Mexico - Zappa 5.5 v Rybka 4.5- Zappa wins $10,000

[Mark Uniacke]

[Event "Blitz:10'+15"]
[Site "?"]
[Date "2007.09.26"]
[Round "9"]
[White "Zappa Mexico X64 "]
[Black "Rybka"]
[Result "1/2-1/2"]
[ECO "B17"]
[PlyCount "180"]
[EventDate "2007.??.??"]
[TimeControl "600+15"]

{4MB, ZapM X Wihte.ctg, ZAPPA} 1. e4 {B/0 0} c6 {8} 2. d4 {B/0 0} d5 {6} 3. Nc3
{B/0 0} dxe4 {11} 4. Nxe4 {B/0 0} Nd7 {7} 5. Bc4 {B/0 0} Ngf6 {10} 6. Ng5 {
B/0 0} e6 {9} 7. Qe2 {B/0 0} Nb6 {9} 8. Bd3 {B/0 0} h6 {9} 9. N5f3 {B/0 0} c5 {
10} 10. dxc5 {B/0 0} Bxc5 {8} 11. Ne5 {B/0 0} Nbd7 {9} 12. Ngf3 {B/0 0} Nxe5 {
13} 13. Nxe5 {B/0 0} O-O {7} 14. O-O {B/0 0} b6 {8} 15. Bf4 {B/0 0} Qd4 {707}
16. Bg3 {0.38/18 160} Bb7 {190} 17. c4 {0.39/19 0} Qd8 {74} 18. Rad1 {
0.41/19 50} Qe7 {104} 19. Kh1 {0.38/19 58} Rfd8 {136} 20. a3 {0.27/18 130} Bd4
{(Sd7) 127} 21. Bh4 {0.24/18 278} Rd6 {(Lxe5) 73} 22. f4 {0.30/18 169} Rad8 {
(a5) 10} 23. Bb1 {0.28/17 186} Kf8 {35} 24. Rd3 {0.28/17 103} Qc7 {12} 25. b4 {
0.26/17 118} a5 {(La6) 72} 26. Re1 {0.12/17 165} Ba1 {99} 27. Rc1 {0.19/18 314}
Rd4 {(axb4) 72} 28. Rcd1 {0.16/17 134} Rxd3 {18} 29. Bxd3 {0.00/18 187} Bc3 {9}
30. bxa5 {-0.09/18 144} Bxa5 {64} 31. Bg3 {-0.06/18 33} Kg8 {61} 32. Bf2 {
-0.20/18 72} Bc3 {(Se4) 59} 33. Bg1 {-0.23/18 132} Ne4 {196} 34. Qc2 {
-0.19/20 0} Bxe5 {321} 35. fxe5 {-0.16/21 0} Qxe5 {(Sc5) 45} 36. Bxb6 {
-0.24/18 159} Rd6 {(Td7) 9} 37. Bg1 {-0.14/18 104} Qh5 {(Td7) 23} 38. Be3 {
-0.21/17 108} Ng3+ {(Lc6) 10} 39. Kg1 {0.28/7 1} Nf5 {(De5) 22} 40. Bf4 {
-0.80/17 131} Rd4 {458} 41. Be2 {-0.96/19 26} Qh4 {13} 42. Bc1 {-0.95/18 102}
Qd8 {10} 43. Rf1 {-0.95/18 114} Qc7 {11} 44. Rf2 {-0.96/18 118} Rh4 {18} 45. h3
{-0.95/19 105} Nd4 {32} 46. Qb2 {-0.96/19 107} Nxe2+ {8} 47. Qxe2 {-0.96/19 30}
Re4 {25} 48. Qf1 {-0.96/18 4} Rxc4 {(La6) 33} 49. Bf4 {-0.91/17 34} Qc6 {11}
50. Rb2 {-0.89/17 27} Rc3 {44} 51. Qb5 {-0.99/18 0} Qxb5 {14} 52. Rxb5 {
-1.04/18 0} Bd5 {21} 53. Bd6 {-1.03/18 19} Rc1+ {30} 54. Kf2 {-1.11/18 6} Rc2+
{13} 55. Ke1 {-1.10/19 72} Rxg2 {(Lxg2) 14} 56. Rc5 {-1.25/18 34} g5 {11} 57.
Rc3 {-1.32/18 18} Ra2 {16} 58. h4 {-1.36/18 29} f5 {(gxh4) 12} 59. hxg5 {
-1.24/17 31} hxg5 {7} 60. Rg3 {-1.25/18 22} g4 {26} 61. Rc3 {-1.43/18 3} Kf7 {
30} 62. Be5 {-1.74/18 24} Kg6 {(Lf3) 37} 63. Rc8 {-1.95/18 47} Rxa3 {(Lf3) 29}
64. Rg8+ {-1.61/17 29} Kf7 {15} 65. Rg7+ {-1.61/18 0} Kf8 {13} 66. Kd2 {
-1.93/18 38} Ra4 {(Ta5) 9} 67. Kc3 {-1.93/17 59} Re4 {(Ta1) 11} 68. Bf6 {
-1.92/16 1} Bc4 {(Tc4+) 18} 69. Kd2 {-2.14/18 48} Be2 {14} 70. Rg5 {-2.34/18 35
} Kf7 {(Lf3) 10} 71. Bh8 {-2.49/18 42} f4 {(Lf3) 35} 72. Re5 {-1.98/19 35} Rxe5
{(Lf3) 29} 73. Bxe5 {-1.98/12 1} f3 {10} 74. Ke3 {-2.15/26 26} Kg6 {10} 75. Bd6
{-2.15/26 30} Kf6 {(Kf5) 56} 76. Bg3 {-2.15/26 24} Kf5 {(Kg5) 136} 77. Bh2 {
-2.15/28 42} e5 {12} 78. Bg3 {-2.15/28 0} Bc4 {(Ld1) 64} 79. Bh4 {-2.15/30 60}
Ba6 {(Lb3) 56} 80. Bg3 {-2.15/28 35} Ke6 {(Kf6) 49} 81. Be1 {-2.15/27 23} Kd5 {
(Kf6) 47} 82. Bh4 {-2.15/29 40} Bb5 {(Ke6) 11} 83. Be1 {-2.15/27 30} Be2 {
(Lc6) 95} 84. Bf2 {-2.15/28 27} Bc4 {94} 85. Be1 {-2.15/28 0} Bb3 {(Ke6) 83}
86. Kd3 {-2.15/26 22} Kc6 {(Lc4+) 41} 87. Ke4 {-2.15/23 22} Kd6 {13} 88. Bb4+ {
-2.15/24 9} Kd7 {(Ke6) 24} 89. Be1 {-2.15/25 24} Ke6 {13} 90. Ke3 {-2.15/27 8}
Bc2 {(Ld5) 44} 1/2-1/2

