Author: Dana Turnmire
Date: 21:44:52 06/03/01
Go up one level in this thread
On June 03, 2001 at 15:24:05, John Merlino wrote: >On June 03, 2001 at 01:54:50, Dana Turnmire wrote: > >> Genius 2 is several years old now but notice how it crushed the Q-side castled >>position of Chessmaster 8000 (both on default settings). This is typical of how >>Genius plays. (30 min/game) >> >> White "Chessmaster" >> Black "Genius 2" >> >>1. e4 d6 2. d4 Nf6 3. Nc3 g6 4. f4 Bg7 5. Nf3 c5 6. dxc5 Qa5 7. Bd3 Qxc5 8. Qe2 >>Nc6 9. Be3 Qa5 10. 0-0-0 0-0 11. Bc4 a6 12. Nd5 Nxd5 13. exd5 b5 14. Bb3 Nb8 15. >>Bd4 Re8 16. Bxg7 Kxg7 17. Rhe1 Nd7 18. Kb1 Nc5 19. h3 Qb6 20. Qe3 a5 21. a3 b4 >>22. Bc4 Rb8 23. Qd4+ Kg8 24. Re3 bxa3 25. Rxa3 Qc7 26. Re1 Rb4 27. Nd2 Bd7 28. >>Rae3 Bb5 29. b3 Bxc4 30. Nxc4 a4 31. Qc3 Rb5 32. Nd2 axb3 33. cxb3 Qb7 34. Qc4 >>Rb4 35. Qc2 Rxf4 36. Qc3 Rf2 37. g4 e5 38. dxe6ep fxe6 39. R3e2 Rxe2 40. Rxe2 >>Qh1+ 41. Qc1 Qxh3 42. Qc4 d5 43. Qc2 Qg3 44. Nf1 Qxg4 45. Re1 Qb4 46. Nd2 Nxb3 >>47. Qxb3 Qxd2 48. Rd1 Qe2 49. Rc1 Qe4+ 50. Kb2 Qe5+ 51. Qc3 Rb8+ 52. Ka1 d4 53. >>Qc5 Qe3 54. Qc8+ Rxc8 55. Rxc8+ Kg7 56. Kb2 d3 57. Rc7+ Kh6 58. Rc1 Qd2+ 59. Kb1 >>e5 60. Rh1+ Kg7 61. Rc1 Qb4+ 62. Ka2 d2 63. Rh1 e4 64. Rg1 e3 65. Rb1 Qa4+ 66. >>Kb2 e2 67. Kc3 d1=Q 68. Rb7+ Kh6 69. Rxh7+ Kxh7 70 Kb2 Qaa1# > >For the record, 10.0-0-0 is the first out-of-book move, and it takes CM8000 just >over 2 1/2 minutes on my PIII-733 to change its mind to 10.Bd2. From the score, >it appears that the book was not the problem here (nor, necessarily, is the >queenside castle): > >Time Depth Score Positions Moves >0:00 2/5 0.33 19369 10. O-O-O Ng4 11. Bd2 Be6 12. h3 >0:01 2/6 0.42 62978 10. O-O-O Be6 11. Bc4 Bxc4 12. > Qxc4 O-O 13. Kb1 Rac8 >0:04 3/7 0.37 309965 10. O-O-O O-O 11. Bc4 a6 12. a3 > b5 13. Bb3 Bg4 >0:13 3/8 0.40 918179 10. O-O-O Be6 11. Kb1 O-O 12. Bc4 > Ng4 13. Bxe6 fxe6 14. e5 d5 >0:48 4/9 0.47 3520856 10. O-O-O O-O 11. Bc4 Bd7 12. h3 > a6 13. Bb3 b5 14. g4 >1:59 4/10 0.30 8696139 10. O-O-O Be6 11. Bd2 O-O 12. a3 > a6 13. Nd5 Qd8 14. Nxf6+ Bxf6 15. > Qf2 Bg7 >2:32 4/10 0.36 11298675 10. Bd2 Nb4 11. Bb5+ Bd7 12. Kd1 > Rc8 13. Bxd7+ Kxd7 14. a4 Qc7 15. > Qb5+ Nc6 16. Be3 >6:51 5/11 0.32 33433057 10. Bd2 Nb4 11. Bb5+ Bd7 12. Kd1 > Rc8 13. Bxd7+ Kxd7 14. a3 Qa6 15. > Rc1 Qxe2+ 16. Kxe2 Nc6 17. Be3 >25:02 6/12 0.43 125449401 10. Bd2 O-O 11. Nd5 Qd8 12. Nxf6+ > Bxf6 13. O-O-O Bg4 14. Kb1 e6 15. > Be3 Nd4 16. Qf2 Nxf3 17. gxf3 > >But, there is something odd here, and perhaps it can all be attributed to time >management. Either way, here's the evidence: > >-- On my PIII-733, 18.Kb1 is discarded in favor of 18.h3 after 39 seconds: > >Time Depth Score