Winning starts with what you know
The new version 18 offers completely new possibilities for chess training and analysis: playing style analysis, search for strategic themes, access to 6 billion Lichess games, player preparation by matching Lichess games, download Chess.com games with built-in API, built-in cloud engine and much more.
Pawn and Two Move Handicap MatchIn the mid 1800s, there were no chess ratings. Players were generally classified based on how much handicap they needed from the top players to have fairly even chances. There were six major handicaps used for serious play:
Now that Rybka 3 is rated so far above all human players, it makes sense to try to determine what is a fair handicap for players at different levels of skill against her. In order to determine this, we have been running some matches at different handicaps against strong players at various levels, assuming a reasonably serious rate of play. We used 30’+30” per move increment for each side as our standard.
Since Rybka 3 won a match from GM Dzindzichashvili at pawn and move by 2.5-1.5 it seems plausible that a typical IM should take pawn and two moves, a typical FM pawn and three moves, and a typical Candidate Master knight odds. We had tried to give FM John Meyer (FIDE 2284) knight odds in a four game match a while back with an intermediate Rybka version, but he was too strong for this and Rybka lost 0-4. So we tried a rematch at pawn and three moves, and this time Rybka won by 3-1.
Since Rybka was able to win a match vs. an FM at pawn and three, I reasoned that pawn and two should be appropriate for an IM. I chose IM Eugene Meyer (the younger brother of John Meyer) for this match. Eugene is an amateur player, rated 2443 FIDE, but back in the 1980s he came very close to earning the GM title; he exceeded the required 2500 rating and made two norms. He surely would have become a GM if he made chess his career. He even tied for first in the World Open Championship back then. Now he’s 56 years old and rather inactive in chess, but he is still quite a strong player. He agreed to play a four game match this way in one day, for a prize fund of $200 if he won the match (half for a drawn match). As in other recent matches, Rybka ran on an Octal computer at 3 Ghz, and used my handicap book (enhanced for this match).
Pawn and two moves: the Meyer-Rybka starting position
Two things to note about this match: First, I set a high contempt value (100) so that Meyer could not easily make a draw if things started to go badly. In general for handicap play I recommend setting contempt at about 2/3 of Rybka’s evaluation of the initial handicap position. Second, I switched around between the three different Rybka versions to maximize variety and to look for any differences. I can’t say that I could tell much difference, but it’s hard to play aggressively when giving pawn and two moves.
In the first game Meyer was surprised on move one by …Nc6, and I believe that his response f4 was already inaccurate due to the strength of …d5 in reply. Still he kept a solid advantage and declined to repeat moves around move 30 with reasonable justification.
Meyer,Eugene B - Rybka 3
Exhibition Match pawn and 2 moves
Potomac/Maryland (1), 09.08.2008
1.d4 Nc6 2.f4 d5 3.e5 Bf5 4.c3 Nh6 5.Nf3 e6 6.Be2 Be7 7.Ng5 Qd7 8.Be3
Bg6 9.Nd2 0-0-0 10.g4 Nf7 11.h4 Nxg5 12.fxg5 Rdf8 13.h5 Be8 14.Qc2 g6 15.0-0-0
Rhg8 16.Nf3 Na5 17.b3 Kb8 18.Rdf1 Qd8 19.Bd3 a6 20.Kb2 Bb5 21.hxg6 hxg6 22.Rh6
Bxd3 23.Qxd3 Rf7 24.Nd2 Rxf1 25.Nxf1 Qe8 26.Nd2 Qf7 27.Qf1 Qg7 28.Qh3 Qf8 29.Qf3
Qg7 30.Qh3 Qf8 31.Nf3 Ba3+ 32.Kc2 c5 33.dxc5 Bxc5 34.Bxc5 Qxc5 35.Nd4 Nc6
Here 36.Nxc6+ kept the edge, but Meyer went wrong with 36.Qe3? Nxd4+ 37.Qxd4 Qa3 38.Qb6?? (he could still equalize or keep a small edge by 38.Kb1 Rf8 39.Rh1). 38...Qxa2+ 39.Kd3 Qb1+ 40.Kd2 Qb2+ 41.Kd3 Rc8 42.Qb4 a5 43.Qxa5 d4 44.Kxd4 Qd2+. Here Rybka announced (silently) mate in 14! 45.Ke4 Rf8 0-1. [Click to replay]
In the second game Rybka Dynamic chose 1…d6 and Meyer replied 2.Nf3. After the match he said that f4 should be played first against the missing f-pawn, to minimize the value of the open file, and at least in this position I won’t disagree. The placement of the bishop on c4 is a little questionable due to …d5, but still White kept a nice advantage. He chose to return the pawn to reach a favorable good bishop vs. bad bishop endgame, but some inaccurate moves gave Rybka equality and then Meyer overlooked 23…e5 and was hopelessly lost.
