When can we Expect CCRL to start testing No Castling ?

[AdminX]

Zahak parses all fields of fen, an respects them. If you cannot reproduce the issue with Zahak then it is an engine issue, if you could then it is a GUI issue

Was able to reproduce the issue in CB14 even with Zahak. So maybe a GUI Issue after all.

[pohl4711]

[d]knbrqbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQRBNK w - - 0 1
[d]knbqrbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBRQBNK w - - 0 1
[d]rnbqrbnk/pppppppp/8/8/8/8/PPPPPPPP/KNBRQBNR w - - 0 1
[d]rnbrqbnk/pppppppp/8/8/8/8/PPPPPPPP/KNBQRBNR w - - 0 1

These are the 4 starting positions of my Drawkiller openings. If anybody wants to play chess without castling, these openings are much better for interesting games.
In my AntiDraw openings collection (in the Drawkiller V5 folder), you find unbalanced Drawkiller openings, starting from these 4 positions. Some with pawn-moves, only, and others followed by classical human opening moves. All analyzed and evaluated.

Example line (Drawkiller V5, dk_+0.90_+0.99_human_6mvs.pgn-File):
[pgn]
[Event "Valjevo"]
[Site "Valjevo"]
[Date "2007.06.20"]
[Round "8"]
[White "Ivanisevic, Ivan"]
[Black "Roiz, Michael"]
[Result "1/2-1/2"]
[ECO "D15"]
[WhiteElo "2614"]
[BlackElo "2605"]
[Annotator "depth=26 eval=+097"]
[SetUp "1"]
[FEN "rnbqrbnk/pppppppp/8/8/8/8/PPPPPPPP/KNBRQBNR w - - 0 1"]
[PlyCount "12"]
[EventDate "2007.??.??"]

1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. Nc3 a6 5. h3 e6 6. e3 Nbd7 1/2-1/2
[/pgn]

So, no-castle chess is much better (less draws, more spectacular games) with these openings. And there is no need for special chess-rules (forbidding castlings in the normal chess starting positions needs a change in chess rules). Here, the kings are on the a- or h-line, so castling is illegal, using normal chess rules. So, no problem here for GUIs or engines). Humans, playing no-castle chess, can choose one of the 4 Drawkiller starting positions and play directly from there.
Download my AntiDrawOpenings here:
https://www.sp-cc.de/files/pohl_antidra ... gs_v1.5.7z

See the spectacular testing results on my website:
https://www.sp-cc.de/anti-draw-openings.htm

[AdminX]

Repeated test with Banksia GUI and Zahak and castle never showed up in analysis.

PS: I used the following PGN

Code: Select all

[Event "NC World Masters 2021"]
[Site "Dortmund"]
[Date "2021.07.18"]
[Round "4.1"]
[White "Kramnik, Vladimir"]
[Black "Anand, Viswanathan"]
[Result "1/2-1/2"]
[ECO "D32"]
[WhiteElo "2753"]
[BlackElo "2753"]
[Annotator "ChessBase"]
[PlyCount "79"]
[EventDate "2021.07.12"]
[EventType "match"]
[EventCountry "GER"]
[SourceTitle "playchess.com"]
[Source "ChessBase"]
[SourceQuality "1"]
[TimeControl "40/5400+30:1800+30"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w - - 0 1"]

1. c4 {5} e6 {7} 2. d4 {18} d5 {27} 3. Nc3 {13} c5 {11} 4. cxd5 {42} exd5 {9}
5. Nf3 {10} Be6 {14} 6. Bf4 {177} Nc6 {112} 7. e3 {68} a6 {297} 8. dxc5 {763}
Bxc5 {16} 9. Ng5 {309} Nf6 {144} 10. Nxe6 {196} fxe6 {9} 11. Bd3 {7} Bd6 {434}
12. Bg3 {221} Bxg3 {164} 13. hxg3 {313} Ke7 {17} 14. Qe2 {225} h5 {232} 15. Rd1
{202} Ne5 {417} 16. Bc2 {571} Qb6 {277} 17. Bb3 {35} Rad8 {44} 18. Rd4 $11 {
390 remains equal.} Nc6 {415} 19. Rdh4 {187} Ne5 {299} 20. g4 {360} Nexg4 {123}
21. Rxh5 $1 {225 Remove Defender} Rxh5 {394} 22. Rxh5 {124} Nxe3 {232} 23. Rg5
{121} d4 {268} 24. Rxg7+ {131} Kf8 {25} 25. Rg6 {32} dxc3 {131} 26. Qf3 $1 {453
} Qd4 {130} 27. Rxf6+ {201} Kg7 {9} 28. Rf7+ {52} Kg8 {10} 29. fxe3 {195} Qd2+
{10 Double Attack} 30. Kf1 {2} cxb2 {10} 31. Qg4+ {92 Double Attack} Kxf7 {5}
32. Qxe6+ {48} Kg7 {5} 33. Qe7+ {178} Kh8 {52} 34. Qf6+ {18} Kh7 {5} 35. Qf7+ {
27} Kh8 {6} 36. Qh5+ $2 {17} Kg7 {6} 37. Qg5+ {10} Kh7 {5} 38. Qh5+ {16} Kg7 {6
} 39. Qf7+ $11 {4} Kh8 {6} 40. Qh5+ {6} 1/2-1/2

[amanjpro]

Repeated test with Banksia GUI and Zahak and castle never showed up in analysis.

