Author: Robert Hyatt
Date: 07:18:16 09/26/98
Go up one level in this thread
On September 25, 1998 at 18:04:35, Danniel Corbit wrote: >Here is a position from the kurrent kkup2 kontest from >http://www.gambitsoft.com/kup/chess1/index1.html [which I have carped about >before]: >[Event "Komputer Kup 2"] >[Site "?"] >[Date "1998.2.9"] >[Round "?"] >[White "Crafty (C)"] >[Black "Rebel 9 (C)"] >[Result "*"] > >1. e4 e6 2. d4 d5 3. Nc3 Nf6 4. Bg5 Be7 5. e5 Nfd7 6. Bxe7 Qxe7 7. f4 O-O 8. >Nf3 c5 9. Qd2 Nc6 10. dxc5 Qxc5 11. O-O-O Nb6 12. Bd3 f6 13. exf6 Rxf6 14. g3 >Bd7 15. Rhe1 Nc4 16. Bxc4 Qxc4 17. Qe2 b5 18. Qxc4 bxc4 19. Ne5 Nxe5 20. Rxe5 >Raf8 21. Ne4 Rh6 22. Nc5 Bc8 23. Rd2 Kf7 24. h4 Re8 25. b3 cxb3 26. axb3 Re7 >27. Re3 Rc7 28. Nd3 Rf6 29. Ne5+ Ke7 30. Rde2 Kd6 31. Re1 Ba6 32. Kd2 Bb5 33. >Ra1 d4 34. Ree1 Rf5 35. Kc1 Rc3 36. Rxa7 Rxg3 37. Nf7+ Kd5 38. Ng5 h6 39. Nxe6 >Re3 40. Nc7+ Ke4 41. Rg1 Rf3 42. Rxg7 R3xf4 43. Ra5 Bf1 44. Re7+ Kf3 45. Rxf5 >Rxf5 46. Ne6 Ke4 47. Ng7+ Re5 48. Rxe5+ Kxe5 49. Kd2 Bb5 50. Nh5 Be8 51. Ng3 >Bd7 52. Ne2 Ke4 53. Nc1 Be8 54. Nd3 Bb5 55. Nb2 Kd5 * > >I like: >56. c4+ > >Crafty played: >56. Kc1 > >Now, the reason I like this move over the one Crafty chose is many-fold. As >far as material, it is a near tie, as I would rather have the bishop than >the knight at this stage, but technically Crafty is a pawn up. I believe >that the black pawn on d4 is much more valuable than white pawn on d2, >because of two reasons. First, it is far advanced towards the enemy's first >rank. This will cause white to have to tie up material to prevent queening. >Second, the black king and bishop are nearby for protection of the pawn, and >can make things difficult. In short, the pawn on d4 will be a constant >worry until it is eliminated. By playing c4+, the *only* sensible choice is >an e.p. capture, which trades pawns, but [in my -not so expert- view] a >great pawn for black is eliminated in return for a ho-humm pawn for white. > >I think most would agree that the pawns are *not* of the same value. Is >there any program that takes this into account? Or is my analysis all >washed up? > >I realize that in this case, it is a tempest in a teapot, since the game is >probably a draw anyway. But in other situations not so. Opinions? This is a difficult question to answer. The short answer is that yes, pawn values are very dynamic, at least in crafty. How far they have advanced, whether they can be defended by pawns (are they isolated or artificially isolated because they have advanced too far to be defended by pawns) are they passed, passed and protected, if weak, are they on open files, can they be used as a "lever" or are they blocked from doing so... EvaluatePawns() is a large piece of code... and it is probably easier to read the comments to at least see what Crafty does...
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.