Author: Laurence Chen
Date: 01:19:59 12/13/98
Go up one level in this thread
On December 12, 1998 at 12:56:48, James T. Walker wrote:
>On December 11, 1998 at 11:48:36, Laurence Chen wrote:
>
>>On December 10, 1998 at 16:34:20, Georg Langrath wrote:
>>
>>>In the begining chesscomputers never made real sacrifices for position. Super
>>>Constellation was the first chesscomputer that was able to do that.
>>>
>>>Dear members of CCC. Have you nice examples of genuine sacrifices from your
>>>chesscomputers? They are nice to look at, and it is interesting to examine if
>>>other programs make the sacrifice. Of course all sacrifices are not healthy, but
>>>they are always nice to look at, especially from chesscomputers.
>>>
>>>Please answer this notice with sacrifices from your computers. I think we all
>>>like to examine them.
>>>
>>>Georg
>>Junior 5.0 sacrifices a Knight on move 20 for the initiative and attack.
>>
>>[Event "Level=Blitz:30'+5"]
>>[Site "?"]
>>[Date "1998.12.11"]
>>[Round "?"]
>>[White "CM 6000, 16MB."]
>>[Black "Junior 5.0, 8MB."]
>>[Result "0-1"]
>>[ECO "B89"]
>>[PlyCount "116"]
>>[EventDate "1998.12.11"]
>>
>>1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 d6 6. Bc4 e6 7. Be3 a6 8. Qe2
>>Qc7 9. O-O-O Be7 10. Bb3 O-O 11. Nxc6 {0} 11... Qxc6 {0.03/13 88} 12. f3 {48}
>>12... b5 {0.06/15 96} 13. Qf2 {39} 13... a5 {-0.28/14 48} 14. a3 {15} 14... b4
>>{-0.61/14 99} 15. Ba4 {8} 15... Qb7 {-0.74/13 78} 16. Nb1 {11} 16... d5 {
>>-0.74/13 68} 17. e5 {49} 17... bxa3 {-1.06/13 145} 18. Nxa3 {6} 18... Rb8 {
>>-1.05/11 39} 19. Nb5 {62} 19... Bd7 {-1.22/12 0} 20. exf6 {41} 20... Bxf6 {
>>-1.65/12 0} 21. c4 {22} 21... Rfc8 {-2.40/12 60} 22. Bc5 {8} 22... Bxb5 {
>>-2.70/12 23} 23. Bxb5 {105} 23... dxc4 {-3.24/14 0} 24. Bd6 {39} 24... Qxb5 {
>>-3.73/13 40} 25. Bxb8 {6} 25... Rxb8 {-3.73/10 10} 26. Qc2 {81} 26... Bxb2+ {
>>-3.69/12 0} 27. Kd2 {27} 27... Qe5 {-3.72/11 15} 28. Qxc4 {48} 28... Ba3 {
>>-3.64/12 0} 29. Rb1 {31} 29... Qg5+ {-3.69/11 0} 30. Ke2 {19} 30... Qxg2+ {
>>-3.59/11 18} 31. Ke3 {28} 31... Bb4 {-3.61/11 0} 32. Rhd1 {34} 32... Qxh2 {
>>-3.74/11 52} 33. Qd4 {7} 33... h5 {-3.48/10 59} 34. Rh1 {42} 34... Qc2 {
>>-3.59/11 0} 35. Rbc1 {20} 35... Qf5 {-3.65/10 20} 36. f4 {36} 36... Ba3 {
>>-4.95/11 116} 37. Rc3 {35} 37... Qg6 {-4.80/11 0} 38. Rf1 {127} 38... Qg3+ {
>>-6.85/10 31} 39. Ke2 {9} 39... Qg2+ {-6.85/9 7} 40. Qf2 {8} 40... Rb2+ {
>>-8.30/10 8} 41. Ke3 {11} 41... Rxf2 {-8.70/11 0} 42. Rc8+ {42} 42... Kh7 {
>>-9.33/13 35} 43. Rxf2 {2} 43... Qg3+ {-9.31/11 66} 44. Ke2 {9} 44... e5 {
>>-9.78/11 156} 45. Rc2 {6} 45... exf4 {-9.75/8 3} 46. Rf3 {32} 46... Qg2+ {
>>-11.70/11 48} 47. Rf2 {5} 47... Qe4+ {-11.70/9 12} 48. Kf1 {4} 48... f3 {
>>-11.31/9 8} 49. Rfd2 {18} 49... Qe3 {-13.92/11 23} 50. Rh2 {8} 50... Qd3+ {
>>-18.38/9 10} 51. Ke1 {7} 51... Bb4+ {-18.72/9 14} 52. Kf2 {6} 52... Qxc2+ {
>>-19.63/8 5} 53. Kxf3 {2} 53... Qxh2 {-20.35/8 10} 54. Ke3 {1} 54... Qd2+ {
>>-#6/8 26} 55. Kf3 {5} 55... Qd3+ {-#4/8 7} 56. Kf2 {5} 56... Bc5+ {-#3/8 4} 57.
