Castling long

[Lyudmil Tsvetkov]

To post just another small message to cool my necessities in the heat.

Two of the best chess players in history, Kasparov and Alekhine, used to castle long much more frequently than other players: by Kasparov the percentage of games with long castling might be somewhere 15-20, while I suppose (without strict measurements done) that Alekhine would even reach some 30%, an unprecedented score. Other prominent and less prominent chess players would castle long much less frequently, unless, of course, they are inveterate white Sicilian players. I would also notice that, in general, engines would prefer castling short in the great variety of cases, much unlike Kasparov and Alekhine.

I would ask myself what could the reason for such a behaviour be? Were Kasparov and Alekhine weak players, or maybe they had bigger imagination than others? Are engines always correct when they choose to castle short instead of allowing themselves an option of long castling, or maybe they are missing some valuable unexplored alternatives? It is obvious that castling long would immediately expose the king to bigger danger, but maybe there might be some trade-offs for that in the sense of good activity in the center and good attacking chances on the king side.

For the time being I am very much undecided on this issue. I could suppose, though, that most engines, just like most humans, do not make the most of castling long. Maybe they have some larger than necessary penalties for long castling, or maybe lacking a bit in imagination Or maybe the engine side is right after all, with deeper search... Difficult to say, I would intuitively go for the first option. Interesting to know your opinion on the matter, does anyone by chance have some data (when this is possible to collect, for example, with very short books and castling still available) on which engines would exhibit a higher castling long rate than the average, and if those engines would be among the leading pack?

Below is a game of Alekhine on the theme.

[pgn][Site "Mexico City (08)"]
[Date "1932.??.??"]
[White "Alekhine Alexander"]
[Black "Brunner J"]
[Result "1-0"]
[ECO "D36"]
[MLNrOfMoves "25"]
[MLFlags "100000"]

1. d4 Nf6 2. c4 e6 3. Nc3 d5 4. Bg5 Nbd7 5. cxd5 exd5 6. e3 c6 7. Bd3 Be7 8. Qc2
O-O 9. Nge2 h6 10. h4 Re8 11. Bf4 a6 12. O-O-O c5 13. g4 cxd4 14. Nxd4 Nc5 15.
g5 Nxd3+ 16. Rxd3 Nh5 17. gxh6 Nxf4 18. exf4 g6 19. h5 Kh7 20. hxg6+ fxg6 21.
Rg3 Rg8 22. Rhg1 Qb6 23. Nxd5 Qxd4 24. Rxg6 Qc5 25. Rg7+ Kxh6 26. Rh7# 1-0
[/pgn]

[d]r1bq1rk1/pp1nbpp1/2p2n1p/3p2B1/3P4/2NBP3/PPQ1NPPP/R3K2R w KQ - 0 10

It is interesting to know how many engines would play now 10.h4, like Alekhine, deciding to castle long in the aftermath.

I very much hope you are not bored by this, I think it has some relevance.

Best, Lyudmil

[Don]

Castling statistics will have a lot to do with the book.

Komodo has a wide but very shallow book of exactly 5 moves - with about 9% of the openings having castle short in them. Only 2 position have long castling out of over 35,000 openings.

I looked at the castling statistics and about 95.66% of all castling incidents were short side castling and this included those dictated by the book.

It would be interested to check this by level - but this is not an interesting enough experiment for me to take time running although I am curious.

I have not checked this for other programs.


0.9566482233548657

[Lyudmil Tsvetkov]

Castling statistics will have a lot to do with the book.

Komodo has a wide but very shallow book of exactly 5 moves - with about 9% of the openings having castle short in them. Only 2 position have long castling out of over 35,000 openings.

I looked at the castling statistics and about 95.66% of all castling incidents were short side castling and this included those dictated by the book.

It would be interested to check this by level - but this is not an interesting enough experiment for me to take time running although I am curious.

I have not checked this for other programs.


0.9566482233548657

Thanks, Don, interesting.

I hope Komodo is not losing any strength because of a very low percent of long castling. But, of course, Komodo should know better.
Do you have by chance special penalties for castling long? 5% seems a very intriguing figure to me. You know, I just got curious by those Alekhine games, where he was castling long in an extremely wide variety of openings (almost all), both as black and white.

Best, Lyudmil

[Don]

Castling statistics will have a lot to do with the book.

Komodo has a wide but very shallow book of exactly 5 moves - with about 9% of the openings having castle short in them. Only 2 position have long castling out of over 35,000 openings.

I looked at the castling statistics and about 95.66% of all castling incidents were short side castling and this included those dictated by the book.

It would be interested to check this by level - but this is not an interesting enough experiment for me to take time running although I am curious.

I have not checked this for other programs.


0.9566482233548657

Thanks, Don, interesting.

I hope Komodo is not losing any strength because of a very low percent of long castling. But, of course, Komodo should know better.
Do you have by chance special penalties for castling long? 5% seems a very intriguing figure to me. You know, I just got curious by those Alekhine games, where he was castling long in an extremely wide variety of openings (almost all), both as black and white.

Best, Lyudmil

Castling long gets rewarded in Komodo. If castling short were not legal Komodo would try to castle long. Komodo will also become aggressive about attacking the opposite castled king with a pawn storm.

I'm sure you already know this, but the king tends to be more vulnerable on the queen side of the board, it is harder to prepare long castling properly and after you castle you will inevitably have to lose a tempo playing Kb1. The reason Komodo doesn't castle long more often is in part because it would rather the king be on b1 than c1 and would rather be making active moves.

I looked in the game collection I have and human players castle long almost 11% of the time. I don't know what it means that Komodo castles long so much less.