PS: I used the following PGN

Code: Select all

[Event "NC World Masters 2021"]
[Site "Dortmund"]
[Date "2021.07.18"]
[Round "4.1"]
[White "Kramnik, Vladimir"]
[Black "Anand, Viswanathan"]
[Result "1/2-1/2"]
[ECO "D32"]
[WhiteElo "2753"]
[BlackElo "2753"]
[Annotator "ChessBase"]
[PlyCount "79"]
[EventDate "2021.07.12"]
[EventType "match"]
[EventCountry "GER"]
[SourceTitle "playchess.com"]
[Source "ChessBase"]
[SourceQuality "1"]
[TimeControl "40/5400+30:1800+30"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w - - 0 1"]

1. c4 {5} e6 {7} 2. d4 {18} d5 {27} 3. Nc3 {13} c5 {11} 4. cxd5 {42} exd5 {9}
5. Nf3 {10} Be6 {14} 6. Bf4 {177} Nc6 {112} 7. e3 {68} a6 {297} 8. dxc5 {763}
Bxc5 {16} 9. Ng5 {309} Nf6 {144} 10. Nxe6 {196} fxe6 {9} 11. Bd3 {7} Bd6 {434}
12. Bg3 {221} Bxg3 {164} 13. hxg3 {313} Ke7 {17} 14. Qe2 {225} h5 {232} 15. Rd1
{202} Ne5 {417} 16. Bc2 {571} Qb6 {277} 17. Bb3 {35} Rad8 {44} 18. Rd4 $11 {
390 remains equal.} Nc6 {415} 19. Rdh4 {187} Ne5 {299} 20. g4 {360} Nexg4 {123}
21. Rxh5 $1 {225 Remove Defender} Rxh5 {394} 22. Rxh5 {124} Nxe3 {232} 23. Rg5
{121} d4 {268} 24. Rxg7+ {131} Kf8 {25} 25. Rg6 {32} dxc3 {131} 26. Qf3 $1 {453
} Qd4 {130} 27. Rxf6+ {201} Kg7 {9} 28. Rf7+ {52} Kg8 {10} 29. fxe3 {195} Qd2+
{10 Double Attack} 30. Kf1 {2} cxb2 {10} 31. Qg4+ {92 Double Attack} Kxf7 {5}
32. Qxe6+ {48} Kg7 {5} 33. Qe7+ {178} Kh8 {52} 34. Qf6+ {18} Kh7 {5} 35. Qf7+ {
27} Kh8 {6} 36. Qh5+ $2 {17} Kg7 {6} 37. Qg5+ {10} Kh7 {5} 38. Qh5+ {16} Kg7 {6
} 39. Qf7+ $11 {4} Kh8 {6} 40. Qh5+ {6} 1/2-1/2

So, for a chess non-varient, your FEN is technically invalid as it has no castling right, but the move counter is 1. That is why probably some GUIs do not respect the castling rights.

This fen might be better?

[d]rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w Qq - 8 5

[AdminX]

I made one more test run using Banksia GUI, but recorded it this time. Castle never shows up in game analysis unless you see something that I might have missed.

Not sure when Youtube makes higher quality video available, but here is the link.

[amanjpro]

Repeated test with Banksia GUI and Zahak and castle never showed up in analysis.

PS: I used the following PGN

Code: Select all

[Event "NC World Masters 2021"]
[Site "Dortmund"]
[Date "2021.07.18"]
[Round "4.1"]
[White "Kramnik, Vladimir"]
[Black "Anand, Viswanathan"]
[Result "1/2-1/2"]
[ECO "D32"]
[WhiteElo "2753"]
[BlackElo "2753"]
[Annotator "ChessBase"]
[PlyCount "79"]
[EventDate "2021.07.12"]
[EventType "match"]
[EventCountry "GER"]
[SourceTitle "playchess.com"]
[Source "ChessBase"]
[SourceQuality "1"]
[TimeControl "40/5400+30:1800+30"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w - - 0 1"]

