GM Bartlomiej Macieja Vs Chess Engines

[Michael Sherwin]

Do you happen to know the time control for those games?

15 minutes per game, plus 10 seconds per move. Four games in the same morning, with very short breaks between. Very difficult conditions for a human, and I am actually quite impressed by Macieja's performance.

I'll write a detailed report from the Macieja vs engines match and the IOPCCC some time after I get home.

Tord

Tord,
I am not impressed at all with the GM performance,what conditions are you talking about
I mean this is a 2600 rated proffesional chess player playing against amateur engines,strong yes,but still,and the programs don't use even a decent cutting edge hardware.....
The GM result is a disasterin my opinion,similar to Adam's one against Hydra....

I am starting to believe that humans are not really very good at chess. Relative to each other, some humans seem fantastically great, but even the games of the 'fantastically great' are full of errors. The modern chess engines do not let many mistakes go unpunished. Now just think back to 1996 when Wchess and other top engines were decimating human grandmasters while the programs were running on 90MHz pentiuns. This should put in perspective what todays grandmasters are up against.

What is the strength of a human grandmaster anyway? For the most part it is creativity. A grandmaster just creates things that are beyond me when I play them and I loose. Even if my chess is more technically correct, I fail to take advantage of their mistakes, then I loose to what I did not see coming! Or something like that.

A human playing creativly plays directly into the strength of a computer which will punish any mistakes. The human is damned if he does and damned if he doesn't. If a creative human plays with out creativity he usually looses interest in the game and then self destructs.

I personally do well against free Rybka. Rybka completely stiffles any creativity that I may have and I am bored to tears, but I refuse to selfdestruct. I have never beaten Rybka--only drawn.

Edit: I loose also, but only about half the games. My score against free Rybka is about 25%.

[Dr.Wael Deeb]

Do you happen to know the time control for those games?

15 minutes per game, plus 10 seconds per move. Four games in the same morning, with very short breaks between. Very difficult conditions for a human, and I am actually quite impressed by Macieja's performance.

I'll write a detailed report from the Macieja vs engines match and the IOPCCC some time after I get home.

Tord

Tord,
I am not impressed at all with the GM performance,what conditions are you talking about
I mean this is a 2600 rated proffesional chess player playing against amateur engines,strong yes,but still,and the programs don't use even a decent cutting edge hardware.....
The GM result is a disasterin my opinion,similar to Adam's one against Hydra....

I am starting to believe that humans are not really very good at chess. Relative to each other, some humans seem fantastically great, but even the games of the 'fantastically great' are full of errors. The modern chess engines do not let many mistakes go unpunished. Now just think back to 1996 when Wchess and other top engines were decimating human grandmasters while the programs were running on 90MHz pentiuns. This should put in perspective what todays grandmasters are up against.

What is the strength of a human grandmaster anyway? For the most part it is creativity. A grandmaster just creates things that are beyond me when I play them and I loose. Even if my chess is more technically correct, I fail to take advantage of their mistakes, then I loose to what I did not see coming! Or something like that.

A human playing creativly plays directly into the strength of a computer which will punish any mistakes. The human is damned if he does and damned if he doesn't. If a creative human plays with out creativity he usually looses interest in the game and then self destructs.

I personally do well against free Rybka. Rybka completely stiffles any creativity that I may have and I am bored to tears, but I refuse to selfdestruct. I have never beaten Rybka--only drawn.

Edit: I loose also, but only about half the games. My score against free Rybka is about 25%.

Well said Michael,an interesting point of view
I play the engines myself,but I include myself in the league where I belong,but I am quite pleased by some games of mine even against weak engines....
In my rating list I am rated 2032 right now when Fruit 2.2.1 is rated 2720


[Event "DCCW_RL 2008"]
[Date "2008.06.01"]
[White "Atak 4.99 Linka wb_2038"]
[Black "Dr.Wael Deeb_2042"]
[Result "1/2-1/2"]
[ECO "D35"]
[Opening "Queen’s Gambit_Exchange Variation"]

1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. cxd5 exd5 5. Bg5 c6 6. e3
h6 7. Bxf6 Qxf6 8. Nf3 Bd6 9. Be2 O-O 10. Qb3 Qg6 11. O-O
Na6 12. Bxa6 bxa6 13. Qa4 Bb7 14. Rac1 Rfe8 15. Qc2 Qxc2
16. Rxc2 a5 17. Na4 Ba6 18. Rxc6 Bxf1 19. Rxd6 Bb5 20. Nc3
Rab8 21. Rxd5 Bc6 22. Rxa5 Bxf3 23. gxf3 Rxb2 24. Rxa7 Rc8 25. Ra3 f5 26. Kg2 Rd2 27. h3 Rd3 28. Nb5 Rxa3 29. Nxa3 Ra8 30. Nc4 Rxa2 31. d5 Ra4 32. Ne5 Kf8 33. Nc6 Rc4 34. Nd4 g6 35. Nc6 Rc5 36. e4 Kf7 37. Ne5+ Kg7 38. d6 Rc8 39. exf5 gxf5 40. d7 Rd8 41. f4 Kf6 42. Kg3 Kg7 1/2-1/2

[Event "DCCW_RL 2008"]
[Date "2008.06.02"]
[White "Dr.Wael Deeb_2042"]
[Black "Atak 4.99 Linka wb_2038"]
[Result "1-0"]
[ECO "A03"]
[Opening "Bird’s Opening"]