[Lyudmil Tsvetkov]

Castling statistics will have a lot to do with the book.

Komodo has a wide but very shallow book of exactly 5 moves - with about 9% of the openings having castle short in them. Only 2 position have long castling out of over 35,000 openings.

I looked at the castling statistics and about 95.66% of all castling incidents were short side castling and this included those dictated by the book.

It would be interested to check this by level - but this is not an interesting enough experiment for me to take time running although I am curious.

I have not checked this for other programs.


0.9566482233548657

Thanks, Don, interesting.

I hope Komodo is not losing any strength because of a very low percent of long castling. But, of course, Komodo should know better.
Do you have by chance special penalties for castling long? 5% seems a very intriguing figure to me. You know, I just got curious by those Alekhine games, where he was castling long in an extremely wide variety of openings (almost all), both as black and white.

Best, Lyudmil

Castling long gets rewarded in Komodo. If castling short were not legal Komodo would try to castle long. Komodo will also become aggressive about attacking the opposite castled king with a pawn storm.

I'm sure you already know this, but the king tends to be more vulnerable on the queen side of the board, it is harder to prepare long castling properly and after you castle you will inevitably have to lose a tempo playing Kb1. The reason Komodo doesn't castle long more often is in part because it would rather the king be on b1 than c1 and would rather be making active moves.

I looked in the game collection I have and human players castle long almost 11% of the time. I don't know what it means that Komodo castles long so much less.

Thanks Don, again, for the interesting statistics. I wonder how you are able to get all this stat in such a short while.

I thought Komodo (and other engines btw) would have some bonus points for castling short (or the equivalent of the king going to a safe square as quickly as possible). In such a case, it would be very natural that engines would castle more often short than humans.

The difference between castling short and long is just 2 tempos: one piece developing move and one king move to a safer square (c1-b1). The question is: would not it be possible not to compromise on quality while trying to castle long instead, for example, by developing in a priority way the queen side; losing some tempos for queen side development might retard a bit the king going to a safer square, but this might not be all that decisive, while at the same time other factors, like good control of center, might compensate for the temporarily unsafe king. So the query would be: is not it possible to assume that by castling much too often short, hurrying to move the king to a safer square, engines actually miss to enter some very promising variations (in many cases much more promising than the short castling ones)?

[Don]

Castling statistics will have a lot to do with the book.

Komodo has a wide but very shallow book of exactly 5 moves - with about 9% of the openings having castle short in them. Only 2 position have long castling out of over 35,000 openings.

I looked at the castling statistics and about 95.66% of all castling incidents were short side castling and this included those dictated by the book.

It would be interested to check this by level - but this is not an interesting enough experiment for me to take time running although I am curious.

I have not checked this for other programs.


0.9566482233548657

Thanks, Don, interesting.

I hope Komodo is not losing any strength because of a very low percent of long castling. But, of course, Komodo should know better.
Do you have by chance special penalties for castling long? 5% seems a very intriguing figure to me. You know, I just got curious by those Alekhine games, where he was castling long in an extremely wide variety of openings (almost all), both as black and white.

Best, Lyudmil

Castling long gets rewarded in Komodo. If castling short were not legal Komodo would try to castle long. Komodo will also become aggressive about attacking the opposite castled king with a pawn storm.

I'm sure you already know this, but the king tends to be more vulnerable on the queen side of the board, it is harder to prepare long castling properly and after you castle you will inevitably have to lose a tempo playing Kb1. The reason Komodo doesn't castle long more often is in part because it would rather the king be on b1 than c1 and would rather be making active moves.

I looked in the game collection I have and human players castle long almost 11% of the time. I don't know what it means that Komodo castles long so much less.

Thanks Don, again, for the interesting statistics. I wonder how you are able to get all this stat in such a short while.

I'm that good!

Seriously, you can easily do this stuff with command line tools in Linux but I wrote a short tcl script to do this. Any good perl, ruby, python programmer can write a script like this in 5 minutes. If there are subvariations in the PGN file you should extract them out with pgn-extact and then pipe the PGN file through this filter:

Code: Select all

#! /usr/bin/env tclkit

while { [gets stdin s] >= 0 } {

    foreach w [split $s] {
	if { $w == "O-O" } {
	    incr short
	} elseif { $w == "O-O-O" } {
	    incr long 
	}
    }

}

set both [expr $short + $long]
set sr [expr $short / ($both + 0.0)]
set lr [expr $long / ($both + 0.0)]

puts [format "short: %8d  %8.2 perc", $short, [expr $sr * 100.0]]
puts [format "long: %8d  %8.2 perc", $long, [expr $lr * 100.0]]
puts ""

I thought Komodo (and other engines btw) would have some bonus points for castling short (or the equivalent of the king going to a safe square as quickly as possible). In such a case, it would be very natural that engines would castle more often short than humans.

Castling is difficult to get right but we do have incentives for doing so. It's a combination of things - the piece square tables for the king in the opening phase likes the 4 squares in the 2 corners of the board, g1,h1,g2,h2 and of course e1 is penalized. But this has to be balanced with other things so that it is not too eager to castle or to play Kf1 when it cannot right away or trap the rook on h1 or a1.

The difference between castling short and long is just 2 tempos: one piece developing move and one king move to a safer square (c1-b1). The question is: would not it be possible not to compromise on quality while trying to castle long instead, for example, by developing in a priority way the queen side; losing some tempos for queen side development might retard a bit the king going to a safer square, but this might not be all that decisive, while at the same time other factors, like good control of center, might compensate for the temporarily unsafe king. So the query would be: is not it possible to assume that by castling much too often short, hurrying to move the king to a safer square, engines actually miss to enter some very promising variations (in many cases much more promising than the short castling ones)?