1. c4 {5} e6 {7} 2. d4 {18} d5 {27} 3. Nc3 {13} c5 {11} 4. cxd5 {42} exd5 {9}
5. Nf3 {10} Be6 {14} 6. Bf4 {177} Nc6 {112} 7. e3 {68} a6 {297} 8. dxc5 {763}
Bxc5 {16} 9. Ng5 {309} Nf6 {144} 10. Nxe6 {196} fxe6 {9} 11. Bd3 {7} Bd6 {434}
12. Bg3 {221} Bxg3 {164} 13. hxg3 {313} Ke7 {17} 14. Qe2 {225} h5 {232} 15. Rd1
{202} Ne5 {417} 16. Bc2 {571} Qb6 {277} 17. Bb3 {35} Rad8 {44} 18. Rd4 $11 {
390 remains equal.} Nc6 {415} 19. Rdh4 {187} Ne5 {299} 20. g4 {360} Nexg4 {123}
21. Rxh5 $1 {225 Remove Defender} Rxh5 {394} 22. Rxh5 {124} Nxe3 {232} 23. Rg5
{121} d4 {268} 24. Rxg7+ {131} Kf8 {25} 25. Rg6 {32} dxc3 {131} 26. Qf3 $1 {453
} Qd4 {130} 27. Rxf6+ {201} Kg7 {9} 28. Rf7+ {52} Kg8 {10} 29. fxe3 {195} Qd2+
{10 Double Attack} 30. Kf1 {2} cxb2 {10} 31. Qg4+ {92 Double Attack} Kxf7 {5}
32. Qxe6+ {48} Kg7 {5} 33. Qe7+ {178} Kh8 {52} 34. Qf6+ {18} Kh7 {5} 35. Qf7+ {
27} Kh8 {6} 36. Qh5+ $2 {17} Kg7 {6} 37. Qg5+ {10} Kh7 {5} 38. Qh5+ {16} Kg7 {6
} 39. Qf7+ $11 {4} Kh8 {6} 40. Qh5+ {6} 1/2-1/2

So, for a chess non-varient, your FEN is technically invalid as it has no castling right, but the move counter is 1. That is why probably some GUIs do not respect the castling rights.

This fen might be better?

[d]rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w Qq - 8 5

Even that FEN was wrong, Q-side castle was still on, this one turns them both off:

[d]rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w - - 16 9

I made one more test run using Banksia GUI, but recorded it this time. Castle never shows up in game analysis unless you see something that I might have missed.

Not sure when Youtube makes higher quality video available, but here is the link.

Ah, will watch it later. But the thing is, as the FEN strictly speaking is not correct, probably each engine/UI interprets it as they wish

[Chessqueen]

I remembered when I asked the same question back 15 years or 16 years ago and 3 months later CCRL started testing Chess960(FRC)

Note: I also noticed that in Chess960(FRC) Most top 3 Engines that start with 1.d4 in CCRL out of 64 of the 112 CCRL games White has a a success rate of 75%, How come ?

It is very easy to start a non castling chess tournament among Engines
All that is needed is this FEN which does not allow castling ===> rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w - - 0 1
NOT the normal FEN which allows castling==================> rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1

I've just tried this no castling FEN in the Fritz 17 GUI.
Also through the set up position screen with castling options turned off.

In both cases ran Stockfish 14 with infinite analysis, castling moves appeared in the anaysis.

Wondering if this is a problem with the GUI or the engine?

Andrew

It could be that different GUI handle the NON Castling Chess different
Under Arena GUI click Position ==> Set up position then ==> under Castling Possibilities disable all castling for long and short for the White and Black side click Okay ==>and under GAME select DEMO

[pgn][Event "Computer chess game"]
[Site "MININT-UB2PIMJ"]
[Date "2021.07.07"]
[Round "?"]
[White "Stockfish_21070310_x64_bmi2"]
[Black "Stockfish_21070310_x64_bmi2"]
[Result "1/2-1/2"]
[BlackElo "3520"]
[ECO "C47"]
[Opening "Four Knights"]
[Time "18:33:46"]
[Variation "Scotch, Main Line, 7...d5"]
[WhiteElo "3520"]
[TimeControl "60+0"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w - - 0 1"]
[Termination "adjudication"]
[PlyCount "130"]
[WhiteType "program"]
[BlackType "program"]

