: 32%)
-------------------------------------------------------------------------------
       1        2        3        4        5        6        7        8     
-------------------------------------------------------------------------------
 1  b4       Lb2      e4       a3[1]    Sc3[2]   Le2      Sa4      Kf1       5:
    e5       f6       Lb4      Lc5      c6[3]    Db6      Lf2      Da5[4]   40%

 2  ...      ...      ...      ...      Lc4      Dh5      Df3      Se2       7:
    ...      ...      ...      La5      c6[5]    g6       Sh6[6]   d6[7]    21%

 3  ...      ...      ...      Lc4      Dh5[8]   Df3      Se2[9]   OO        5:
    ...      ...      ...      Se7      g6       Tf8      d6[10]   f5[11]   40%

 4  ...      ...      ...      ...      Dh5      Df3[12]  Se2      OO        5:
    ...      ...      ...      c6       g6       Sh6[13]  De7[14]  Sf7[15]  50%

 5  ...      ...      ...      ...      c3[16]   d4       Lg8      de5[17]   5:
    ...      ...      ...      Sc6      Le7      Sa5      Tg8      Kf8[18]  30%

 6  ...      ...      ...      ...      a3       Sf3      OO[20]   d4[21]    5:
    ...      ...      ...      ...      Lc5      Sh6[19]  Sf7      ed4[22]  50%

 7  ...      ...      ...      ...      ...      Sc3[23]  Sf3      ef5       6:
    ...      ...      ...      ...      La5      d6[24]   f5[25]   Lf5[26]   8%

 8  ...      ...      ...      ...      ...      Df3      Se2      g4[28]    4:
    ...      ...      ...      ...      ...      d6[27]   f5       Dh4[29]  25%

-------------------------------------------------------------------------------
 

Scores
Ranking Crosstable, Sokolsky

Code: Select all

GGTJan10_4221_10+12_A00sokolsky  2010

                                  1  2  3  4  5  6  7  
1   Rybka 3                       ** ½½ ½1 1½ 11 ½1 1½   9.0/12
2   Stockfish 1.6 JA 64bit_dec09  ½½ ** 10 01 1½ 0½ ½1   6.5/12  37.50
3   FireBird 1.0 beta x64 mp      ½0 01 ** 0½ 01 1½ 11   6.5/12  34.00
4   Deep Fritz 11                 0½ 10 1½ ** ½0 01 1½   6.0/12
5   Naum 4 MPx64                  00 0½ 10 ½1 ** 01 ½1   5.5/12
6   Deep Shredder 12 x64 FrGui    ½0 1½ 0½ 10 10 ** ½0   5.0/12
7   Zappa Mexico II               0½ ½0 00 0½ ½0 ½1 **   3.5/12
(42 games)

--------------------------------------------------------

Ranking Crosstable: 3 gambits, Amar+Ware+Sokolsky
(FRITZ11-GUI)

Code: Select all

GGTJan10_4221_10+12_A00 2010

                                  1      2      3      4      5      6      7      
1   FireBird 1.0 beta x64 mp      ****** ½1½½½0 11111½ 10½½01 11½½01 11½10½ 1½1111  25.0/36
2   Rybka 3                       ½0½½½1 ****** 01½1½1 1½1011 1010½½ 11011½ ½½011½  22.5/36
3   Deep Shredder 12 x64 FrGui    00000½ 10½0½0 ****** 110110 111½1½ 10½½10 1½10½0  17.5/36
4   Naum 4 MPx64                  01½½10 0½0100 001001 ****** 01010½ 1½½½½1 0½1½½1  16.5/36
5   Stockfish 1.6 JA 64bit_dec09  00½½10 0101½½ 000½0½ 10101½ ****** 1½0½01 0½½1½1  16.0/36
6   Deep Fritz 11                 00½01½ 00100½ 01½½01 0½½½½0 0½1½10 ****** 010½1½  14.5/36
7   Zappa Mexico II               0½0000 ½½100½ 0½01½1 1½0½½0 1½½0½0 101½0½ ******  14.0/36
(126 games)

Here is the pgn-link for the interested folk:

http://www.file-upload.net/download-222 ... y.pgn.html

Next: Swiss Gambit

Test conditions
Starting positions: pool of 100 gambits sorted by Elo number.
Time Control: short tournament level with 40/20', 20/10', 10'+12''.
System: Intel Core i7 920, oc 3700 MHz, 6 GB RAM. Hyperthreading off, Turbo Mode off. Vista 64Bit
Parameter: 3 Threads, ponder off, 1.2 MB Hash, EGTB 3-4-5 (not used by Firebird and Stockfish, but sometimes used by the GUI, even for both engines)
Fritz11-GUI: remis never, resign late.

Rainer Neuhäusler

[govert]

Thankyou! Keep them coming

P.S. You only use 1.2 MB of Hash?

[kingliveson]

This is good info...it would be nice if they all are kept withing a single thread to make it easier to follow.

http://www.talkchess.com/forum/viewtopic.php?t=32301
http://www.talkchess.com/forum/viewtopic.php?t=32218

Franklin

[rainhaus]

M. Helmer: P.S. You only use 1.2 MB of Hash?

Yes, it's gonna be close. I would say, just enough. On a longer time control or with 4 CPU I would take more. Firebird, Naum, Rybka don't display hash filling. Shredder is a slow filler, Stockfish fills in 3 minutes on 3 CPU, Zappa engine is the fastest, it fills 1024 MB within 90 sec. (Checked with Arena bcause Fritz11 doesn't show)

[rainhaus]

This is good info...it would be nice if they all are kept withing a single thread to make it easier to follow.
Franklin

It's very hard reading a forum site in the thread view mode, if long or over long threads are to scroll.
There is another possibility..., the headline method. Put "GGT" in the search box and you get all posts in one second : )

[govert]

M. Helmer: P.S. You only use 1.2 MB of Hash?

Yes, it's gonna be close. I would say, just enough. On a longer time control or with 4 CPU I would take more. Firebird, Naum, Rybka don't display hash filling. Shredder is a slow filler, Stockfish fills in 3 minutes on 3 CPU, Zappa engine is the fastest, it fills 1024 MB within 90 sec. (Checked with Arena bcause Fritz11 doesn't show)

I still think you mean 1.2 GB, not 1.2MB

[rainhaus]

M. Helmer: P.S. You only use 1.2 MB of Hash?

Yes, it's gonna be close. I would say, just enough. On a longer time control or with 4 CPU I would take more. Firebird, Naum, Rybka don't display hash filling. Shredder is a slow filler, Stockfish fills in 3 minutes on 3 CPU, Zappa engine is the fastest, it fills 1024 MB within 90 sec. (Checked with Arena bcause Fritz11 doesn't show)

I still think you mean 1.2 GB, not 1.2MB

Exact, I must fix this typo in the test conditions.THX. But as you see, even with 1,2 GB it's a relevant question of you