We could experiment with this, but we usually are pretty obsessive about getting everything just right to the best of our ability. It's been a long time since we worked on that code but it's very likely we experimented with much more aggressive and slightly more aggressive queen side castling and ended up where we are now.

[velmarin]

If not misunderstand, Mr. Lyudmil, it seems that you think that simply making a "long castling" is a virtue.
The motors can look the other way, but "long castling" is done, I wanted to show a game and have found several on the same page.
According to the conditions chosen engines.

As noted by Mr Daley, in a similar way Bouquet_beta(1.7) changes the whole of static pawns, second positions of kings,
this may facilitate long castling, pawn storm also prepared in case of different sides of castling, this provides more space and more aggressiveness in the game.


Url: http://www.chess2u.com/t6680p75-short-and-sweet
five games examples
[pgn]
[Event "3 Minutes/Game"]
[Site "Stockfish 26-07-2013 64bit Ga"]
[Date "2013.07.27"]
[Round "56.5"]
[White "Bouquet 1.7 beta 1 x64"]
[Black "Stockfish 26-07-2013 64bit"]
[Result "1-0"]
[PlyCount "43"]
[EventDate "2013.??.??"]

1. e4 {book 0s} e6 {book 0s} 2. d4 {book 0s} d5 {book 0s} 3. Nc3 {book 0s} Nf6
{book 0s} 4. Bg5 {book 0s} dxe4 {book 0s} 5. Nxe4 {book 0s} Be7 {book 0s} 6.
Bxf6 {book 0s} Bxf6 {book 0s} 7. Nf3 {book 0s} O-O {book 0s} 8. Qd2 {book 0s}
Be7 {book 0s} 9. O-O-O {0.09/18 29} Nc6 {0.08/16 5} 10. Bc4 {0.24/17 4} b6 {0.
16/17 4} 11. Kb1 {0.34/14 1} Bb7 {0.16/18 2} 12. Qe2 {0.28/16 5 (Rhe1)} Nb4 {
0.12/16 2} 13. h4 {0.23/16 5 (Rhe1)} Rb8 {0.08/15 4} 14. Ne5 {0.36/14 4 (c3)}
a5 {0.36/15 5} 15. f3 {0.49/16 10 (c3)} Bxh4 {0.54/15 4} 16. g3 {1.66/17 17
(a3)} Be7 {0.22/17 2} 17. Rxh7 {3.30/15 3 (c3)} Kxh7 {-0.36/17 0} 18. Rh1+ {3.
30/14 0} Kg8 {4.88/18 3} 19. Ng6 {3.41/16 1} Bh4 {4.88/18 3} 20. Rxh4 {3.41/14
0} Qxh4 {5.03/19 3} 21. Nxh4 {3.44/15 0} Bd5 {5.21/18 2} 22. f4 {3.53/15 1 (c3)
} 1-0
[Event "3 Minutes/Game"]
[Site "Komodo(s) 64-bit Gauntlet, US"]
[Date "2013.07.19"]
[Round "54.1"]
[White "Komodo 5.1r2 64-bit"]
[Black "Komodo CCT 64-bit"]
[Result "0-1"]
[PlyCount "66"]
[EventDate "2013.??.??"]

1. e4 {book 0s} c5 {book 0s} 2. Nf3 {book 0s} e6 {book 0s} 3. d4 {book 0s} cxd4
{book 0s} 4. Nxd4 {book 0s} a6 {book 0s} 5. Nc3 {book 0s} Qc7 {book 0s} 6. g3 {
book 0s} Bb4 {book 0s} 7. Bd2 {book 0s} Nc6 {book 0s} 8. Nb3 {book 0s} Nf6 {0.
04/14 1} 9. Bg2 {0.15/16 6 (a3)} Be7 {book 0s} 10. f4 {0.13/17 3} d6 {0.15/16 2
} 11. Qf3 {0.11/16 7 (Qe2)} Nd7 {0.17/16 7} 12. O-O-O {0.18/17 6} b5 {0.17/16 2
} 13. Kb1 {0.11/17 12 (Be3)} Nb6 {0.02/16 3 (O-O)} 14. Ne2 {0.07/16 4} Nc4 {0.
10/16 4} 15. Bc3 {0.00/16 5} O-O {0.08/15 2 (e5)} 16. Nd2 {0.03/16 5} Nxd2+ {
0.06/16 5} 17. Bxd2 {0.02/17 5} Bb7 {0.08/16 4} 18. h4 {-0.03/16 15} Rfc8 {-0.
05/16 9} 19. g4 {0.02/15 5 (h5)} Na5 {-0.08/15 3} 20. Qd3 {0.06/16 8 (Bxa5)}
Nc4 {-0.15/13 1} 21. g5 {0.06/15 2 (h5)} d5 {-0.23/16 13} 22. e5 {0.02/14 4
(Bc3)} b4 {-0.51/14 3 (a5)} 23. h5 {-0.50/14 7} a5 {-0.67/14 2} 24. Nd4 {-0.60/
13 2} Bc5 {-0.76/14 3} 25. Nb3 {-0.95/14 5} Ba6 {-1.19/15 3} 26. Qg3 {-0.91/15
3 (Qf3)} a4 {-1.28/14 6 (Ba7)} 27. Na1 {-1.34/14 2} Rab8 {-2.98/14 4 (a3)} 28.
g6 {-2.11/12 1} Na3+ {-4.24/14 2 (Bd4)} 29. bxa3 {-4.06/12 1} bxa3+ {-4.45/14 2
} 30. Nb3 {-4.35/13 1} axb3 {-4.55/14 1} 31. gxf7+ {-4.51/13 1 (axb3)} Kxf7 {
-8.88/13 3} 32. Qxb3 {-10.91/14 9} Bd4 {-12.64/15 2} 33. Rc1 {-12.15/14 1}
Rxb3+ {-13.54/15 2} 0-1
[Event "3 Minutes/Game"]
[Site "Stockfish 26-07-2013 64bit Ga"]
[Date "2013.07.26"]
[Round "7.2"]
[White "Houdini 3 Pro x64"]
[Black "Stockfish 26-07-2013 64bit"]
[Result "1-0"]
[PlyCount "49"]
[EventDate "2013.??.??"]

