Problem 464 not best option

So after 32-27, 18-15, 27-32, I would do 24-28. The trace of 15-19 is even better and ends up with 5 pieces against 2. 24-28 still gives 5 against 3, but the ‘correct’ move 22-18 will leave you with 4 pieces against 4 pieces, with no clear victory.
I think the engine wrongly evaluated the worth of kings in this case.

Hi arjanboogaart,
Would you please put in your PDN files for the two lines of play you considered (24-28 and the alternate 15-19). I had Kingsrow play against Kingsrow with the 9 piece endgame database (haven’t finished the 10 piece database yet) and it came up with the same play as in Checkercruncher.
Thanks,
Lloyd

I’m definitely interested different moves are better than the problem solution, it might indicate a bug. But I think I’m going to disable this problem either way. It asks for a nice tactics but there are a lot of ways to win this one. Thanks for pointing it out.

Here’s Kingsrow with the 9 piece endgame database. I wouldn’t disable this problem. Getting to the winning position was accomplished with a neat sequence of moves. The end game was also very instructive. I wouldn’t get rid of any puzzles unless there are clear errors.

[FEN “W:WK18,K20,22,K24,26,30:B7,13,17,21,K27,K31”]

  1. 18-15 27-32 2. 22-18 31x22 3. 15-19 22x15 4. 19x3 17-22 5. 3-7 22-25 6. 7-10 25-29 7.
    24-28 29-25 8. 20-24 25-22 9. 10-14 22-25 10. 14-18 13-17 11. 18-14 25-29 12. 14-9 17-22
  2. 9-14 22-26 14. 30x23 21-25 15. 14-17 25-30 16. 17-22 30-25 17. 23-18 25-30 18. 24-19
    32-27 19. 19-23 27-32 20. 28-24 32-28 21. 24-27 30-25 22. 22-17 25-30 23. 18-14 29-25 24.
    27-31 25-21 25. 17-22 28-24 26. 23-18 24-28 27. 31-26 30x23 28. 18x27 28-32 29. 22-18
    32x23 30. 18x27 21-17 31. 14-10 17-14 32. 10-6 14-18 33. 6-1 18-15 34. 27-23 15-10 35.
    23-18 10-7 36. 18-15 7-3 37. 15-11 3-8 38. 11x4 *

The moves were definitely neat, I liked the tactic. But they weren’t necessary and I’m tempted to disable this one because I’m trying cut down on the number of problems with lots of alternative moves. I’m ok with a few moves that get ‘try again’ but ‘try again, try again, try again, try again, you lose!’ is extremely frustrating.

I know I don’t have enough puzzles yet and disabling them means more users getting more puzzles they’ve seen before. But a new version of the problem generator is already running on nights and weekends. I should have thousands more puzzles to upload in the near future.

I think the problem lies with the multiple solutions part. Also, while there is a forced win available with the given solution, there are quicker wins with the other moves, so in my opinion those solutions would be better.

Besides, since there are no multiple correct answers possible at the moment, in game design terms its not good to keep these in. It’s indeed very frustrating to make 4 ‘not good enough’ moves and then lose, even though the moves you made would win you the game.

Possibly in the future, the problem engine could be extended to allow multiple solutions with different rewards for the quality of the solution, but lets not get ahead of ourselves :slight_smile:

I looked about after 24-28 and15-19. They didn’t seem to lead to easier wins. Please post the PDN files after these moves so I can see what you mean. Thanks.