>>Ke1 {5} 57... Qc2 {-#2/6 0} 58. Kf1 {4} 58... Qf2# {-#1/3 0} 0-1
>>>>>>>>>>>>>>>>>>>>>>>>>
>Hello Laurence,
>The "sacrifice" you mention on move 20 is a very nice move. Actually the
>sacrifice is by move 19..... Bd7 which allows the pawn to capture the knight.
>However, I'm not sure you can call this a "Sacrifice" since most computer
>programs will make the Bd7 move and allow the capture of the Knight. The
>problem is the Knight on b5 is attacked 3 times and defended once so it is still
>in danger plus the recapture of the pawn on f6 puts another attacker on the b2
>square. Mate seems to be threatened if white attempts to save the Knight. This
>is only my opinion as an amateur so I may be wrong. In any case it brings up
>the question of what constitutes a "sacrifice" by a program. Since the score
>went up by making the moves the computer thinks it is a good move which will
>eventually lead to gain of material back with additional positional advantage in
>the bargain. I would be interested in what programers say they consider a
>sacrifice by their programs "If they do sacrifice" at all.
>By the way the programs I tested your postion on were:
>
>Program Time
>Fritz5.16 0:01
>Rebel 10 0:06
>Rebel Dec. 0:18
>Crafty15.20 3:05
>CM5500 0:50
>CM6000 1:06
>
>In any case it was a good move and a nice game to view.
>Regards,
>Jim Walker
That's a good point, what a chess engine considers to be a "sacrifice"? One
thing that I noticed was that whenever a chess engine made a sacrifice, it did
it because it had a positive PV, not a negative one. One would assume then that
the sacrifice is not a true one because it is lead to believe that the engine
saw everything. I don't think that is the case, sometimes the PV is positive and
the engine makes the sacrifice, and it turns out several moves later, the
sacrifice was unsound, and because chess engines suffer from horizon blindness,
it may fail to see that such sacrifices are unsound. Positional sacrifices are
very common in GM games, not so much in masters games, and sometimes one finds
in amateur games only because it was a blunder move which lead to such
sacrifices. Take a look at this game which CM 6000 makes a sacrifice but it
finds out in later moves that it cannnot win the game, and therefore goes for a
forced draw by repitition of moves. In move 25, CM makes a sacrifice or better
plays a combo but it backfires. I was puzzled by CM6000 sacrifice. The problem
became apparent that simplifying the position to a K&P endgame would be lost for
white. It was nice to see that Fritz returned the favor by giving back material.
[Event "Level=Blitz:30'+5"]
[Site "?"]
[Date "1998.12.09"]
[Round "?"]
[White "CM 6000, 16MB"]
[Black "Fritz 5.16, 8MB"]
[Result "1/2-1/2"]
[ECO "B33"]
[PlyCount "79"]
[EventDate "1998.12.09"]
{8128kB, f5book.ctg} 1. e4 1... c5 2. Nf3 e6 3. d4 cxd4 4. Nxd4 Nc6 5. Nb5 Nf6
6. N1c3 d6 7. Bf4 e5 8. Bg5 a6 9. Na3 b5 10. Bxf6 gxf6 11. Nd5 f5 12. c3 Bg7
13. exf5 Bxf5 14. Qf3 {25} 14... Bd7 {0} 15. Be2 {54} 15... Rc8 {0.34/11 113}
16. O-O {52} 16... Be6 {0.06/10 106} 17. Nc2 {51} 17... Nd4 {-0.06/10 85} 18.
Nxd4 {62} 18... exd4 {-0.31/9 0} 19. cxd4 {62} 19... Bxd4 {0.06/10 96} 20. Rac1
{6} 20... Bxb2 {-0.22/11 89} 21. Rxc8 {6} 21... Qxc8 {-0.59/9 5} 22. Bd3 {23}
22... Bd4 {-0.66/9 46} 23. Re1 {51} 23... Qc5 {-0.63/9 0} 24. Nf4 {49} 24...
O-O {-0.25/10 82} 25. Bxh7+ {7} 25... Kxh7 {0.00/12 47} 26. Nxe6 {6} 26...
Bxf2+ {0.00/11 23} 27. Kf1 {11} 27... fxe6 {0.00/12 77} 28. Qxf8 {7} 28... Bxe1
{0.00/12 32} 29. Qe7+ {2} 29... Kg6 {-0.56/11 45} 30. Qxe6+ {5} 30... Kg5 {
-0.34/10 41} 31. Qg8+ {6} 31... Kf6 {-0.09/10 37} 32. Qh8+ {42} 32... Ke6 {
-0.03/11 65} 33. Qe8+ {49} 33... Kf5 {-0.09/10 0} 34. Qf7+ {33} 34... Ke5 {
0.00/11 61} 35. Qh5+ {7} 35... Kd4 {0.00/10 14} 36. Qd1+ {22} 36... Kc4 {
0.03/9 16} 37. Qb3+ {21} 37... Kd4 {0.00/1 0} 38. Qd1+ {30} 38... Kc4 {
0.00/11 3} 39. Qb3+ {41} 39... Kd4 {0.00/1 0} 40. Qd1+ {37} 1/2-1/2
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.