1. e4 {book 0s} c6 {book 0s} 2. d4 {book 0s} d5 {book 0s} 3. e5 {book 0s} Bf5 {
book 0s} 4. Nc3 {book 0s} e6 {book 0s} 5. g4 {book 0s} Bg6 {book 0s} 6. Nge2 {
book 0s} Qh4 {book 0s} 7. Nf4 {0.57/15 4} Bb4 {0.22/15 3} 8. Be2 {0.29/16 8}
Ne7 {0.42/17 7 (Nd7)} 9. O-O {0.52/16 3} h5 {0.68/19 11} 10. gxh5 {0.50/17 4}
Bxc3 {0.66/19 2} 11. bxc3 {0.54/17 17} Nd7 {0.66/18 4 (Bf5)} 12. Nxg6 {0.50/17
13} Nxg6 {0.68/17 3} 13. Bg4 {0.62/16 2} O-O-O {0.52/18 5} 14. h3 {0.42/16 2}
Ne7 {0.94/19 4} 15. Qf3 {0.32/15 2} Nf5 {0.98/20 3} 16. Kh2 {0.57/11 2} g6 {0.
84/19 2 (Rdg8)} 17. Rg1 {0.69/15 1} Rdg8 {0.98/18 4} 18. h6 {0.75/15 1} f6 {1.
27/18 5} 19. Bxf5 {0.94/16 2} gxf5 {1.21/20 3} 20. Bf4 {0.93/15 0} fxe5 {1.35/
20 7} 21. dxe5 {0.94/14 0} Kd8 {1.41/20 11 (Nc5)} 22. Rab1 {1.14/15 2 (c4)} Nb6
{1.45/17 4 (b6)} 23. Rxg8+ {5.70/13 1 (Be3)} Rxg8 {5.95/16 1} 24. Rg1 {5.26/12
0} Rxg1 {7.69/20 4} 25. Kxg1 {5.05/11 0} 1-0
[Event "3 Minutes/Game"]
[Site "Stockfish 26-07-2013 64bit Ga"]
[Date "2013.07.27"]
[Round "101.3"]
[White "Stockfish 26-07-2013 64bit"]
[Black "Komodo 5.1r2 64-bit"]
[Result "1-0"]
[PlyCount "47"]
[EventDate "2013.??.??"]

1. d4 {book 0s} Nf6 {book 0s} 2. Nf3 {book 0s} a6 {book 0s} 3. Bg5 {book 0s} e6
{book 0s} 4. e4 {0.46/16 3} h6 {0.29/16 6} 5. Bxf6 {0.44/17 3} Qxf6 {0.23/16 4}
6. Nc3 {0.36/18 5 (Nbd2)} Nc6 {0.13/16 2} 7. Be2 {0.34/17 4 (e5)} d6 {0.10/15
2 (Bb4)} 8. O-O {0.12/16 7} Bd7 {-0.09/14 2} 9. Rb1 {0.30/18 6 (d5)} Be7 {0.03/
15 5} 10. a3 {0.14/19 9 (Qd3)} O-O-O {-0.27/15 3 (O-O)} 11. b4 {0.66/17 4 (d5)}
g5 {0.13/15 6 (Kb8)} 12. Qd3 {0.80/18 5 (b5)} Kb8 {-0.10/14 5 (g4)} 13. b5 {2.
68/16 1 (d5)} axb5 {0.33/13 1} 14. Qxb5 {3.29/17 1} b6 {2.27/16 5} 15. Qa6 {3.
15/18 0} Na7 {2.49/16 4} 16. a4 {3.29/18 3 (e5)} Bc8 {1.62/14 3} 17. Qc4 {3.33/
18 1} Bb7 {1.61/14 7 (d5)} 18. a5 {3.83/15 3} Rc8 {1.94/15 11 (d5)} 19. d5 {4.
50/14 2} Rh7 {2.00/15 11 (Bd8)} 20. Rb3 {5.07/14 4 (a6)} Bd8 {2.00/13 5 (exd5)}
21. Rfb1 {5.13/15 2 (a6)} Qf4 {2.44/12 4 (h5)} 22. Qa4 {8.64/14 0 (g3)} Qf6 {
5.47/13 3} 23. Ra1 {12.24/17 4 (e5)} b5 {7.84/13 3} 24. Nxb5 {13.73/13 0 (Bxb5)
} 1-0
[Event "stoc 29.07"]
[Site "Jarqurgan"]
[Date "2013.07.30"]
[Round "20.3"]
[White "Houdini 3 Pro x64"]
[Black "Stockfish 290713 64"]
[Result "0-1"]
[ECO "B90"]
[Annotator "0.21;0.06"]
[PlyCount "64"]
[EventDate "2013.07.30"]
[EventType "simul"]
[Source "Sharafutdinov"]
[TimeControl "180"]

