Author: Uri Blass
Date: 14:25:41 02/11/01
Go up one level in this thread
On February 11, 2001 at 16:49:07, Tim Foden wrote: >Just thought of this one too: > >[D]3k4/4b3/1p1p1p1p/1P1P1P1P/1P1P1P1P/8/3BB3/3K4 w > >.. this should be quite difficult :-) > >Here is the tail end of the analysis with Crafty 17.11, 920MHz Duron, 192MB >hash: > > 59 2:33 8.12 1. Kc2 Kd7 2. Kd3 Bf8 3. Ke4 Bg7 4. > Bc4 Ke7 5. Bc3 Kd7 6. Be1 Ke7 7. Bd2 > Kd7 8. Be2 Ke8 9. Bc3 Ke7 10. Be1 Kf7 > 11. Bf3 Kg8 12. Kd3 Kf7 13. Be4 Ke7 > 14. Ke3 Kd8 15. Bc3 Ke7 16. Bd3 Kd7 > 17. Ke4 Ke7 18. Be2 Kf7 19. Kd3 Ke7 > 20. Bf3 Ke8 21. Ke3 Ke7 22. Bg4 Kd7 > 23. Be2 Ke7 24. Bf3 Kf7 25. Kd2 Ke7 > 26. Bg2 Kd7 27. Be4 Ke7 28. Ke2 Kd7 > 29. Kf3 Ke7 30. Ke3 > 59-> 2:39 8.12 1. Kc2 Kd7 2. Kd3 Bf8 3. Ke4 Bg7 4. > Bc4 Ke7 5. Bc3 Kd7 6. Be1 Ke7 7. Bd2 > Kd7 8. Be2 Ke8 9. Bc3 Ke7 10. Be1 Kf7 > 11. Bf3 Kg8 12. Kd3 Kf7 13. Be4 Ke7 > 14. Ke3 Kd8 15. Bc3 Ke7 16. Bd3 Kd7 > 17. Ke4 Ke7 18. Be2 Kf7 19. Kd3 Ke7 > 20. Bf3 Ke8 21. Ke3 Ke7 22. Bg4 Kd7 > 23. Be2 Ke7 24. Bf3 Kf7 25. Kd2 Ke7 > 26. Bg2 Kd7 27. Be4 Ke7 28. Ke2 Kd7 > 29. Kf3 Ke7 30. Ke3 > 60 2:44 8.12 1. Kc2 Kd7 2. Kd3 Bf8 3. Ke4 Bg7 4. > Bc4 Ke7 5. Bc3 Kd7 6. Be1 Ke7 7. Bd2 > Kd7 8. Be2 Ke8 9. Bc3 Ke7 10. Be1 Kf7 > 11. Bf3 Kg8 12. Kd3 Kf7 13. Be4 Ke7 > 14. Ke3 Kd8 15. Bc3 Ke7 16. Bd3 Kd7 > 17. Ke4 Ke7 18. Be2 Kf7 19. Kd3 Ke7 > 20. Bf3 Ke8 21. Ke3 Ke7 22. Bg4 Kd7 > 23. Be2 Ke7 24. Bf3 Kf7 25. Kd2 Ke7 > 26. Bg2 Kd7 27. Be4 Ke7 28. Ke2 Kd7 > 29. Kf3 Ke7 30. Ke3 Kd7 > 60-> 2:49 8.12 1. Kc2 Kd7 2. Kd3 Bf8 3. Ke4 Bg7 4. > Bc4 Ke7 5. Bc3 Kd7 6. Be1 Ke7 7. Bd2 > Kd7 8. Be2 Ke8 9. Bc3 Ke7 10. Be1 Kf7 > 11. Bf3 Kg8 12. Kd3 Kf7 13. Be4 Ke7 > 14. Ke3 Kd8 15. Bc3 Ke7 16. Bd3 Kd7 > 17. Ke4 Ke7 18. Be2 Kf7 19. Kd3 Ke7 > 20. Bf3 Ke8 21. Ke3 Ke7 22. Bg4 Kd7 > 23. Be2 Ke7 24. Bf3 Kf7 25. Kd2 Ke7 > 26. Bg2 Kd7 27. Be4 Ke7 28. Ke2 Kd7 > 29. Kf3 Ke7 30. Ke3 Kd7 > time=2:50 cpu=99% mat=7 n=119315477 fh=95% nps=699141 > ext-> checks=0 recaps=0 pawns=0 1rep=0 thrt:0 > predicted=0 nodes=119315477 evals=2214248 > endgame tablebase-> probes done=0 successful=0 > hashing-> trans/ref=30% pawn=99% used=58% >analyze.White(1): > >As you can see, crafty still thinks white is winning. > >(I ran crafty because I don't really trust GLC in these kind of positions. I >think I have a bug somewhere... still haven't found it). > >Cheers, Tim. Programs should solve it by seeing that it is a draw by the 50 move rule. The number of relevant positions: 2*8*15*(12*11*16+16*15*12)=1,198,080 In every ply you need to search at most 1,198,080 positions so in order to search 100 ply forward you need to search less than 119,808,000 positions. Explanation to my numbers: 2*8*15 2 is the side to move. 8 is the number of squares for the black bishop. 15 is the number of squares for the black king. 12*11*16 12 is the number of black squares for the white king. 11 is the number of black squares for the white bishop when the white king is on a black square. 16 is the number of white squares for the white bishop after knowing that the white king is on black square. 16*15*12 16 is the number of white squares for the white king and the rest is similiar. Uri
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.