Positions Moves >0:00 1/4 1.10 4796 18. Kb1 Nc5 19. Nd4 e6 20. dxe6 > fxe6 >0:01 2/5 0.94 21583 18. Nd4 Bb7 19. Kb1 Rac8 20. Nf3 > Nf6 >0:01 2/6 1.02 39911 18. Nd4 Bb7 19. Kb1 Rac8 20. Nf3 > Nf6 21. Ng5 >0:05 3/7 0.69 162909 18. Nd4 Nf6 19. h3 Qc7 20. Kb1 > Bb7 21. Nf3 Kf8 >0:06 3/7 0.83 204114 18. Kb1 Nf6 19. h3 Bf5 20. g4 Be4 > 21. Ng5 >0:09 3/8 0.93 467436 18. Kb1 Nf6 19. h3 h6 20. g4 e6 > 21. dxe6 Bxe6 >0:18 4/9 0.64 1134111 18. Kb1 Nf6 19. h3 h5 20. Rd3 Bf5 > 21. Re3 Qb4 22. g3 Ng8 >0:39 4/9 0.79 2735989 18. h3 Nf6 19. g4 h5 20. g5 Ng8 > 21. h4 Bf5 22. Kb1 Rac8 > >-- 23.Qd4+ is discarded in favor of 23.axb4 (with a brief PV in favor of 23.a4) >after 39 seconds: > >Time Depth Score Positions Moves >0:00 1/3 ---- 2735 ---- >0:00 1/4 0.89 10668 23. Ng5 bxa3 24. Qxa3 e5 25. dxe6 > fxe6 >0:01 2/5 0.89 52453 23. Nd4 Bd7 24. Nb3 Nxb3 25. Qxb3 > Bf5 >0:01 2/6 0.59 104155 23. Nd4 Bd7 24. axb4 Qxb4 25. Bb3 > e6 26. dxe6 fxe6 >0:02 2/6 0.67 139594 23. Qd4+ f6 24. axb4 axb4 25. Qe3 > b3 26. Qc3 e5 27. dxe6 Bxe6 28. > cxb3 >0:05 3/7 0.63 348389 23. Qd4+ f6 24. axb4 Qxb4 25. Ng5 > Na4 26. Bb3 Qxd4 27. Rxd4 Nxb2 > 28. Kxb2 fxg5 29. fxg5 >0:14 3/8 0.40 1062157 23. Qd4+ f6 24. axb4 axb4 25. g4 > Qa7 26. Ng5 h6 27. Ne6+ Bxe6 28. > dxe6 >0:35 3/8 0.50 2635740 23. a4 Bd7 24. Bb3 Kg8 25. g4 e6 > 26. dxe6 fxe6 >0:39 3/8 0.61 2924219 23. axb4 Qxb4 24. Qd4+ f6 25. Re2 > Na4 26. Bb3 Nc5 27. Qxb4 Rxb4 28. > Nd4 > >-- 34.Qc4 is discarded in favor of 34.f5 after 25 seconds (note that the game >is, in all likelihood, already lost here, but f5 could have made a big >difference): > >Time Depth Score Positions Moves >0:00 1/4 0.15 8365 34. g3 h6 35. Qd4 Nxb3 >0:00 2/5 -0.04 38606 34. g3 Rc8 35. Qb2 e6 36. dxe6 Nxe6 >0:01 2/6 -0.15 93827 34. g3 Rc8 35. Rc1 Rb8 36. Re2 Qxd5 >0:04 3/7 -0.88 373130 34. g3 Rc8 35. Rc1 Rc7 36. Qd4 > Nxb3 37. Nxb3 Rxc1+ 38. Kxc1 Rxb3 >0:06 3/7 -0.85 601184 34. g4 Rc8 35. Rc1 Rc7 36. Qd4 > Nxb3 37. Nxb3 Rxb3+ 38. Ka2 >0:08 3/7 -0.67 773735 34. R1e2 Rc8 35. Rxe7 Ne4 36. Qxc8+ > Qxc8 37. R7xe4 Qf5 38. Kc1 Rc5+ > 39. Kd1 Qxd5 40. Re8+ Kg7 >0:10 3/7 -0.37 946851 34. Qc4 e6 35. dxe6 fxe6 36. Rf3 > Ra8 37. Qc3 Qa7 >0:18 3/8 -0.33 1640523 34. Qc4 e6 35. dxe6 fxe6 36. g3 > Rb8 37. Rc3 Rb4 >0:25 3/8 -0.25 2276307 34. f5 Qxd5 35. fxg6 e5 36. gxf7+ > Qxf7 37. Rg3+ Kh8 38. Rf1 Qb7 39. > Qc4 > >-- BUT, 36.Qc3 (the move in the game) is NOT a PV until after 1:04 (unless you >count the ones in less than a second): > >Time Depth Score Positions Moves >0:00 1/3 -1.22 2020 36. Qc3 e5 37. dxe6 fxe6 >0:00 2/5 -1.28 28682 36. Qc3 e5 37. dxe6 fxe6 38. Rf3 > Ref8 39. Rxf4 Rxf4 >0:01 2/6 -1.15 69689 36. Nc4 Rb8 37. Na5 Qxd5 38. Rxe7 > Rd4 >0:03 3/7 -1.20 259981 36. Nc4 Qxd5 37. Rxe7 Rxe7 38. > Rxe7 Rf1+ 39. Ka2 Qa8+ 40. Na3 > Rg1 41. g4 Rg2 >0:07 3/8 -1.44 665267 36. Nc4 Qxd5 37. Rxe7 Rxe7 38. > Rxe7 Nxb3 39. Ne3 Qb5 40. Qc8+ > Kg7 41. Qa8 Rf2 >0:27 4/9 -1.83 2607731 36. Nc4 Qxd5 37. Rxe7 Rxe7 38. > Rxe7 Nxb3 39. Ne3 Qa5 40. Re8+ > Kg7 41. Qb2+ Nd4 42. Re7 d5 43. > Rd7 Qe1+ 44. Qc1 >1:04 4/9 -1.54 6048732 36. Qc3 Rf5 37. Qc4 Qxd5 38. g4 > Qxc4 39. Nxc4 Rd5 40. Rxe7 Rxe7 > 41. Rxe7 Nxb3 > >Just something to muddle on.... > >jm I wasn't trying to bash CM8000 but was pointing out the popular style of play of Genius 2. It was well known for its playing strength especially in the faster time controls and can probably hold its own even today in that category as evidenced in this game in 30 minutes. However, at the longer times Genius 2 starts losing strength. I started this game at move 10 where Chessmaster castled queenside (and went out of book) and changed the time to 40/120min. The outcome was quite different with Genius losing its queen on the 25th move. I believe Chessmaster 8000 is a very strong program even though it seems the opening book could be a little better and certainly the learning feature wouldn't hurt a bit. While on the subject my wish list for CM9000 would include: 1. The option of a simple interface similar to that of Genius 2 where I could simply put in a game and it would analyze it with the Chessmaster engine for whatever side it is to play without having 3 different buttons to click. If White is winning by 5.83 when it is Black's turn the evaluation will go to -5.83. I don't want to worry if I have multiple engines running in the background (if professional testers like those at SSDF can't figure out how to use it it simply is to complicated). Maybe have an "advanced user" button on the task bar for those who wish to get involved in depth. 2. I already mentioned the learning feature. There is simply too much evidence the opening book can be critical in matches with humans or other computers if the program insists on playing the same losing lines. 3. I have never used endgame tables but it certainly seems to be a standard feature on all the top programs now. 4. I believe you can keep all the nice features CM8000 has but one can get so bogged down in so many features you can't see the forest for the trees. Here is how the game went starting at move 10 at 40/120min. 10. Bd2 0-0 11. Nd5 Qd8 12. 0-0-0 a5 13. Bc3 Nxd5 14. exd5 Bxc3 15. bxc3 Nb8 16. Rhe1 Re8 17. Qe3 Qc7 18. Nd4 Bg4 19. Bb5 Bd7 20. f5 Bxb5 21. Nxb5 Qd7 22. Qh6 Na6 23. Nd4 Nc7 24. Ne6 Qxe6 and Genius loses its queen or gets mated.
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.