{Pentium(R) Dual-Core  CPU      E5700  @ 3.00GHz 2999 MHz  W=15.8 plies; 2
896kN/s; Perfect2012t.ctg  B=17.5 plies; 2 048kN/s; Perfect2012t.ctg} 1. e4 {
[%eval 0,0] [%emt 0:00:00]} c5 {[%eval 0,0] [%emt 0:00:00]} 2. Nf3 {[%eval 0,0]
[%emt 0:00:00]} d6 {[%eval 0,0] [%emt 0:00:00]} 3. d4 {[%eval 0,0] [%emt 0:00:
00]} cxd4 {[%eval 0,0] [%emt 0:00:00]} 4. Nxd4 {[%eval 0,0] [%emt 0:00:00]} Nf6
{[%eval 0,0] [%emt 0:00:00]} 5. Nc3 {[%eval 0,0] [%emt 0:00:00]} a6 {[%eval 0,
0] [%emt 0:00:00]} 6. Be3 {[%eval 0,0] [%emt 0:00:00]} e5 {[%eval 0,0] [%emt 0:
00:00]} 7. Nb3 {[%eval 0,0] [%emt 0:00:00]} Be6 {[%eval 0,0] [%emt 0:00:00]} 8.
f3 {[%eval 0,0] [%emt 0:00:00]} Be7 {[%eval 0,0] [%emt 0:00:00]} 9. Qd2 {
[%eval 21,17] [%emt 0:00:05]} Nbd7 {[%eval 6,18] [%emt 0:00:05] (h5)} 10. O-O-O
{[%eval 15,18] [%emt 0:00:03] (g4)} b5 {[%eval -10,20] [%emt 0:00:05]} 11. Kb1
{[%eval 17,18] [%emt 0:00:03] (g4)} O-O {[%eval -30,17] [%emt 0:00:03] (h5)}
12. g4 {[%eval 24,19] [%emt 0:00:06] (Bd3)} b4 {[%eval -28,21] [%emt 0:00:03]}
13. Ne2 {[%eval 20,17] [%emt 0:00:02] (Na4)} a5 {[%eval -32,19] [%emt 0:00:03]}
14. g5 {[%eval 11,17] [%emt 0:00:05]} Nh5 {[%eval -18,20] [%emt 0:00:04]} 15.
Ng3 {[%eval 38,18] [%emt 0:00:03]} a4 {[%eval 8,20] [%emt 0:00:06]} 16. Nc1 {
[%eval 38,16] [%emt 0:00:00]} Nf4 {[%eval 0,20] [%emt 0:00:08]} 17. h4 {[%eval
33,18] [%emt 0:00:08] (Nf5)} Qc7 {[%eval -26,18] [%emt 0:00:03] (b3)} 18. Nf5 {
[%eval 27,18] [%emt 0:00:06]} Bxf5 {[%eval -6,19] [%emt 0:00:05]} 19. exf5 {
[%eval 27,17] [%emt 0:00:00]} Rfb8 {[%eval 4,20] [%emt 0:00:13] (Rfc8)} 20. Re1
{[%eval 44,15] [%emt 0:00:05] (Bxf4)} Bf8 {[%eval -8,17] [%emt 0:00:03]} 21. f6
{[%eval 45,15] [%emt 0:00:07] (g6)} Ne6 {[%eval 6,15] [%emt 0:00:03]} 22. Rg1 {
[%eval 34,16] [%emt 0:00:04] (Bh3)} Qc6 {[%eval -40,16] [%emt 0:00:05] (gxf6)}
23. f4 {[%eval 34,14] [%emt 0:00:02]} exf4 {[%eval -141,16] [%emt 0:00:04]} 24.
Bxf4 {[%eval -24,13] [%emt 0:00:03]} b3 {[%eval -137,19] [%emt 0:00:05]} 25.
Be3 {[%eval -38,15] [%emt 0:00:03] (cxb3)} d5 {[%eval -208,16] [%emt 0:00:03]
(bxa2+)} 26. c3 {[%eval -32,15] [%emt 0:00:02] (Nd3)} a3 {[%eval -468,18]
[%emt 0:00:12] (bxa2+)} 27. Nxb3 {[%eval -336,14] [%emt 0:00:06] (axb3)} axb2 {
[%eval -511,13] [%emt 0:00:01]} 28. Bg2 {[%eval -397,13] [%emt 0:00:04] (Bd3)}
Qa6 {[%eval -638,16] [%emt 0:00:01]} 29. Na5 {[%eval -607,15] [%emt 0:00:17]
(Qxb2)} Qxa5 {[%eval -626,15] [%emt 0:00:01]} 30. Qxd5 {[%eval -828,15] [%emt
0:00:14] (Bxd5)} Qa3 {[%eval -981,18] [%emt 0:00:03] (Qa4)} 31. Kc2 {[%eval
-835,12] [%emt 0:00:01] (Qxa8)} Ra6 {[%eval -1173,14] [%emt 0:00:02] (b1Q+)}
32. Be4 {[%eval -981,13] [%emt 0:00:04] (Bd2)} Rd6 {[%eval -1218,14] [%emt 0:
00:01] (Qa4+)} 0-1
[/pgn]

[Lyudmil Tsvetkov]

I'm that good!

I am sure about it

Any good perl, ruby, python programmer can write a script like this in 5 minutes.

All I can think of is 'Pearl Harbour', 'Ruby-Gate', I do not know about pythons.
Seriously, I am ashamed I am so backward about programming, but here patience is required.

It's been a long time since we worked on that code but it's very likely we experimented with much more aggressive and slightly more aggressive queen side castling and ended up where we are now.