Absolutely incredible.

If nothing else Zappa and Erdo have proven that you always have a chance against Rybka!

71...f4???? perhaps the worst blunder in modern computer chess history.

[fen]7B/5k2/4p3/5pR1/4r1p1/8/3Kb3/8 b - - 0 71[/fen]

[Dylan Sharp]

So, what happens if Rybka wins tomorrow and we have a 5-5? Will there be a tie-break game?

[Ted Summers]

So, what happens if Rybka wins tomorrow and we have a 5-5? Will there be a tie-break game?

No, the $10,000 is to be split between both teams

[Dylan Sharp]

So, Zappa can no longer lose, and only needs a draw to win. Was anybody expecting a result like this before the match started?

[Ted Summers]

So, Zappa can no longer lose, and only needs a draw to win. Was anybody expecting a result like this before the match started?

No, I was not expecting this. But I am happy that Rybka is finally making a match of it. This coming from someone who wants Zappa to win. The mistakes have been so human like. Who would have thought.

[Anthony C]

A black day for the Zappa team. Zappa got a very good position with Black in a Ruy Lopez, got in time trouble, blundered, and lost. After Qd8 Qb8 Qg8, White has the amazing plan of g5-g6, Kh5, and Bxh6, and can win despite being down a full exchange. Black could probably play for a win after e4 instead of Qxb5, or a draw after c3 instead of Qd8.

In the second game, Rybka simply outplayed Zappa, and again got a position where its deep search was able to find some tricks. If you look at the position 5 full moves before Rxc4 it is hard to believe that Black is winning. However in the endgame Rybka traded rooks and despite having an eval of +7 (Zappa only +2 for White) the game was drawn.

