Author: Dieter Buerssner
Date: 16:50:14 08/18/03
Go up one level in this thread
On August 18, 2003 at 04:12:23, Uri Blass wrote: >On August 18, 2003 at 03:55:15, Andreas Stabel wrote: > >>I've made a program that scans PGN files and gives a lot of statistics. >>This is one of them. I scanned what was called the Granit collection, and >>there the following game has 92 possible moves for black at move 53. >>2r4k/3r3p/5q1P/8/5P2/5K2/6P1/qq2q3 b - - 0 53 >It is not >a 'sensible' position(with no gross material imbalance) The discussion was also around the question of how many moves a chess engine should expect. In the search tree, very unbalanced positions will happen often. Anyway, as I said already, I think nobody wants his engine to crash, when the user loads some weird position. >Maybe you should try to add the condition that the advantage of the stronger >side is not more than 2 pawns. I tried this "rule" (of course in a static sense, just material on board, ignoring for example obvious recaptures, that would make the advantage of one side much bigger, or that would bring the material balance from a too unbalanced position by this static rule). With material values of Yace, and abs(material_balance < 200 centi pawns) I get from the games downloadable at Dann Corbit's ftp (3420168 games, 277380975 moves): [D] QQQqQ1Q1/6qP/5k2/1q6/6Q1/1QK5/p1Q5/1q1q1qqq w - - 1 50 105 legal moves Which fits your "rule". But it is obviously a made up game. [Event "Zoltan Blazsik"] [Site "?"] [Date "1997.??.??"] [Round "?"] [White "Queen_50"] [Black "task"] [Result "0-1"] [ECO "A00e"] [Variation "Ware Opening"] 1. a4 b5 2. a5 Bb7 3. a6 b4 4. Na3 bxa3 5. axb7 a5 6. b4 h5 7. g4 h4 8. g5 Nh6 9. Bg2 h3 10. gxh6 hxg2 11. h4 g5 12. f4 e5 13. f5 e4 14. Nf3 exf3 15. f6 Bg7 16. fxg7 c5 17. d4 c4 18. d5 Nc6 19. dxc6 d5 20. e4 c3 21. Bb2 cxb2 22. c4 f5 23. e5 d4 24. b5 g4 25. b6 g3 26. c5 f4 27. g8=Q+ Ke7 28. h7 g1=Q+ 29. Kd2 b1=Q 30. Qgb3 Qg6 31. h5 a4 32. h6 Rg8 33. hxg8=Q a2 34. Rb1 axb1=Q 35. e6 a3 36. h7 Kf6 37. e7 g2 38. b8=Q a2 39. b7 d3 40. Kc3 d2 41. Qdc2 f2 42. c7 f1=Q 43. c8=Q Qb5 44. Qg4 f3 45. c6 f2 46. c7 f1=Q 47. c8=Q d1=Q 48. e8=Q gxh1=Q 49. bxa8=Q Qg7 50. h8=Q a1=Q+ 0-1 If I ignore this game: [D] r3r1k1/1pN2pp1/2n2n1p/p1b2b2/P4B2/3q1NP1/1P3PBP/R2Q1RK1 b - - 0 17 75 moves, in [Event "crafty-1"] [Site "home"] [Date "2002.01.30"] [Round "2"] [White "Crafty 16.15"] [Black "Crafty 18.07"] [Result "1-0"] [WhiteElo "2500"] [BlackElo "2500"] [ECO "A29"] [Variation "English: Four Knights, 4.g3 Bc5"] [EventDate "2002.01.27"] [PlyCount "84"] [Source "Mahamood"] 1. c4 e5 2. Nc3 Nf6 3. Nf3 Nc6 4. g3 Bc5 5. Bg2 O-O 6. O-O d6 7. d3 h6 8. a3 a5 9. e3 Re8 10. d4 exd4 11. exd4 Ba7 12. Nb5 Bb6 13. Bf4 Bf5 14. c5 dxc5 15. dxc5 Qd3 16. a4 Bxc5 17. Nxc7 Nh5 18. Bd2 Nf6 19. Rc1 Ne4 20. Be3 Qxd1 21. Rfxd1 Bxe3 22. fxe3 Rab8 23. Nxe8 Rxe8 24. Nd2 Nxd2 25. Bxc6 Rd8 26. Bxb7 Bg4 27. Re1 Bf3 28. Rc8 Rxc8 29. Bxc8 Nc4 30. Rf1 Be4 31. Bf5 Bb7 32. Rc1 Nxe3 33. Rc5 g6 34. Bd3 Bd5 35. Rxa5 Bb3 36. Ra8+ Kg7 37. a5 Bd5 38. Rb8 Kf6 39. a6 Ng4 40. a7 Ne5 41. Ba6 Nf3+ 42. Kf2 Nxh2 1-0 The last maximum I saw ignoring engines (but it is most probably not the real maximum): [D] r3r3/pppq1bbk/2n3pp/1NB2p2/5Q2/3P2P1/PPP1R1BP/R5K1 w - - 5 20 69 moves in [Event "Ch Brazil team"] [Site "Curitiba BRA"] [Date "1984.??.??"] [Round "4"] [White "Ribeiro, Regina"] [Black "Bittencourt F Luiz Ruppel"] [Result "0-1"] [ECO "A00u"] [Variation "Benko Opening"] [Annotator ""] [Source ""] [Remark ""] 1. g3 e5 2. Bg2 g6 3. d3 Bg7 4. e4 Nc6 5. Nc3 Nge7 6. Nge2 O-O 7. O-O d6 8. Bg5 h6 9. Be3 Be6 10. Qd2 Kh7 11. f4 f5 12. Rf2 d5 13. exd5 Nxd5 14. Bc5 Re8 15. fxe5 Nxe5 16. Nf4 Nxf4 17. Qxf4 Qd7 18. Re2 Nc6 19. Nb5 Bf7 20. Rae1 Rxe2 21. Rxe2 Rc8 22. Nxa7 Bd4+ 23. Bxd4 Nxd4 24. Re5 Ra8 25. Qe3 Nxc2 26. Qc5 Qd4+ 27. Qxd4 Nxd4 28. Re7 Kg8 29. Rxc7 Rxa7 30. Rd7 Ne6 31. Bd5 Ng5 32. Bxb7 Rxa2 33. h4 Ne6 34. d4 Rxb2 35. Bg2 Rd2 36. d5 Nc5 37. Rc7 Ne4 0-1 I remember that I saw in previous runs games with more legal moves in games between well known opponents (but I did not use the 200 cp rule). Regards, Dieter
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.