This should suggest then that Komodo is right about the 5%. Still, I am a bit sceptic (or is skeptic the right word)?

[Lyudmil Tsvetkov]

If not misunderstand, Mr. Lyudmil, it seems that you think that simply making a "long castling" is a virtue.
The motors can look the other way, but "long castling" is done, I wanted to show a game and have found several on the same page.
According to the conditions chosen engines.

As noted by Mr Daley, in a similar way Bouquet_beta(1.7) changes the whole of static pawns, second positions of kings,
this may facilitate long castling, pawn storm also prepared in case of different sides of castling, this provides more space and more aggressiveness in the game.


Url: http://www.chess2u.com/t6680p75-short-and-sweet
five games examples
[pgn]
[Event "3 Minutes/Game"]
[Site "Stockfish 26-07-2013 64bit Ga"]
[Date "2013.07.27"]
[Round "56.5"]
[White "Bouquet 1.7 beta 1 x64"]
[Black "Stockfish 26-07-2013 64bit"]
[Result "1-0"]
[PlyCount "43"]
[EventDate "2013.??.??"]

1. e4 {book 0s} e6 {book 0s} 2. d4 {book 0s} d5 {book 0s} 3. Nc3 {book 0s} Nf6
{book 0s} 4. Bg5 {book 0s} dxe4 {book 0s} 5. Nxe4 {book 0s} Be7 {book 0s} 6.
Bxf6 {book 0s} Bxf6 {book 0s} 7. Nf3 {book 0s} O-O {book 0s} 8. Qd2 {book 0s}
Be7 {book 0s} 9. O-O-O {0.09/18 29} Nc6 {0.08/16 5} 10. Bc4 {0.24/17 4} b6 {0.
16/17 4} 11. Kb1 {0.34/14 1} Bb7 {0.16/18 2} 12. Qe2 {0.28/16 5 (Rhe1)} Nb4 {
0.12/16 2} 13. h4 {0.23/16 5 (Rhe1)} Rb8 {0.08/15 4} 14. Ne5 {0.36/14 4 (c3)}
a5 {0.36/15 5} 15. f3 {0.49/16 10 (c3)} Bxh4 {0.54/15 4} 16. g3 {1.66/17 17
(a3)} Be7 {0.22/17 2} 17. Rxh7 {3.30/15 3 (c3)} Kxh7 {-0.36/17 0} 18. Rh1+ {3.
30/14 0} Kg8 {4.88/18 3} 19. Ng6 {3.41/16 1} Bh4 {4.88/18 3} 20. Rxh4 {3.41/14
0} Qxh4 {5.03/19 3} 21. Nxh4 {3.44/15 0} Bd5 {5.21/18 2} 22. f4 {3.53/15 1 (c3)
} 1-0
[Event "3 Minutes/Game"]
[Site "Komodo(s) 64-bit Gauntlet, US"]
[Date "2013.07.19"]
[Round "54.1"]
[White "Komodo 5.1r2 64-bit"]
[Black "Komodo CCT 64-bit"]
[Result "0-1"]
[PlyCount "66"]
[EventDate "2013.??.??"]

1. e4 {book 0s} c5 {book 0s} 2. Nf3 {book 0s} e6 {book 0s} 3. d4 {book 0s} cxd4
{book 0s} 4. Nxd4 {book 0s} a6 {book 0s} 5. Nc3 {book 0s} Qc7 {book 0s} 6. g3 {
book 0s} Bb4 {book 0s} 7. Bd2 {book 0s} Nc6 {book 0s} 8. Nb3 {book 0s} Nf6 {0.
04/14 1} 9. Bg2 {0.15/16 6 (a3)} Be7 {book 0s} 10. f4 {0.13/17 3} d6 {0.15/16 2
} 11. Qf3 {0.11/16 7 (Qe2)} Nd7 {0.17/16 7} 12. O-O-O {0.18/17 6} b5 {0.17/16 2
} 13. Kb1 {0.11/17 12 (Be3)} Nb6 {0.02/16 3 (O-O)} 14. Ne2 {0.07/16 4} Nc4 {0.
10/16 4} 15. Bc3 {0.00/16 5} O-O {0.08/15 2 (e5)} 16. Nd2 {0.03/16 5} Nxd2+ {
0.06/16 5} 17. Bxd2 {0.02/17 5} Bb7 {0.08/16 4} 18. h4 {-0.03/16 15} Rfc8 {-0.
05/16 9} 19. g4 {0.02/15 5 (h5)} Na5 {-0.08/15 3} 20. Qd3 {0.06/16 8 (Bxa5)}
Nc4 {-0.15/13 1} 21. g5 {0.06/15 2 (h5)} d5 {-0.23/16 13} 22. e5 {0.02/14 4
(Bc3)} b4 {-0.51/14 3 (a5)} 23. h5 {-0.50/14 7} a5 {-0.67/14 2} 24. Nd4 {-0.60/
13 2} Bc5 {-0.76/14 3} 25. Nb3 {-0.95/14 5} Ba6 {-1.19/15 3} 26. Qg3 {-0.91/15
3 (Qf3)} a4 {-1.28/14 6 (Ba7)} 27. Na1 {-1.34/14 2} Rab8 {-2.98/14 4 (a3)} 28.
g6 {-2.11/12 1} Na3+ {-4.24/14 2 (Bd4)} 29. bxa3 {-4.06/12 1} bxa3+ {-4.45/14 2
} 30. Nb3 {-4.35/13 1} axb3 {-4.55/14 1} 31. gxf7+ {-4.51/13 1 (axb3)} Kxf7 {
-8.88/13 3} 32. Qxb3 {-10.91/14 9} Bd4 {-12.64/15 2} 33. Rc1 {-12.15/14 1}
Rxb3+ {-13.54/15 2} 0-1
[Event "3 Minutes/Game"]
[Site "Stockfish 26-07-2013 64bit Ga"]
[Date "2013.07.26"]
[Round "7.2"]
[White "Houdini 3 Pro x64"]
[Black "Stockfish 26-07-2013 64bit"]
[Result "1-0"]
[PlyCount "49"]
[EventDate "2013.??.??"]

