Fugitive King puzzle

Discussion and analysis about certain positions.

Fugitive King puzzle

Postby Alex_Moiseyev on Sat Jan 21, 2006 5:03 pm

This puzzle was composed originally by Brian Hinkle while analysing the game. Article about "Fugitive King" with solution will be included in the coming soon book SIXTH. You are welcome to use programs to solve this puzzle, but I really doubt they help you much, if you don't have ending base more than 8 pieces.

Image
Either play, white wins

Regards,

Alex
I am playing checkers, not chess.
User avatar
Alex_Moiseyev
 
Posts: 4091
Joined: Sat Nov 12, 2005 5:03 pm

Re: Fugitive King puzzle

Postby matthewkooshad on Sat Jan 21, 2006 6:41 pm

Alex, do you explain the solution in your book?
User avatar
matthewkooshad
 
Posts: 289
Joined: Tue Nov 15, 2005 3:08 pm
Location: Mississippi, USA

Re: Fugitive King puzzle

Postby Alex_Moiseyev on Sat Jan 21, 2006 8:15 pm

Hi, Matthew ! Good question, thanks :P Article about Figutive King was written by Mr. Hinkle for my book. Like regular game, it has solution (about 40-50 moves) and comments. His comments to solution explains :kermit: the main ideas in position. The whole article with diagram occupies 1 page.

Regards,

:angel13: Alex Moiseyev
I am playing checkers, not chess.
User avatar
Alex_Moiseyev
 
Posts: 4091
Joined: Sat Nov 12, 2005 5:03 pm

Re: Fugitive King puzzle

Postby matthewkooshad on Sun Jan 22, 2006 1:06 am

Interesting, thanks for the reply :)
User avatar
matthewkooshad
 
Posts: 289
Joined: Tue Nov 15, 2005 3:08 pm
Location: Mississippi, USA

Re: Fugitive King puzzle

Postby Ed Gilbert on Sun Jan 22, 2006 9:38 am

Hi Alex,

shr wrote:You are welcome to use programs to solve this puzzle, but I really doubt they help you much, if you don't have ending base more than 8 pieces.


I set kingsrow 1.15q up with an 8-piece db and let it play the strong side of this fugitive king ending against another instance of kingsrow defending the losing side using a 10-piece db. At 15sec/move the 8-piece instance playing white was able to win the game. Here is the game as it played out with white making the first move. Note that the 29-25 pitch at white's first move is the only move to win.

[Event "Fugitive king ending"]
[Black "kingsrow/10-piece"]
[White "kingsrow/8-piece"]
[Result "0-1"]
[Setup "1"]
[FEN "W:W32,29,23,20,9:BK22,16,12,11,8"]
1. 29-25 22x29 2. 9-6 29-25 3. 6-2 25-22 4. 2-7 22-26 5. 23-18 26-23 6. 18-14 23-18 7. 14-9
18-14 8. 9-6 14-9 9. 6-1 9-14 10. 1-6 14-18 11. 32-27 18-22 12. 27-23 22-26 13. 23-18 26-22
14. 18-14 22-18 15. 14-9 18-23 16. 9-5 23-19 17. 5-1 19-23 18. 1-5 23-19 19. 5-9 19-15 20.
9-14 15-19 21. 14-18 19-15 22. 18-22 15-19 23. 6-9 19-15 24. 9-13 15-19 25. 13-17 19-15
26. 17-21 15-19 27. 21-25 19-23 28. 25-30 23-27 29. 22-18
{8-piece kingsrow now sees a database win} 27-31 30. 30-25 31-27 31. 25-22 27-24 32.
18-23 24-19 33. 22-26 19-24 34. 26-30 24-28 35. 7-10 28-24 36. 10-14 24-28 37. 14-18 28-24
38. 18-22 24-28 39. 22-26 28-24 40. 26-31 24-28 41. 31-27 28-24 42. 27-32 24-27 43. 23-18
27-24 44. 32-28 16-19 45. 18-23 19x26 46. 30x23 24-27 47. 23x32 0-1


-- Ed
Ed Gilbert
 
Posts: 129
Joined: Mon Nov 14, 2005 7:37 am
Location: Morristown, New Jersey

Re: Fugitive King puzzle

Postby Alex_Moiseyev on Sun Jan 22, 2006 11:34 am

Hi, Ed !

Thanks for correct solution ! I know - 15sec/move is alot :lol: for KR, and I congratulate program. My Nemesis with 8 pieces only and "home-PC" :roll: can't solve this at reasonable time, because I may need many hours and days to search 100 (!!!) plies ahead.

29-25 change a move. If, instead, you play 9-6 22-26 6-2 26-19 2-7 - this is a whole different story !

The winning plan KR choose is absolutely correct. But solution in the book at the very end, when we have trade is slight different to demonstrate beauties.

It is an amazing fact, that program made a pitch w/o seem everything to the end :idea:

Regards,

Alex
I am playing checkers, not chess.
User avatar
Alex_Moiseyev
 
Posts: 4091
Joined: Sat Nov 12, 2005 5:03 pm

Re: Fugitive King puzzle

Postby Ed Gilbert on Sun Jan 22, 2006 12:01 pm

shr wrote:Thanks for correct solution ! I know - 15sec/move is alot :lol: for KR, and I congratulate program. My Nemesis with 8 pieces only and "home-PC" :roll: can't solve this at reasonable time, because I may need many hours and days to search 100 (!!!) plies ahead.


Kingsrow doesn't need a fast PC or need to see hundreds of plies ahead to solve this because it has heuristics that know about this class of 4x2 cramps. The heuristics guided the search for the first 30 moves until it saw the database win on a 15sec search. Brian Hinkle helped me develop these heuristics for the program.

-- Ed
Ed Gilbert
 
Posts: 129
Joined: Mon Nov 14, 2005 7:37 am
Location: Morristown, New Jersey

Re: Fugitive King puzzle

Postby Patrick Parker on Mon Mar 27, 2006 4:18 pm

this is off topic but what are heuristics
User avatar
Patrick Parker
 
Posts: 901
Joined: Tue Nov 15, 2005 11:32 pm
Location: amite, louisiana

Re: Fugitive King puzzle

Postby Ed Gilbert on Mon Mar 27, 2006 5:00 pm

Patrick Parker wrote:this is off topic but what are heuristics


A heuristic is an algorithm for solving a problem. It usually means an algorithm that is not necessarily perfect or optimal, but one that "gets the job done". In this context, it means that kingsrow has an algorithm that understands some things about this class of 4x2 cramps. It doesn't mean that it has perfect knowledge of every position that exhibits this formation, but it knows enough to understand that it has to pitch a man in this particular position, something that it otherwise would not do. It can also successfully finish the win, something it would not be able to do without the heuristic.

-- Ed
Ed Gilbert
 
Posts: 129
Joined: Mon Nov 14, 2005 7:37 am
Location: Morristown, New Jersey


Return to Positions

Who is online

Users browsing this forum: No registered users and 1 guest

cron
class=