I actually think this match showcases the differences between Rybka and Zappa quite well. Zappa's knowledge is in its evaluation and its relatively accurate guesses about positions. Rybka's knowledge is in its intelligent search which lets it see very deeply very quickly. Interestingly most people associate a gradual score rise with a superiority in evaluation, when in fact it usually indicates a superiority in search, as the opponent is making many little blunders. In contrast when one engine has a superiority in evaluation the score will usually rise sharply after a move because suddenly the engine can secure a positional advantage (whatever the term is that the other engine is unaware of).

Two years ago I would have told you that a good evaluation is more important than a good search, but now I am not so sure. At the very least Rybka has proven that even today's engines on big hardware (Zappa) still make many search errors, especially at such quick timecontrols as 60+20.

cheers,

anthony

[Terry McCracken]

A black day for the Zappa team. Zappa got a very good position with Black in a Ruy Lopez, got in time trouble, blundered, and lost. After Qd8 Qb8 Qg8, White has the amazing plan of g5-g6, Kh5, and Bxh6, and can win despite being down a full exchange. Black could probably play for a win after e4 instead of Qxb5, or a draw after c3 instead of Qd8.

In the second game, Rybka simply outplayed Zappa, and again got a position where its deep search was able to find some tricks. If you look at the position 5 full moves before Rxc4 it is hard to believe that Black is winning. However in the endgame Rybka traded rooks and despite having an eval of +7 (Zappa only +2 for White) the game was drawn.

I actually think this match showcases the differences between Rybka and Zappa quite well. Zappa's knowledge is in its evaluation and its relatively accurate guesses about positions. Rybka's knowledge is in its intelligent search which lets it see very deeply very quickly. Interestingly most people associate a gradual score rise with a superiority in evaluation, when in fact it usually indicates a superiority in search, as the opponent is making many little blunders. In contrast when one engine has a superiority in evaluation the score will usually rise sharply after a move because suddenly the engine can secure a positional advantage (whatever the term is that the other engine is unaware of).

Two years ago I would have told you that a good evaluation is more important than a good search, but now I am not so sure. At the very least Rybka has proven that even today's engines on big hardware (Zappa) still make many search errors, especially at such quick timecontrols as 60+20.

cheers,

anthony

I'm very glad Anthony that you take the time and give your analysis on the match and individual games.

I wish Vas would do the same! Of course he does have his own forum, still it would nice if he'd give his take on the match here as well.

Best,
Terry

[Steve B]

A black day for the Zappa team. Zappa got a very good position with Black in a Ruy Lopez, got in time trouble, blundered, and lost. After Qd8 Qb8 Qg8, White has the amazing plan of g5-g6, Kh5, and Bxh6, and can win despite being down a full exchange. Black could probably play for a win after e4 instead of Qxb5, or a draw after c3 instead of Qd8.

In the second game, Rybka simply outplayed Zappa, and again got a position where its deep search was able to find some tricks. If you look at the position 5 full moves before Rxc4 it is hard to believe that Black is winning. However in the endgame Rybka traded rooks and despite having an eval of +7 (Zappa only +2 for White) the game was drawn.

I actually think this match showcases the differences between Rybka and Zappa quite well. Zappa's knowledge is in its evaluation and its relatively accurate guesses about positions. Rybka's knowledge is in its intelligent search which lets it see very deeply very quickly. Interestingly most people associate a gradual score rise with a superiority in evaluation, when in fact it usually indicates a superiority in search, as the opponent is making many little blunders. In contrast when one engine has a superiority in evaluation the score will usually rise sharply after a move because suddenly the engine can secure a positional advantage (whatever the term is that the other engine is unaware of).

Two years ago I would have told you that a good evaluation is more important than a good search, but now I am not so sure. At the very least Rybka has proven that even today's engines on big hardware (Zappa) still make many search errors, especially at such quick timecontrols as 60+20.

cheers,

anthony

I'm very glad Anthony that you take the time and give your analysis on the match and individual games.

I wish Vas would do the same! Of course he does have his own forum, still it would nice if he'd give his take on the match here as well.

Best,
Terry

and of course we also are treated to Mark's Comments and analysis here as well

its very special to have a Premiere Elite World programmer commenting on the games of other commercial engines as they are locked in fatal mortal combat with mucho dinero on the line

History In the making regards
Steve

[Father]

Hello.