1. e4 {book 0s} c6 {book 0s} 2. d4 {book 0s} d5 {book 0s} 3. e5 {book 0s} Bf5 {
book 0s} 4. Nc3 {book 0s} e6 {book 0s} 5. g4 {book 0s} Bg6 {book 0s} 6. Nge2 {
book 0s} Qh4 {book 0s} 7. Nf4 {0.57/15 4} Bb4 {0.22/15 3} 8. Be2 {0.29/16 8}
Ne7 {0.42/17 7 (Nd7)} 9. O-O {0.52/16 3} h5 {0.68/19 11} 10. gxh5 {0.50/17 4}
Bxc3 {0.66/19 2} 11. bxc3 {0.54/17 17} Nd7 {0.66/18 4 (Bf5)} 12. Nxg6 {0.50/17
13} Nxg6 {0.68/17 3} 13. Bg4 {0.62/16 2} O-O-O {0.52/18 5} 14. h3 {0.42/16 2}
Ne7 {0.94/19 4} 15. Qf3 {0.32/15 2} Nf5 {0.98/20 3} 16. Kh2 {0.57/11 2} g6 {0.
84/19 2 (Rdg8)} 17. Rg1 {0.69/15 1} Rdg8 {0.98/18 4} 18. h6 {0.75/15 1} f6 {1.
27/18 5} 19. Bxf5 {0.94/16 2} gxf5 {1.21/20 3} 20. Bf4 {0.93/15 0} fxe5 {1.35/
20 7} 21. dxe5 {0.94/14 0} Kd8 {1.41/20 11 (Nc5)} 22. Rab1 {1.14/15 2 (c4)} Nb6
{1.45/17 4 (b6)} 23. Rxg8+ {5.70/13 1 (Be3)} Rxg8 {5.95/16 1} 24. Rg1 {5.26/12
0} Rxg1 {7.69/20 4} 25. Kxg1 {5.05/11 0} 1-0
[Event "3 Minutes/Game"]
[Site "Stockfish 26-07-2013 64bit Ga"]
[Date "2013.07.27"]
[Round "101.3"]
[White "Stockfish 26-07-2013 64bit"]
[Black "Komodo 5.1r2 64-bit"]
[Result "1-0"]
[PlyCount "47"]
[EventDate "2013.??.??"]

1. d4 {book 0s} Nf6 {book 0s} 2. Nf3 {book 0s} a6 {book 0s} 3. Bg5 {book 0s} e6
{book 0s} 4. e4 {0.46/16 3} h6 {0.29/16 6} 5. Bxf6 {0.44/17 3} Qxf6 {0.23/16 4}
6. Nc3 {0.36/18 5 (Nbd2)} Nc6 {0.13/16 2} 7. Be2 {0.34/17 4 (e5)} d6 {0.10/15
2 (Bb4)} 8. O-O {0.12/16 7} Bd7 {-0.09/14 2} 9. Rb1 {0.30/18 6 (d5)} Be7 {0.03/
15 5} 10. a3 {0.14/19 9 (Qd3)} O-O-O {-0.27/15 3 (O-O)} 11. b4 {0.66/17 4 (d5)}
g5 {0.13/15 6 (Kb8)} 12. Qd3 {0.80/18 5 (b5)} Kb8 {-0.10/14 5 (g4)} 13. b5 {2.
68/16 1 (d5)} axb5 {0.33/13 1} 14. Qxb5 {3.29/17 1} b6 {2.27/16 5} 15. Qa6 {3.
15/18 0} Na7 {2.49/16 4} 16. a4 {3.29/18 3 (e5)} Bc8 {1.62/14 3} 17. Qc4 {3.33/
18 1} Bb7 {1.61/14 7 (d5)} 18. a5 {3.83/15 3} Rc8 {1.94/15 11 (d5)} 19. d5 {4.
50/14 2} Rh7 {2.00/15 11 (Bd8)} 20. Rb3 {5.07/14 4 (a6)} Bd8 {2.00/13 5 (exd5)}
21. Rfb1 {5.13/15 2 (a6)} Qf4 {2.44/12 4 (h5)} 22. Qa4 {8.64/14 0 (g3)} Qf6 {
5.47/13 3} 23. Ra1 {12.24/17 4 (e5)} b5 {7.84/13 3} 24. Nxb5 {13.73/13 0 (Bxb5)
} 1-0
[Event "stoc 29.07"]
[Site "Jarqurgan"]
[Date "2013.07.30"]
[Round "20.3"]
[White "Houdini 3 Pro x64"]
[Black "Stockfish 290713 64"]
[Result "0-1"]
[ECO "B90"]
[Annotator "0.21;0.06"]
[PlyCount "64"]
[EventDate "2013.07.30"]
[EventType "simul"]
[Source "Sharafutdinov"]
[TimeControl "180"]

