|
| Page: [Previous] [Next] 1 2 3 |
| (21) Posted by Joost de Heer [Thursday, Aug 17, 2023 21:34] |
Codex Article 17: 50 Moves-Rule
Unless expressly stipulated, the 50 moves-rule does not apply to the solution of chess compositions except for retro-problems.
So for studies, unless explicitly stated, the 50 move rule does NOT apply.
|
|
| (22) Posted by Kevin Begley [Thursday, Aug 17, 2023 21:39] |
@Joost,
That's my whole point -- if it can be expressly stipulated, it should be an option.
Right?
The default position is clear. No argument there.
|
|
| (23) Posted by Mu-Tsun Tsai [Friday, Aug 18, 2023 11:05] |
The Syzygy tablebase tool is now on! It is capable of handling all studies with up to 7 pieces.
A few notes:
1. Through experiments, I realized that "choosing the Black defense that leads to a unique White continuation" is not always the best strategy, as such defenses can sometimes be completely trivial. Instead, I judge the quality of a defense both by the number of White continuations and the effectiveness of the defense itself (i.e. how well it prolongs the game).
2. To keep the API usage under control and not getting blocked by lichess, the tool is limited to make 1 API request per second. This of course slows down the search quite a bit, but better than getting blocked and have to wait for one minute.
Enjoy!
|
|
| (24) Posted by Viktoras Paliulionis [Friday, Aug 18, 2023 12:12] |
I get the error message "Unable to connect to the server. Please check your network connection."
|
|
| (25) Posted by Mu-Tsun Tsai [Friday, Aug 18, 2023 12:45] |
@(24)
Should be fixed now, sorry about that!
|
|
| (26) Posted by Viktoras Paliulionis [Friday, Aug 18, 2023 13:25] |
Thank You!
I try to solve this study:
https://yacpdb.org/#580894
 (= 2+4 )
=
Solution:
1.Kc6! Kg5! 2.Sf7+ Kf6 3.Sd8 Ke7 4.Kxc7! Sd5+ 5.Kc6 Sb4+ 6.Kc7 Sa6+ 7.Kc8 Bg2! 8.Sc6+! Bxc6 stalemate
The tool hasn't finished the calculation yet, but I suspect it also analyzes branches with duals. Correct studies usually do not have duals, so such not unique branches should not be analyzed.
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Kxd8 6. Kxd5=! Kd7=! 7. Kd4= Kd6=! 8. Kc3= Kc5=! 9. Kb2= ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nf6 6. Nb7! Bg2+! 7. Kb6 Bxb7! 8. Kxb7=! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nc3 6. Nb7! Bg2+! 7. Kb6 Bxb7! 8. Kxb7=! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Be2 8. Nc6+! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bd3 8. Nc6+! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bc4 8. Nc6+! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bb5 8. Nc6+! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bg2 8. Nc6+! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bh3+ 8. Kb7! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Nd5+ 7. Kc6! Kxd8 8. Kxd5=! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Nd5+ 7. Kc6! Nf6 8. Nb7! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Nd5+ 7. Kc6! Nc3 8. Nb7! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Nd5+ 7. Kc6! Nb4+ 8. Kc7! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Nd5+ 7. Kc6! Bg2 8. Nb7! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Bg2 6. Nb7! Nf6+ 7. Kb6 Bxb7! 8. Kxb7=! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Bg2 6. Nb7! Nf4+ 7. Kb6 Bxb7! 8. Kxb7=! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Bg2 6. Nb7! Ne3+ 7. Kb6 Bxb7! 8. Kxb7=! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Bg2 6. Nb7! Nc3+ 7. Kb6 Bxb7! 8. Kxb7=! ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Nd5 4. Kxd5! Bg2+! 5. Kc5! Ke7! 6. Nc6+! Bxc6! 7. Kxc6! Ke6! 8. Kxc7= ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Be2 4. Kxc7!
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Bc4 4. Kxc7!
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ba6 4. Kxc7!
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Bg2+ 4. Kxc7!
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Bh3 4. Kxc7!