Some weeks ago, I said, that Zappa will be winning the match in "talkchess". And I said that Rybka was as Karpov in his best moment, and that, Zappa as Kasparov in his best moment, "in rybka forum".
"Rybka will be thinking how to have a promotion of a pawn, and Zappa will be atacking the Rybka King", the Zappa will be the winner.
I want to be sincere.
This level of chess is too high for my intelligence, and my owns limits as chess player.
Does have had the human beings any person on history that could have played as Rybka an Zappa has played.?
How many elo, could have these two machines, for example, if they will be playing, with the eight chess human beings that are playing in Mexico for the WCCA?
Thank in advance,
Pablo

NOTE.
CONGRATULLATIONS TO VASIK AND RYBKA TEAM, AND FOR, ANTONY, ERDO, HARVEY, AND ALL ZAPPA TEAM.
PERSONALLY, I HAVE RECIVED A TOP CHESS LESSON.

[Terry McCracken]

Hello.

Some weeks ago, I said, that Zappa will be winning the match in "talkchess". And I said that Rybka was as Karpov in his best moment, and that, Zappa as Kasparov in his best moment, "in rybka forum".
"Rybka will be thinking how to have a promotion of a pawn, and Zappa will be atacking the Rybka King", the Zappa will be the winner.
I want to be sincere.
This level of chess is too high for my intelligence, and my owns limits as chess player.
Does have had the human beings any person on history that could have played as Rybka an Zappa has played.?
How many elo, could have these two machines, for example, if they will be playing, with the eight chess human beings that are playing in Mexico for the WCCA?
Thank in advance,
Pablo

NOTE.
CONGRATULLATIONS TO VASIK AND RYBKA TEAM, AND FOR, ANTONY, ERDO, HARVEY, AND ALL ZAPPA TEAM.
PERSONALLY, I HAVE RECIVED A TOP CHESS LESSON.

There are errors from both engines that humans would never make.

Trying to Set the Record Straight Regards,
Terry

P.S. As for the last game, flip a quarter

[Terry McCracken]

A black day for the Zappa team. Zappa got a very good position with Black in a Ruy Lopez, got in time trouble, blundered, and lost. After Qd8 Qb8 Qg8, White has the amazing plan of g5-g6, Kh5, and Bxh6, and can win despite being down a full exchange. Black could probably play for a win after e4 instead of Qxb5, or a draw after c3 instead of Qd8.

In the second game, Rybka simply outplayed Zappa, and again got a position where its deep search was able to find some tricks. If you look at the position 5 full moves before Rxc4 it is hard to believe that Black is winning. However in the endgame Rybka traded rooks and despite having an eval of +7 (Zappa only +2 for White) the game was drawn.

I actually think this match showcases the differences between Rybka and Zappa quite well. Zappa's knowledge is in its evaluation and its relatively accurate guesses about positions. Rybka's knowledge is in its intelligent search which lets it see very deeply very quickly. Interestingly most people associate a gradual score rise with a superiority in evaluation, when in fact it usually indicates a superiority in search, as the opponent is making many little blunders. In contrast when one engine has a superiority in evaluation the score will usually rise sharply after a move because suddenly the engine can secure a positional advantage (whatever the term is that the other engine is unaware of).

Two years ago I would have told you that a good evaluation is more important than a good search, but now I am not so sure. At the very least Rybka has proven that even today's engines on big hardware (Zappa) still make many search errors, especially at such quick timecontrols as 60+20.

cheers,

anthony

I'm very glad Anthony that you take the time and give your analysis on the match and individual games.

I wish Vas would do the same! Of course he does have his own forum, still it would nice if he'd give his take on the match here as well.

Best,
Terry

and of course we also are treated to Mark's Comments and analysis here as well

its very special to have a Premiere Elite World programmer commenting on the games of other commercial engines as they are locked in fatal mortal combat with mucho dinero on the line

History In the making regards
Steve

Indeed, Steve!

Waiting for the Rematch Regards,
Terry

[joseMfc]

Both engines have made some mistakes, but also we have seen very good games, (Zappa's game 3, Rybka's game .

But Rybka's errors have been quite astonishing ... game 4 and game 9.

The end game with different colours bishops is, for humans, very easy to understand ...

Is this a bug in Rybka code or is so difficult for computers?.

Today last game .... how exciting match!

[Uri Blass]

A black day for the Zappa team. Zappa got a very good position with Black in a Ruy Lopez, got in time trouble, blundered, and lost. After Qd8 Qb8 Qg8, White has the amazing plan of g5-g6, Kh5, and Bxh6, and can win despite being down a full exchange. Black could probably play for a win after e4 instead of Qxb5, or a draw after c3 instead of Qd8.

