Problem 962 Spoiler Alert

#1

The first few moves are correct. However 5-9 is very weak. 6-10 is much stronger and leads to a much harder win by White. Here’s one nice sequence of moves. There are others which all lead to a White win.

  1. List item

[Event “”]
[Date “”]
[Black “”]
[White “”]
[Result “*”]
[Setup “1”]
[FEN “B:W7,13,22,25,26,28,30:B2,5,6,8,10,17”]

  1. 2x11 25-21 2. 10-14 22-18 3. 14x23 26x19 4. 17-22 28-24 5. 6-10 21-17 6. 11-15 17-14 7.
    10x17 19x10 8. 8-11 10-7 9. 11-16 7-3 10. 17-21 3-7 11. 22-25 24-20 12. 16-19 20-16 13.
    19-23 7-10 14. 23-27 16-12 15. 25-29 10-15 16. 29-25 15-18 17. 27-31 12-8 18. 31-27 8-3
  2. 27-24 3-7 20. 24-19 7-10 21. 19-24 10-14 22. 24-19 13-9 23. 19-24 9-6 24. 24-27 6-2
  3. 27-31 2-6 26. 31-27 14-17 27. 27-24 17-22 28. 25-29 6-10 29. 24-27 10-14 30. 27-32
    18-23 31. 32-28 23-27 32. 28-32 22-26 33. 32x23 26x19 34. 29-25 14-18 35. 5-9 19-23 36.
    9-13 23-27 37. 13-17 27-31 38. 17-22 18-23 39. 25-29 23-26 *
#2

In these early problems the engine wasn’t running nearly as a long on each position as it is now. Maybe it did just play a bad move? Or Is this a case where it played some ‘objectively’ best move, choosing the slow easy loss instead of a quicker but more difficult one?

If it’s the former, I can write a note about recreating this puzzle with the new generator. I don’t have a mechanism for doing this yet but I will eventually and next time it should choose the better move. If it’s the later, we’re probably stuck with it. I haven’t found a way to prevent the engine from being stupid sometimes.

Either way, should I disable the problem for now?