{Pentium(R) Dual-Core  CPU      E5700  @ 3.00GHz 2999 MHz  W=15.8 plies; 2
896kN/s; Perfect2012t.ctg  B=17.5 plies; 2 048kN/s; Perfect2012t.ctg} 1. e4 {
[%eval 0,0] [%emt 0:00:00]} c5 {[%eval 0,0] [%emt 0:00:00]} 2. Nf3 {[%eval 0,0]
[%emt 0:00:00]} d6 {[%eval 0,0] [%emt 0:00:00]} 3. d4 {[%eval 0,0] [%emt 0:00:
00]} cxd4 {[%eval 0,0] [%emt 0:00:00]} 4. Nxd4 {[%eval 0,0] [%emt 0:00:00]} Nf6
{[%eval 0,0] [%emt 0:00:00]} 5. Nc3 {[%eval 0,0] [%emt 0:00:00]} a6 {[%eval 0,
0] [%emt 0:00:00]} 6. Be3 {[%eval 0,0] [%emt 0:00:00]} e5 {[%eval 0,0] [%emt 0:
00:00]} 7. Nb3 {[%eval 0,0] [%emt 0:00:00]} Be6 {[%eval 0,0] [%emt 0:00:00]} 8.
f3 {[%eval 0,0] [%emt 0:00:00]} Be7 {[%eval 0,0] [%emt 0:00:00]} 9. Qd2 {
[%eval 21,17] [%emt 0:00:05]} Nbd7 {[%eval 6,18] [%emt 0:00:05] (h5)} 10. O-O-O
{[%eval 15,18] [%emt 0:00:03] (g4)} b5 {[%eval -10,20] [%emt 0:00:05]} 11. Kb1
{[%eval 17,18] [%emt 0:00:03] (g4)} O-O {[%eval -30,17] [%emt 0:00:03] (h5)}
12. g4 {[%eval 24,19] [%emt 0:00:06] (Bd3)} b4 {[%eval -28,21] [%emt 0:00:03]}
13. Ne2 {[%eval 20,17] [%emt 0:00:02] (Na4)} a5 {[%eval -32,19] [%emt 0:00:03]}
14. g5 {[%eval 11,17] [%emt 0:00:05]} Nh5 {[%eval -18,20] [%emt 0:00:04]} 15.
Ng3 {[%eval 38,18] [%emt 0:00:03]} a4 {[%eval 8,20] [%emt 0:00:06]} 16. Nc1 {
[%eval 38,16] [%emt 0:00:00]} Nf4 {[%eval 0,20] [%emt 0:00:08]} 17. h4 {[%eval
33,18] [%emt 0:00:08] (Nf5)} Qc7 {[%eval -26,18] [%emt 0:00:03] (b3)} 18. Nf5 {
[%eval 27,18] [%emt 0:00:06]} Bxf5 {[%eval -6,19] [%emt 0:00:05]} 19. exf5 {
[%eval 27,17] [%emt 0:00:00]} Rfb8 {[%eval 4,20] [%emt 0:00:13] (Rfc8)} 20. Re1
{[%eval 44,15] [%emt 0:00:05] (Bxf4)} Bf8 {[%eval -8,17] [%emt 0:00:03]} 21. f6
{[%eval 45,15] [%emt 0:00:07] (g6)} Ne6 {[%eval 6,15] [%emt 0:00:03]} 22. Rg1 {
[%eval 34,16] [%emt 0:00:04] (Bh3)} Qc6 {[%eval -40,16] [%emt 0:00:05] (gxf6)}
23. f4 {[%eval 34,14] [%emt 0:00:02]} exf4 {[%eval -141,16] [%emt 0:00:04]} 24.
Bxf4 {[%eval -24,13] [%emt 0:00:03]} b3 {[%eval -137,19] [%emt 0:00:05]} 25.
Be3 {[%eval -38,15] [%emt 0:00:03] (cxb3)} d5 {[%eval -208,16] [%emt 0:00:03]
(bxa2+)} 26. c3 {[%eval -32,15] [%emt 0:00:02] (Nd3)} a3 {[%eval -468,18]
[%emt 0:00:12] (bxa2+)} 27. Nxb3 {[%eval -336,14] [%emt 0:00:06] (axb3)} axb2 {
[%eval -511,13] [%emt 0:00:01]} 28. Bg2 {[%eval -397,13] [%emt 0:00:04] (Bd3)}
Qa6 {[%eval -638,16] [%emt 0:00:01]} 29. Na5 {[%eval -607,15] [%emt 0:00:17]
(Qxb2)} Qxa5 {[%eval -626,15] [%emt 0:00:01]} 30. Qxd5 {[%eval -828,15] [%emt
0:00:14] (Bxd5)} Qa3 {[%eval -981,18] [%emt 0:00:03] (Qa4)} 31. Kc2 {[%eval
-835,12] [%emt 0:00:01] (Qxa8)} Ra6 {[%eval -1173,14] [%emt 0:00:02] (b1Q+)}
32. Be4 {[%eval -981,13] [%emt 0:00:04] (Bd2)} Rd6 {[%eval -1218,14] [%emt 0:
00:01] (Qa4+)} 0-1
[/pgn]

Gracias, Don Jose!
Estas preparando un motor excepcional, a juzgar de esta partida.

17. Rh7 is a hammer (fantastico); obviously, Stockfish does not see 19. Ng6 (also a very exceptional move).
It is interesting that at move 17 Stockfish sees itself 0.36 pawns in advantage, and one move later already losing by 5 full pawns!!

Me alegro mucho de esta partida, Jose, Bouquet va adelante de una manera pujante.

Seriously, great game.
Bye, Lyudmil

[velmarin]

Thank you, Lyudmil.
I also felt surprised me.

Remember if clikea to points above the " A8 black rook" has more games other engines.
Total 5 games whith "O.O.O", on the same page.