1. f4 d5 2. Nf3 g6 3. g3 Bg7 4. Bg2 c6 5. O-O Nf6 6. d4 O-O
7. c3 Bf5 8. Qb3 b6 9. Ne5 e6 10. Nd2 Qc7 11. e3 c5 12. Re1
Nbd7 13. Ndf3 Nxe5 14. Nxe5 c4 15. Qd1 a5 16. g4 Be4 17. g5
Nd7 18. Bxe4 dxe4 19. Qe2 Nxe5 20. fxe5 Rad8 21. Qg2 Qc6
22. Rf1 Rfe8 23. Rf4 h6 24. Rxe4 Qa4 25. h4 Rf8 26. Rf4 b5 27. Rf1 hxg5 28. hxg5 Rd7 29. Qh2 b4 30. Qe2 b3 31. a3 Qc6 32. Qg4 Rfd8 33. e4 Qb6 34. Kg2 {12s} Kf8 35. Rf3 Ke7 36. Be3 Qc7 37. Raf1 Rf8 38. Qh4 Kd8 39. Qh7 Bh8 40. Qxg6 Kc8 41. Qh6 Rfd8 42. Qh7 Rf8 1-0

[Event "DCCW_RL 2008"]
[Date "2008.06.04"]
[White "Dr.Wael Deeb_2042"]
[Black "Bearded 06_29 wb_2040"]
[Result "1/2-1/2"]
[ECO "A03"]
[Opening "Bird’s Opening"]


1. f4 d5 2. Nf3 g6 3. g3 Bg7 4. Bg2 c6 5. O-O Nf6 6. d4 O-O
7. c3 Bf5 8. Qb3 Qc7 9. c4 dxc4 10. Qxc4 Nbd7 11. Nc3 Rad8
12. e3 a6 13. Nh4 Be6 14. Qe2 b5 15. b3 c5 16. Bb2 cxd4
17. exd4 Qd6 18. Rad1 Bg4 19. Bf3 Bxf3 20. Nxf3 Rc8 21. Ne5
Rfd8 22. Ne4 Nxe4 23. Qxe4 Nf6 24. Qe2 Nd5 25. Rc1 b4
26. Qf2 Nc3 27. Bxc3 bxc3 28. Rfd1 Qa3 29. Rc2 Qb4 30. Qe3
Rc7 31. Rd3 Rdc8 32. Kg2 Bxe5 33. dxe5 Rc6 34. Kf3 e6
35. Qe4 a5 36. Qxb4 axb4 37. Rd4 Rb8 38. Ke3 Ra6 39. Kd3
Kf8 40. Ke4 h5 41. h4 f5+ 42. Kd3 1/2-1/2

[Tord Romstad]

Tord,
I am not impressed at all with the GM performance,what conditions are you talking about

Fast time controls, hardly any breaks between the games. It's extremely tiring.

I mean this is a 2600 rated proffesional chess player playing against amateur engines,strong yes,but still,and the programs don't use even a decent cutting edge hardware.....

Glaurung and Booot are among the best amateur engines, and on modern hardware they are certainly far stronger than the Chess Genius version that beat Kasparov back in the 1990s. About the hardware: Glaurung was running on a 2.4 GHz Core 2 Duo MacBook, which is about as cutting edge as you can get on a laptop today. Booot was running on an even faster dual- or quad-core 3 GHz Intel CPU, but was only using one thread.

Matacz and Butcher are probably slightly weaker, but still formidable opponents in rapid chess, and Macieja played them when he was already tired from battling with Glaurung and Booot.

I think you and most others underestimate the difficulty of playing such a match. It is not at all like playing relaxed games against your favorite programs at home with your own computer. Playing in front of an audience, with money on the table, against unknown engines with unknown opening books, is quite a different thing. I suspect that most GMs wouldn't even have accepted the challenge.

Tord

[Dr.Wael Deeb]

Tord,
I am not impressed at all with the GM performance,what conditions are you talking about

Fast time controls, hardly any breaks between the games. It's extremely tiring.

I mean this is a 2600 rated proffesional chess player playing against amateur engines,strong yes,but still,and the programs don't use even a decent cutting edge hardware.....

Glaurung and Booot are among the best amateur engines, and on modern hardware they are certainly far stronger than the Chess Genius version that beat Kasparov back in the 1990s. About the hardware: Glaurung was running on a 2.4 GHz Core 2 Duo MacBook, which is about as cutting edge as you can get on a laptop today. Booot was running on an even faster dual- or quad-core 3 GHz Intel CPU, but was only using one thread.

Matacz and Butcher are probably slightly weaker, but still formidable opponents in rapid chess, and Macieja played them when he was already tired from battling with Glaurung and Booot.

I think you and most others underestimate the difficulty of playing such a match. It is not at all like playing relaxed games against your favorite programs at home with your own computer. Playing in front of an audience, with money on the table, against unknown engines with unknown opening books, is quite a different thing. I suspect that most GMs wouldn't even have accepted the challenge.

Tord

Agreed and thanks for the reply....
I do praise the courage of the GM,don't get me wrong,but I am still amazed by the result....
Ok,let's put it this way:
The GM in question against Glaurung running on your fast laptop,4 games at longer time controls to be played in two days with a big break between them,now what would be the outcome
My guess:3-1 for Glaurung as the best prediction I can come up with....