Meyer,Eugene B - Rybka 3 "Dynamic"
Exhibition Match pawn and 2 moves
Potomac/Maryland (2), 09.08.2008
1.d4 d6 2.Nf3 Nf6 3.Nc3 e6 4.Bc4 Be7 5.0-0 d5 6.Bd3 0-0 7.Re1 Nc6 8.Bg5
h6 9.Bh4 Nxe4 10.Bxe7 Nxc3 11.Bxd8 Nxd1 12.Raxd1 Nxd8 13.Ne5 Nc6 14.Nxc6 bxc6
15.g3 g5 16.c4 Bd7 17.cxd5 cxd5 18.Rc1 c6 19.Rc3 Rab8 20.Re2 Kg7 21.a3 a5 22.b3
Rf3
23.Rec2? e5 24.Bf1 Rxc3 25.Rxc3 0-1. [Click to replay]
In game three Rybka “Human” Meyer got a normal advantage for the handicap but Rybka’s plan of …a6, …b5, and …c5 was indeed quite human. Probably Meyer should have played 11.b3 to keep more space and then 19.Ng4 was much better than the knight exchange. The move 24.g4?, although natural, turns out to allow Black to reach a decent endgame, and after 27.Kg2?! he was obviously worse but had no trouble holding the draw.
Meyer,Eugene B - Rybka 3 "Human"
Exhibition Match pawn and 2 moves
Potomac/Maryland, 09.08.2008
1.d4 d6 2.f4 g6 3.Nf3 Bg7 4.c3 e6 5.Bd3 Nh6 6.0-0 0-0 7.Kh1 Qe7 8.Na3
a6 9.Bd2 b5 10.Nc2 c5 11.a4 c4 12.Be2 Bb7 13.Ng5 Nf7 14.Nxf7 Qxf7 15.Bf3 Nd7
16.Ne3 Nf6 17.Qc2 d5 18.exd5 Nxd5 19.Nxd5 Bxd5 20.Bxd5 exd5 21.axb5 axb5 22.Rxa8
Rxa8 23.f5 gxf5 24.g4 Qe6 25.Qxf5 Qxf5 26.gxf5 Kf7 27.Kg2 Ra2 28.Bc1 b4 29.Rf2
bxc3 30.bxc3 Rxf2+ 31.Kxf2 Kf6 32.Kf3 Kxf5 33.Bf4 Bf6 34.Bc1 Bh4 35.Bd2 Bd8
36.Bc1 Bc7 37.h3 Bd8 38.Bd2 Bh4 39.Bc1 Be1 40.Bb2 Kg5 41.Ke2 Bg3 42.Kf3 Kh4
43.Kg2 Be1 44.Ba1 h6 45.Bb2 Kg5 46.Kf3 h5 47.Ke2 Bg3 48.Kf3 Kh4 49.Kg2 Bf4 50.Ba3
Kg5 51.Kf3 Kf5 52.Bb4 Bc7 53.Ba3 Kg6 draw. [Click
to replay]
For the final game (also against “Human” Rybka) Meyer decided to go all out for a win, as the match was already lost. However 9.f4? was too aggessive, and after rejecting 10.Bxf7+ due to incorrect analysis he soon landed in a position where Black had the bishop pair and a tremendous position for the pawn. Perhaps another Rybka could have held on, but everything collapsed in a few moves.
Meyer,Eugene B - Rybka 3
Exhibition Match pawn and 2 moves
Potomac/Maryland (4), 09.08.2008
1.d4 Nc6 2.Nf3 d6 3.Bc4 Nf6 4.Qe2 e5 5.dxe5 Nxe5 6.Nxe5 dxe5 7.Nc3
c6 8.Bg5 Bc5 9.f4 Bg4 10.Bxf6 Bxe2 11.Bxd8 Bxc4 12.Bc7 exf4 13.Bxf4 0-0 14.Bg3
Rad8 15.Rd1 Bd4 16.Ne2 Bxb2 17.Rb1 Bf6 18.Rxb7 Bg5 0-1. [Click
to replay]
Meyer was not in good form this match, but I think the 3.5-0.5 score and the previous Dzindzi match indicate that pawn and two moves is probably a fair handicap between Rybka 3 (on good hardware) and a very strong IM or low-rated GM. Pawn and three might be more equitable for an average IM. But since knight odds is too large for an FM, it seems players of that level need some intermediate handicap.
Replay games provided by Holger Lieske, SysOp on Playchess
| Multi-processor version: Deep Rybka 3 | 99.90 Euro | |
| Single processor version: Rybka 3 | 49.99 Euro | |
| Rybka 3 Book | 24.99 Euro |
Note that Rybka 3 includes a database of one million games, and that the purchase of the program entitles you to one year of access to the chess server Playchess.com. Rybka 3 is a UCI engine, with 32 and 64-bit versions included in the package. Rybka can be made the default engine in ChessBase 10.