In the second game, Rybka simply outplayed Zappa, and again got a position where its deep search was able to find some tricks. If you look at the position 5 full moves before Rxc4 it is hard to believe that Black is winning. However in the endgame Rybka traded rooks and despite having an eval of +7 (Zappa only +2 for White) the game was drawn.

I actually think this match showcases the differences between Rybka and Zappa quite well. Zappa's knowledge is in its evaluation and its relatively accurate guesses about positions. Rybka's knowledge is in its intelligent search which lets it see very deeply very quickly. Interestingly most people associate a gradual score rise with a superiority in evaluation, when in fact it usually indicates a superiority in search, as the opponent is making many little blunders. In contrast when one engine has a superiority in evaluation the score will usually rise sharply after a move because suddenly the engine can secure a positional advantage (whatever the term is that the other engine is unaware of).

Two years ago I would have told you that a good evaluation is more important than a good search, but now I am not so sure. At the very least Rybka has proven that even today's engines on big hardware (Zappa) still make many search errors, especially at such quick timecontrols as 60+20.

cheers,

anthony

I think that it is not defined what is more important good evaluation or good search because the question is how much advantage you have.

Note that based on my experience it is not the first time when rybka is simply lying about evaluating endgames.

In 3 out of 12 correspondence games that I played rybka showed big scores for drawn positions during analysis(and when I say big scores I mean at least +2).

I feel that rybka's score in endgame is simply unreliable and this is the main reason that I also think that rybka has inferior evaluation relative to zappa.

I cannot say much about middle game evaluation because unlike endgames I do not think that I know the right evaluation and I think that humans do not know much about the middle game relative to the endgame(in the endgame after analysis humans can often find the expected result when in the middle game it is hard to know).

Uri

[Terry McCracken]

A black day for the Zappa team. Zappa got a very good position with Black in a Ruy Lopez, got in time trouble, blundered, and lost. After Qd8 Qb8 Qg8, White has the amazing plan of g5-g6, Kh5, and Bxh6, and can win despite being down a full exchange. Black could probably play for a win after e4 instead of Qxb5, or a draw after c3 instead of Qd8.

In the second game, Rybka simply outplayed Zappa, and again got a position where its deep search was able to find some tricks. If you look at the position 5 full moves before Rxc4 it is hard to believe that Black is winning. However in the endgame Rybka traded rooks and despite having an eval of +7 (Zappa only +2 for White) the game was drawn.

I actually think this match showcases the differences between Rybka and Zappa quite well. Zappa's knowledge is in its evaluation and its relatively accurate guesses about positions. Rybka's knowledge is in its intelligent search which lets it see very deeply very quickly. Interestingly most people associate a gradual score rise with a superiority in evaluation, when in fact it usually indicates a superiority in search, as the opponent is making many little blunders. In contrast when one engine has a superiority in evaluation the score will usually rise sharply after a move because suddenly the engine can secure a positional advantage (whatever the term is that the other engine is unaware of).

Two years ago I would have told you that a good evaluation is more important than a good search, but now I am not so sure. At the very least Rybka has proven that even today's engines on big hardware (Zappa) still make many search errors, especially at such quick timecontrols as 60+20.

cheers,

anthony

I think that it is not defined what is more important good evaluation or good search because the question is how much advantage you have.

Note that based on my experience it is not the first time when rybka is simply lying about evaluating endgames.

In 3 out of 12 correspondence games that I played rybka showed big scores for drawn positions during analysis(and when I say big scores I mean at least +2).

I feel that rybka's score in endgame is simply unreliable and this is the main reason that I also think that rybka has inferior evaluation relative to zappa.

I cannot say much about middle game evaluation because unlike endgames I do not think that I know the right evaluation and I think that humans do not know much about the middle game relative to the endgame(in the endgame after analysis humans can often find the expected result when in the middle game it is hard to know).

Uri

Interesting, I find complex endgames far more subtle and difficult than any middlegame position. Infact I think most humans do, including computers.

My 2 Cents after 27+ Years of Play Regards,
Terry

[Harvey Williamson]

If you just want all the games without wading through this long thread rhen look here http://hiarcs.net/forums/viewtopic.php?t=320

Although to find this post I guess you have just waded through this long thread