1. Kc6! Kh5 2. Nf7! Nd5 3. Kxd5!
1. Kc6! Kh5 2. Nf7! Bg2+ 3. Kxc7!
1. Kc6! Nd5 2. Kxd5!
1. Kc6! Bc4 2. Ng6+!
1. Kc6! Bg2+ 2. Kxc7! Kg5 3. Nf7+! Kf6! 4. Nd6 Ke7 5. Nb5 Nd5+! 6. Kc6 Nc3+! 7. Kc5 Nxb5! 8. Kxb5=! ...
1. Kc6! Bg2+ 2. Kxc7! Kg5 3. Nf7+! Kf6! 4. Nd6 Ke6 5. Nb5 Nd5+! 6. Kc6 Nc3+! 7. Kc5 Nxb5! 8. Kxb5=! ...
1. Kc6! Bg2+ 2. Kxc7! Bd5 3. Ng6+! Kg5! 4. Ne5 Kf6 5. Nd3 Be4 6. Nc1 Ke7! 7. Ne2 Nd5+! 8. Kb8 ...
1. Kc6! Bg2+ 2. Kxc7! Bd5 3. Ng6+! Kg5! 4. Ne5 Kf6 5. Nd3 Bc4 6. Ne1 Kf5 7. Nf3! Kf4 8. Nd2 ...
1. Kc6! Bg2+ 2. Kxc7! Bd5 3. Ng6+! Kg5! 4. Ne5 Kf6 5. Nd3 Bc4 6. Ne1 Kf5 7. Nf3! Be2 8. Nd2 ...
1. Kc6! Bg2+ 2. Kxc7! Bd5 3. Ng6+! Kg5! 4. Ne5 Kf6 5. Nd3 Bc4 6. Ne1 Ke5 7. Nf3+! Kf4! 8. Nd2 ...
1. Kc6! Bg2+ 2. Kxc7! Bd5 3. Ng6+! Kg5! 4. Ne5 Kf5 5. Nd3 Bc4! 6. Nb2 Bb5! 7. Kb6 Nc4+ 8. Kxb5 ...
1. Kc6! Bg2+ 2. Kxc7! Bd5 3. Ng6+! Kg5! 4. Ne5 Kf5 5. Nd3 Bc4! 6. Nb2 Bb5! 7. Kb6 Nd5+ 8. Kxb5 ...
|
|
| (27) Posted by Viktoras Paliulionis [Friday, Aug 18, 2023 13:57] |
Current solving state:
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Kxd8 6. Kxd5=! Kd7=! 7. Kd4= Kd6=! 8. Kc3= Kc5=! 9. Kb2= Kb4=! 10. Ka1= ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nf6 6. Nb7! Bg2+! 7. Kb6 Bxb7! 8. Kxb7=! Kd7=! 9. Ka6= Kc6=! 10. Ka5= ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nc3 6. Nb7! Bg2+! 7. Kb6 Bxb7! 8. Kxb7=! Kd7=! 9. Ka6= Kc6= 10. Ka5= ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nc3 6. Nb7! Bg2+! 7. Kb6 Bxb7! 8. Kxb7=! Kd7=! 9. Ka6= Kc7= 10. Ka5= ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Be2 8. Nc6+! Kd6! 9. Nd4 Bg4+! 10. Kb7 ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bd3 8. Nc6+! Kd6! 9. Nd4 ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bc4 8. Nc6+! Kd6! 9. Nd4 ...
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bb5 8. Nc6+!
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bg2 8. Nc6+!
...
(More than 60 lines)
|
|
| (28) Posted by Mu-Tsun Tsai [Friday, Aug 18, 2023 14:02] |
Ah, the problem is that the search doesn't stop when it reaches draw! You can stop it for now. I will fix it when I have the time.
|
|
| (29) Posted by Viktoras Paliulionis [Friday, Aug 18, 2023 15:24] |
I think the following line could be stopped after White's 9th move because it is not unique:
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Be2 8. Nc6+! Kd6! 9. Nd4 (stop here!) Bg4+! 10. Kb7 ...
It would be enough to write, for example, like this:
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Be2 8. Nc6+! Kd6! 9. Nd4 (not unique)
I didn't understand what the exclamation mark means in the analysis? He could mean that this is a unique move. For example, for draws, it could mean that other moves lead to loss.
2... Kf6! is with the exclamation mark, but it is not the only move that would result in a draw.
It would be nice if black's last move was indicated for stalemate:
1. Kc6! Kg5 2. Nf7+! Kf6! 3. Nd8! Ke7 4. Kxc7! Nd5+! 5. Kc6! Nb4+ 6. Kc7! Na6+ 7. Kc8! Bb5 8. Nc6+! Bxc6 (stalemate)
|
|
| (30) Posted by Mu-Tsun Tsai [Friday, Aug 18, 2023 17:04] |
> I think the following line could be stopped after White's 9th move
That behavoir was inspired by this study:
 (= 2+2 )
where the mainline is
1. Kc4! a2! 2. Kb3! a1=N+! 3. Kc3 Nc2! 4. Re5 Kc1! 5. Re2 Kd1! 6. Rxc2 Ke1! 7. Kd3 Kd1! 8. Rc4 Ke1! 9. Rf4 Kd1! 10. Rf1#
and 3.Kc3 is indeed not unique (if I choose a defense of Black that leads to a unique White move at move 2, it will be totally trivial). That's why I don't choose the defense solely on the uniqueness.
