Moderators: hgm, Rebel, chrisw
11 years ago, HIARCS found the solution after 835 millions nodes : http://www.talkchess.com/forum/viewtopi ... 285#642285Sean Evans wrote:Ah, so for the first time CCC members and their chess engines cannot solve a chess puzzleVinvin wrote:It's mate in 39 (77 half-plies), http://www.talkchess.com/forum/viewtopi ... 978#710978 Out of reach for current engines.
...
I do not see 39 moves in your Hiarcs solution to mate.Vinvin wrote:11 years ago, HIARCS found the solution after 835 millions nodes : http://www.talkchess.com/forum/viewtopi ... 285#642285
A non buggy engine will find all the solutions of all the possible puzzles. Only the time matters.
Hi, played through your moves and it does come to mate. The problem we have is the database answer posted on this thread has 39 moves, not 14 moves. Move 4...Kg4 for Black seems to be a dilemma to your 14 move solutionArpad Rusz wrote:The study is in the HHdbV (actually there are even five versions of it in the database).
[Event "Schakend Nederland#2207 (v)"]
[Site "?"]
[Date "1990.??.??"]
[Round "?"]
[White "Van Breukelen=G"]
[Black "(+0047.24d6h7)"]
[Result "1-0"]
[SetUp "1"]
[FEN "8/3P3k/n2K3p/2p3n1/1b4N1/2p1p1P1/8/3B4 w - - 0 1"]
[PlyCount "27"]
[EventDate "1990.??.??"]
{HHdbV#23850} 1. Nf6+ Kg7 2. Nh5+ Kg6 (2... Kh8 3. d8=Q+) 3. Bc2+ Kxh5 4. d8=Q $1 Nf7+ (4... Kg4 5. Qf6 Kxg3 6. Qe5+ Kf2 7. Qh2+ Ke1 8. Qg1+ Ke2 9. Qg4+ Ke1 10. Ke5) 5. Ke6 Nxd8+ 6. Kf5 e2 7. Be4 e1=N $1 8. Bd5 $1 c2 9. Bc4 c1=N $1 10. Bb5 Nc7 11. Ba4 $1 Ne2 12. Bd1 Nf3 13. Bxe2 Nce6 14. Bxf3# 1-0
Of course you are right as for the thematic solution, Arpad (even if it's 15 moves, not 14, that's wrong in several solution sites in internet too).Yet the longest way to mate might be of interest too, especially if calculable, don't you think so too?Arpad Rusz wrote:The solution of the study has a main line of only 14 moves. All white moves are unique in this line. The variation starting with 4...Kg4 is only a technical one, it is sufficient to know that white will win somehow. White's answer to 4...Kg4 is not unique so this variation cannot even be part of the study's solution!
You did a great job Martan, keep up the good workpeter wrote:Of course you are right as for the thematic solution, Arpad (even if it's 15 moves, not 14, that's wrong in several solution sites in internet too).Yet the longest way to mate might be of interest too, especially if calculable, don't you think so too?
And one cannot demand from engine to give a solution to a position as a mating output, if the most beautiful way to mate is 24 moves shorter than the best defence is. Showing the mate in 15 as solution to the position as primary line would be simply wrong too.
It's a great study, but you must not demand to see 4...Nf7+ in engine- output because seeing it not from the human point of view as for beauty and thematic clearness only, 4...Nf7+ is simply not the best black move here.
BTW, do you know if Gijs van Breukelen ist the author of this very beautiful piece
All those so-called chessmasters with numerous YouTube videos on this puzzle all now proven wrong! Gijs van Breukelen created the puzzle around 1970. Breukelen published the puzzle in the Netherlands chess magazine, Schakend Nederland, in 1997. You deserve a CCC medal 
I think there is a problem on the initial setup. I think there needs to be a White pawn on h3. If the h3 pawn is there the original 14 move mate might work. I would have to run an engine on it to see.peter wrote:Thanks Sean!
In my database I changed my .pgn according to author's name and solution- variant, which of course has to be the main line in the study's sense yet.
Arpad is simply right about that, and I am a little engine-damgaged as for chess in my old days.
