Author: Mark Young
Date: 18:26:37 05/24/99
Go up one level in this thread
On May 24, 1999 at 07:43:02, Ed Schröder wrote: >REBEL vs GM Rohde (move 23 and 26) > >23. exd4 or Rxb4 > >It has been said that 23.Rxb4 is a better move than 23.exd4 (Rebel's move). >I partly agree but IMO white's position is already lost also after 23.Rxb4. The >fact that computers evaluate the position after Rxb4 with +0.50 or higher is >no promotion for the current state of art of computer chess IMO. To make my statement clear to you, it had nothing to do with hiarcs7 finding Rxb4. I myself can not and would not assume that Rebel 10.5 can represent how other programs can play as was addressing in context of that thread. Why, because Rebel 10.5 is a unknown, a prototype...as Bob would say a work in progress. It is clear to me that Rebel 10.5 by playing the move 26. f4 has some kind of bug. And other proven programs like Hiarcs7, Fritz5 and even Rebel 9 understand the error of 26. f4. This was not ment as a slam against you or Rebel 10.5. Just saying hold on, do not give up on all micros because a new version of Rebel10 had problems. After looking at the game it is becoming clear that Rebel10 plays one mean game of positional chess. And not for a tactical bug, may have found 23. Rxb4 and showing it was positional justified in its rook manouver in the earlier couple of moves. And almost for sure the bug cost rebel a draw in missing 26. Bg2. > >I had a close look at Rebel's moves 23 and 26. In Rebel7 I removed a "mate- >threat" extension which impact I considered none to zero. However this game >seems to be full of it. When I enable Rebel6's extension again (just one >instruction) Rebel sees all the danger much quicker. > >REBEL 10.5 (as played in the game) >1r1qr2k/1p4pp/3p4/1R1B1p2/1pPp4/4P1Pb/1P3P1P/Q3R1K1 w - - >00:00:04 7.00 0.99 e3xd4 Re8xe1+ Qa1xe1 Qd8-e8 Qe1-d2 f5-f4 Bd5xb7 (2) >00:00:10 8.00 1.05 e3xd4 Re8xe1+ Qa1xe1 Qd8-e8 Qe1xe8+ Rb8xe8 f2-f4 (5) >00:00:44 9.00 1.10 e3xd4 h7-h6 Rb5xb7 Rb8xb7 Bd5xb7 Kh8-h7 Bb7-d5 (11) >00:01:09 10.00 0.83 e3xd4 Rb8-a8 Re1xe8+ Qd8xe8 Qa1-c1 Qe8-e2 Rb5xb7 (47) >00:02:16 11.00 0.83 e3xd4 Rb8-a8 Re1xe8+ Qd8xe8 Qa1-c1 Qe8-e2 Rb5xb7 (81) >00:05:10 12.00 0.60 e3xd4 Rb8-a8 Re1xe8+ Qd8xe8 Qa1-c1 Qe8-e2 Rb5xb7 (219) > >REBEL 10.5 (with the Rebel6 extension) >1r1qr2k/1p4pp/3p4/1R1B1p2/1pPp4/4P1Pb/1P3P1P/Q3R1K1 w - - >00:00:05 7.00 1.01 e3xd4 Re8xe1+ Qa1xe1 Qd8-e8 Qe1-d1 f5-f4 Kg1-h1 (2) >00:00:11 8.00 1.05 e3xd4 Re8xe1+ Qa1xe1 Qd8-e8 Qe1xe8+ Rb8xe8 f2-f4 (6) >00:00:27 9.00 1.02 e3xd4 Rb8-a8 Re1xe8+ Qd8xe8 Qa1-c1 Ra8-a7 Rb5-b6 (12) >00:00:46 10.00 0.83 e3xd4 Rb8-a8 Re1xe8+ Qd8xe8 Qa1-c1 Qe8-e2 Rb5xb7 (30) >00:01:25 11.00 0.00 e3xd4 Rb8-a8 Re1xe8+ Qd8xe8 Qa1-c1 Qe8-e2 Rb5xb7 (69) >00:09:58 11.02 0.50 Rb5xb4 d4xe3 Re1xe3 Re8xe3 f2xe3 h7-h6 Bd5xb7 (120) > >Not that it matters much as black gets the crucial e-file anyway. > >Rebel10.5 needs 15 plies (and 3 hours) to see that 23.exd4 is 0.00, the new >version 11 plies (and 1 minute). I think I will keep the change :-) > > >26.f4? (Bg2) > >REBEL10.5 (with the Rebel6 extension) >r6k/1p4pp/3p4/1R1B1p2/1pPP4/6Pb/1P2qP1P/2Q3K1 w - - >00:00:04 7.00 0.47 Rb5xb7 Ra8-e8 f2-f4 Qe2-d3 Rb7xb4 Qd3xd4+ Kg1-h1 (3) >00:00:06 8.00 0.00 Rb5xb7 Ra8-e8 f2-f4 g7-g6 g3-g4 Re8-e4 Rb7-b8+ (4) >00:00:13 9.00 0.00 Rb5xb7 Ra8-e8 f2-f4 h7-h6 Rb7-f7 Re8-e4 Rf7-f8+ (10) >00:00:30 10.00 -0.04 Rb5xb7 Ra8-e8 f2-f4 h7-h6 Rb7-f7 Qe2-d3 Kg1-h1 (19) >00:01:33 11.00 -1.88 Rb5xb7 Ra8-e8 f2-f4 g7-g6 Bd5-h1 Qe2-e1+ Qc1xe1 (47) >00:03:36 11.08 -0.19 Bd5-g2 Qe2-d2 Qc1-f1 Bh3xg2 Kg1xg2 Qd2xb2 Rb5xb7 (175) >00:04:32 12.00 -0.19 Bd5-g2 Qe2-d2 Qc1-f1 Bh3xg2 Kg1xg2 Qd2xb2 Rb5xb7 (234) > >Also here the Rebel6 "mate-threat" extension does miracles and the correct >move is played. I will have a closer look if this old Rebel6 "mate-threat" >extension can be improved. > >Ed Schroder > >-------------------------------------------------------------------------- > >[Event "Monthly GM Challenge"] >[Site "Internet (Rebel Home Page)"] >[Date "1999.05.22"] >[Round "?"] >[White "Rebel 10.5"] >[Black "GM M. Rohde"] >[Result "0-1"] >[WhiteElo "?"] >[BlackElo "?"] >[ECO "E00"] > >1. d4 Nf6 2. c4 e6 3. g3 Bb4+ 4. Bd2 c5 5. Bxb4 cxb4 6. Bg2 O-O 7. Nf3 >d6 8. O-O a5 9. a3 Na6 10. Nbd2 Re8 11. Ne1 e5 12. Nc2 bxa3 13. Rxa3 Rb8 >14. Ne4 Nxe4 15. Bxe4 Bh3 16. Re1 Qb6 17. e3 f5 18. Bd5+ Kh8 19. Qa1 Nb4 >20. Nxb4 axb4 21. Ra5 exd4 22. Rb5 Qd8 23. exd4 Ra8 24. Rxe8+ Qxe8 25. >Qc1 Qe2 26. f4 Qd3 27. Rxb7 Re8 28. Kh1 Qc2 29. Qe3 Qb1+ 30. Qg1 Qxb2 31. >Rb6 h6 32. Ra6 Qc2 33. c5 dxc5 34. dxc5 Qc3 35. Ra1 b3 36. Rc1 Qd3 37. >Rd1 Qe2 38. Rb1 b2 39. Qd1 Qxd1+ 40. Rxd1 Rd8 41. Rb1 0-1
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.