But I will agree that this mechanism only make sense in winning problems. For drawing problem I shall stick to uniqueness criteria.
> I didn't understand what the exclamation mark means in the analysis
For Black's move, it basically means "this is the only defense worth further discussion", based on the criteria for choosing defense moves.
> It would be nice if black's last move was indicated for stalemate
The problem here is that, for this particular problem, I don't see why Black should favor a stalemate move over other defenses that could prolong the game. I can imagine that in general that wouldn't be the best thing to do for Black, and this problem is more like an exception.
|
|
| (31) Posted by Mu-Tsun Tsai [Friday, Aug 18, 2023 17:34] |
Just made another update. The logic should be much improved now. I also implement transposition table to avoid wasting time searching the same position.
For that particular problem, now it will complete the search and the output will be:
1. Kc6!
1... Kg5 2. Nf7+! Kf6! 3. Nd8!
3... Ke7 4. Kxc7! Nd5+! 5. Kc6!
5... Kxd8 6. Kxd5=!
5... Nf6 6. Nb7!
5... Nc3 6. Nb7!
5... Nb4+ 6. Kc7!
6... Na6+ 7. Kc8!
7... Be2 8. Nc6+!
7... Bd3 8. Nc6+!
7... Bc4 8. Nc6+!
7... Bb5 8. Nc6+!
7... Bg2 8. Nc6+!
7... Bh3+ 8. Kb7!
8... Kxd8 9. Kxa6=!
8... Bc8+ 9. Kxc8!
6... Nd5+ 7. Kc6! (transpose)
5... Bg2 6. Nb7!
3... Nd5 4. Kxd5! Bg2+! 5. Kc5! Ke7! 6. Nc6+! Bxc6! 7. Kxc6!
3... Be2 4. Kxc7!
3... Bc4 4. Kxc7!
3... Ba6 4. Kxc7!
3... Bg2+ 4. Kxc7!
3... Bh3 4. Kxc7!
1... Kh5 2. Nf7!
2... Nd5 3. Kxd5!
2... Bg2+ 3. Kxc7!
1... Nd5 2. Kxd5!
1... Bc4 2. Ng6+!
1... Bg2+ 2. Kxc7!
2... Kg5 3. Nf7+!
2... Bd5 3. Ng6+!
which basically includes the intended mainline.
|
|
| (32) Posted by Viktoras Paliulionis [Monday, Aug 21, 2023 03:27] |
@(30)
Your example is not a study, because a correct study must have a single solution and no duals in the main variation.
An EGTB analysis tool would be useful for composers and solvers if it could check the studies for soundness. Ideally, it should show a solution (that matches the author's), and if the solution is not unique, then show duals. Your position has duals, but the tool doesn't show them. Therefore Stockfish analysis of this position is more useful.
BTW, why does the tool sometimes show only the first move and stop? Eg.: https://www.yacpdb.org/#294437
|
|
| (33) Posted by Kevin Begley [Monday, Aug 21, 2023 05:06] |
@(30) is by Philippe Stamma, Essai sur le jeu des Echecs, 1737.
The line given has several duals starting on move 4 (some faster than the line given), but the line given extends well beyond Stamma's solution (assuming the solution given by Win Chloe was the author's intent).
I'm looking through the PDF to confirm this, but it will take time to locate the problem (no diagrams are provided, only text).
Win Chloe (I believe incorrectly) reports a dual starting on move 3.
1.Rç4 a2 2.Rb3 a1=D 3.Té1‡
2…a1=C+ 3.Rç3,Ra3 (dual)
This is NOT a dual. The move 3.Ka3 is nothing but a waste of time, since after 3.Ka3 ...Sc2+, the only win is 4.Kb3, and black can immediately transpose to the previous diagram via 4...Sa1+, when white must find 5.Kc3! (which should be considered the only winning move on move 3).
If Stamma gave 3.Ka3, I suspect his solution was likely a typo (in which case, Win Chloe would be correct to report the problem has a dual).
If Stamma gave 3.Kc3, his solution is entirely correct (and Win Chloe should not consider this a dual).
Update:
The PDF is here: https://ruchess.ru/upload/iblock/4b6/4b6811c0ea18e3fab5513d91661e0167.pdf
I believe this problem may appear on Page 55, but I am unable to discern the solution.
Maybe somebody can confirm. My best guess: I was wrong to assume the solution given by Win Chloe was the author's intent.