1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6 4. d4 exd4 5. Nxd4 Bb4 6. Nxc6 bxc6 7. Bd3
d5 8. Qe2 Bxc3+ 9. bxc3 dxe4 10. Bxe4 Nxe4 11. Qxe4+ Qe7 12. Qxe7+ Kxe7 13.
Bf4 Re8 14. Rd1 Rb8 15. Kd2 Be6 16. a3 Kd7 17. Kc1+ Kc8 18. Rhe1 Rb5 19.
Rd4 Ra5 20. Ree4 Rxa3 21. Ra4 Rxa4 22. Rxa4 Bd5 23. Rxa7 Re4 24. Rxc7+ Kd8
25. Bg3 Re7 26. Rxe7 Kxe7 27. f3 Bc4 28. Kd2 Bf1 29. Bh4+ f6 30. g4 g5 31.
Bf2 f5 32. gxf5 Kf6 33. Ke3 Kxf5 34. Bg3 h5 35. Bc7 Bc4 36. Bd8 Bb5 37. Be7
Bf1 38. c4 Bxc4 39. Kd4 Bb5 40. Ke3 Ba4 41. c3 Bb3 42. Bd8 Bd5 43. Bc7 Kg6
44. Kf2 Ba2 45. Bd6 Kf5 46. Ba3 g4 47. Bf8 Bc4 48. fxg4+ Kxg4 49. Be7 Kh3
50. Bd6 Bb3 51. Bc7 Bc4 52. Bd6 Bb3 53. Bc7 Bd1 54. Ke3 Bc2 55. Be5 Bg6 56.
Kd4 Bf7 57. Bd6 Bd5 58. Bc7 Bg2 59. Kc5 Kg4 60. Kd4 Kh3 61. Bd6 Bh1 62. Kc5
Bf3 63. Kb6 Bg2 64. Kb7 Bd5 65. Kb8 Bf3 {User Adjudication} 1/2-1/2[/pgn]

[Chessqueen]

These are the 4 starting positions of my Drawkiller openings. If anybody wants to play chess without castling, these openings are much better for interesting games.
In my AntiDraw openings collection (in the Drawkiller V5 folder), you find unbalanced Drawkiller openings, starting from these 4 positions. Some with pawn-moves, only, and others followed by classical human opening moves. All analyzed and evaluated.

Example line (Drawkiller V5, dk_+0.90_+0.99_human_6mvs.pgn-File):
[pgn]
[Event "Valjevo"]
[Site "Valjevo"]
[Date "2007.06.20"]
[Round "8"]
[White "Ivanisevic, Ivan"]
[Black "Roiz, Michael"]
[Result "1/2-1/2"]
[ECO "D15"]
[WhiteElo "2614"]
[BlackElo "2605"]
[Annotator "depth=26 eval=+097"]
[SetUp "1"]
[FEN "rnbqrbnk/pppppppp/8/8/8/8/PPPPPPPP/KNBRQBNR w - - 0 1"]
[PlyCount "12"]
[EventDate "2007.??.??"]

1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. Nc3 a6 5. h3 e6 6. e3 Nbd7 1/2-1/2
[/pgn]

So, no-castle chess is much better (less draws, more spectacular games) with these openings. And there is no need for special chess-rules (forbidding castlings in the normal chess starting positions needs a change in chess rules). Here, the kings are on the a- or h-line, so castling is illegal, using normal chess rules. So, no problem here for GUIs or engines). Humans, playing no-castle chess, can choose one of the 4 Drawkiller starting positions and play directly from there.
Download my AntiDrawOpenings here:
https://www.sp-cc.de/files/pohl_antidra ... gs_v1.5.7z

See the spectacular testing results on my website:
https://www.sp-cc.de/anti-draw-openings.htm

I took the challenge using the Expert Level of Honey 14 and honestly to play a non Castling game against a computer of your level is like playing against alien from another Planet. I did not wanted to continue since I did NOT feel comfortable and none of the positions were familiar to me, we humans become familiar playing certain format and Pattern recognition, so this type of Non Castling Chess is even harder than playing Chess960 (FRC}. Uri and Mr. Kaufman can give it a try using this position and setting level Expert for Uri, and Master for Mr. Kaufman and see what I mean .


[pgn][Event "Computer chess game"]
[Site "DESKTOP-OFQ3C0P"]
[Date "2021.07.19"]
[Round "?"]
[White "ChessQueen"]
[Black "Honey-v14-x86-64-modern"]
[Result "*"]
[BlackElo "2000"]
[ECO "Shuffl"]
[Opening "e Chess"]
[Time "16:17:03"]
[WhiteElo "2000"]
[TimeControl "1200+3"]
[SetUp "1"]
[FEN "rnbqrbnk/pppppppp/8/8/8/8/PPPPPPPP/KNBRQBNR w - - 0 1"]
[Termination "unterminated"]
[PlyCount "35"]
[WhiteType "human"]
[BlackType "program"]

1. d4 Nc6 2. d5 Nb8 3. Nf3 Nf6 4. e4 d6 5. e5 dxe5 6. Nxe5 Kg8 7. Bf4 Bf5
8. Bb5 Nbd7 9. Qc3 Nxe5 10. Bxe5 Nxd5 11. Qf3 c6 12. Bxc6 bxc6 13. Qxf5 e6
14. Qe4 f6 15. Bd4 c5 16. Bc3 Rb8 17. b3[/pgn]