Needless to say, duals in study solutions were hardly recognized as a blemish back in 1737.
|
|
| (34) Posted by Joose Norri [Monday, Aug 21, 2023 05:28] |
Posts 29&30: > It would be nice if black's last move was indicated for stalemate
The problem here is that, for this particular problem, I don't see why Black should favor a stalemate move over other defenses that could prolong the game. I can imagine that in general that wouldn't be the best thing to do for Black, and this problem is more like an exception.>>>
It's the other way round. Bxc6 is the only move that prevents a general draw. In some studies, which are the exception, prolonging the "game" produces noteworthy points or even the point of the study.
If prolonging the solution by not accepting the stalemating sacrifice leads to another unique White move, that does not necessarily mean that that branch of the solution is worth noting.
|
|
| (35) Posted by Mu-Tsun Tsai [Monday, Aug 21, 2023 07:53] |
@(32)
> Your example is not a study, because a correct study must have a single solution and no duals in the main variation.
That is by today's standard. The problem was from 1737.
> BTW, why does the tool sometimes show only the first move and stop?
Because you just discovered another bug in the algorithm :P
I just fixed it and it works now.
|
|
| (36) Posted by Mu-Tsun Tsai [Monday, Aug 21, 2023 07:57] |
@(34)
> It's the other way round. Bxc6 is the only move that prevents a general draw.
However, Syzygy doesn't have the notion of a general draw (except for insufficient material which is trivial), so I can't build any logic upon that reasoning.
|
|
| (37) Posted by Joost de Heer [Monday, Aug 21, 2023 12:49] |
Many years ago there was a controversy during the Dutch solving championship, where according to the author (and the tournament director) the main line in a study was a move that allowed a #1, while the solvers chose a different, longer sequence, claiming 'of course allowing a #1 can't be the main variation, black is not playing *that* bad'.
|
|
| (38) Posted by Joost de Heer [Monday, Aug 21, 2023 13:09] |
QUOTE
Update:
The PDF is here: https://ruchess.ru/upload/iblock/4b6/4b6811c0ea18e3fab5513d91661e0167.pdf
I believe this problem may appear on Page 55, but I am unable to discern the solution.
It's problem 15 (Quinzieme partie).
Black: K on the square of the white queen's knight
P on the 5th square of his queen's rook
White: K on the 5th square of his queen
R on the 4th square of his king
The solution given:
1. K to the 4th square of his queen's bishop
Pawn one step
2. K to the 3rd square of his queen's knight
Pawn promotes to queen
3. R to the square of his king, delivering check and mate
First variation:
1. K to the 4th square of his queen's bishop
Pawn one step
2. K to the 3rd square of his queen's knight
Pawn promotes to queen but demands a knight giving check
3. K to the 3rd square of his queen's bishop
S to the 2nd square of the white queen's bishop
4. R to the 2nd square of his king
S to the 3rd square of the white queen's rook
5. K to the 3rd square of the white queen's knight
S to the 2nd square of the white queen's bishop
6. R captures S and wins the game
Second variation
1. K to the 4th square of his queen's bishop
Pawn one step
2. K to the 3rd square of his queen's knight
Pawn promotes to queen but demands a knight giving check
3. K to the 3rd square of his queen's bishop
S to the 2nd square of the white queen's bishop
4. R to the 2nd square of his king
S to the 1st square of the white queen's rook
5. R to the 2nd square of his king's bishop
K to the square of the white queen's bishop
6. R gives check and mate
Or translated to modern notation:
W: Kd5 Re4
B: Kb1 Pa3
1. Kc4 a2 2. Kb3 a1=Q (2... a1=S 3. Kc3 Sc2 4. Re2 Sa3 (4... Sa1 5. Rf2 Kc1 6. Rf1#) 5. Ka3 Sc2 6. Rxc2 win) 3. Re1#
|
|
| (39) Posted by Joose Norri [Monday, Aug 21, 2023 20:10] |
@(34)
> It's the other way round. Bxc6 is the only move that prevents a general draw.
However, Syzygy doesn't have the notion of a general draw (except for insufficient material which is trivial), so I can't build any logic upon that reasoning.
Exactly. You can't, you need a human assessor. In studies it can be only a very useful tool. Syzygy is not the gold standard.
|
|
| (40) Posted by Joose Norri [Monday, Aug 21, 2023 20:15] |
And Joost's point, post 37, is what it means. In a study allowing a mate in one is the main line, if other moves result in a general win. (Naturally there are exceptions.)
|
|
Read more... |
Page: [Previous] [Next] 1 2 3
MatPlus.Net  Forum General FEN Tool for chess composers and solvers |
ISC 2024
