Page 5 of 10
Re: Nakamura vs. Komodo
Posted: Thu Jan 07, 2016 10:01 am
by whereagles
Naka also didn't run out of time
Re: Nakamura vs. Komodo
Posted: Thu Jan 07, 2016 12:26 pm
by duncan
Laskos wrote:I remember when simulating these handicaps that the rule of thumb would be that for human against computer "material odds beat positional odds" (for the same nominal value of odds).
was the human a super grand-master?
Re: Nakamura vs. Komodo
Posted: Thu Jan 07, 2016 3:43 pm
by lkaufman
Lenderman played the opening well.
Re: Nakamura vs. Komodo
Posted: Thu Jan 07, 2016 4:42 pm
by velmarin
A nice article on the event, in my humble opinion Nakamura is the star.
http://worldchess.com/2016/01/02/are-ma ... ass%C3%A9/
A very profound view of Matthew Lai:
Code: Select all
“In my opinion, it’s much more impressive that a human can play at a 2900 level while only searching at three to five moves per second, than that a computer can play at a 3300 level while searching tens of millions of moves per second,” Lai said. “Playing at only 3300 level is pretty disappointing when you are looking at tens of millions of moves per second.”
Re: Nakamura vs. Komodo
Posted: Thu Jan 07, 2016 4:43 pm
by Laskos
duncan wrote:Laskos wrote:I remember when simulating these handicaps that the rule of thumb would be that for human against computer "material odds beat positional odds" (for the same nominal value of odds).
was the human a super grand-master?
IIRC I simulated a weaker player, FM or IM level, in need of 6-7 mover odds. Yes, I am not sure my rule of thumb holds for a 2800 player, since the scaling with time control (strength) was an issue there.
Re: Nakamura vs. Komodo
Posted: Thu Jan 07, 2016 4:47 pm
by lkaufman
velmarin wrote:A nice article on the event, in my humble opinion Nakamura is the star.
http://worldchess.com/2016/01/02/are-ma ... ass%C3%A9/
A very profound view of Matthew Lai:
Code: Select all
“In my opinion, it’s much more impressive that a human can play at a 2900 level while only searching at three to five moves per second, than that a computer can play at a 3300 level while searching tens of millions of moves per second,” Lai said. “Playing at only 3300 level is pretty disappointing when you are looking at tens of millions of moves per second.”
It does suggest there is ample room for improvement.
Round 3 fight now
Posted: Thu Jan 07, 2016 5:04 pm
by sainzlei
Re: Round 3 fight now
Posted: Thu Jan 07, 2016 6:35 pm
by duncan
Re: Round 3 fight now
Posted: Thu Jan 07, 2016 6:36 pm
by mjlef
Game 3:
[pgn][Event "Computer chess game"]
[Site "8CORE"]
[Date "2016.01.07"]
[Round "?"]
[White "Komodo"]
[Black "GM Nakamura"]
[Result "1/2-1/2"]
[BlackElo "2400"]
[Time "11:02:50"]
[WhiteElo "2400"]
[TimeControl "2700+15"]
[SetUp "1"]
[FEN "1rbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/1NBQKBNR w Kk - 0 1"]
[Termination "normal"]
[PlyCount "97"]
[WhiteType "human"]
[BlackType "human"]
1. c4 Nf6 2. Nc3 e5 3. Nf3 e4 4. Nd4 Bc5 5. Nb3 Bb4 6. Qc2 Qe7 7. e3 O-O 8.
Nd4 d6 9. Nde2 c6 10. a3 Ba5 11. b4 Bc7 12. Bb2 d5 13. cxd5 cxd5 14. f4 a5
15. Nb5 Bb6 16. Be5 Bd7 17. Bd6 Qd8 18. bxa5 Bxa5 19. Bxb8 Qxb8 20. Ned4
Rc8 21. Qb2 Ng4 22. Ke2 h5 23. h3 Nh6 24. Rg1 Bb6 25. Ke1 Bxb5 26. Bxb5 Qd6
27. Ne2 d4 28. Nxd4 Qf6 29. Qb4 Rc1+ 30. Kf2 {(Ke1-f2 Nh6-g4+ h3xg4 Qf6-h4+
g2-g3 Qh4-h2+ Rg1-g2 Qh2-h1 Nd4-c2 Bb6xe3+ Kf2xe3 Qh1xg2 Qb4-b2 Rc1-d1
Qb2-b4 h5xg4 Bb5-e2 Qg2xg3+ Ke3xe4 Qg3-g2+ Ke4-d3 Qg2-d5+ Qb4-d4 Qd5-b3+
Qd4-c3 Qb3-d5+ Kd3-e3 Rd1-g1 Qc3-c8+ Kg8-h7 Be2-d3+ g7-g6 Bd3-e4 Rg1-g3+
Ke3-f2 Rg3-f3+ Be4xf3 Qd5xf3+ Kf2-g1 Qf3-g3+ Kg1-f1 Qg3-h3+ Kf1-e1 Qh3-h1+
Ke1-f2 Qh1-f3+) 0.00/37 9} Ng4+ 31. hxg4 {(h3xg4 Qf6-h4+ g2-g3 Qh4-h2+
Rg1-g2 Qh2-h1 Nd4-c2 Rc1xc2 Qb4xe4 Rc2xd2+ Bb5-e2 h5-h4 Qe4-e8+ Kg8-h7
Qe8-e4+ Kh7-h8 Qe4-e8+ Kh8-h7) 0.00/39 21} Qh4+ 32. g3 {(g2-g3 Qh4-h2+
Rg1-g2 Qh2-h1 Nd4-c2 Bb6xe3+ Kf2xe3 Qh1xg2 Qb4-b2 Rc1-d1 Qb2-b4 h5xg4
Bb5-e2 Rd1-c1 Qb4-b2 Rc1-h1 Qb2-b5 Rh1-h2 Nc2-d4 Qg2xg3+ Ke3xe4 Qg3xa3
Qb5-d7 Rh2xe2+ Nd4xe2 Qa3-f3+ Ke4-f5 Qf3xe2 Qd7-d8+ Kg8-h7 Qd8-h4+ Kh7-g8
Qh4-d8+) 0.00/39 25} Qh2+ 33. Rg2 Qh1 34. Nc2 Rxc2 35. Qxe4 Rxd2+ 36. Be2
hxg4 37. Qe8+ Kh7 38. Qb5 g6 39. Qxb6 Rxe2+ 40. Kxe2 Qxg2+ 41. Kd3 Qxg3 42.
Qxb7 Kg7 43. Qb2+ Kg8 44. Qb8+ {(Qb2-e5 Qg3-f3 Qe5-e8+ Kg8-g7 Qe8-e5+
Kg7-h7 Qe5-e7 Qf3-f1+ Kd3-d4 Qf1-a1+ Kd4-d3 Qa1-b1+ Kd3-c3 Qb1-a2 Kc3-d3
Qa2-b1+) 0.00/25 9} Kg7 45. Qe5+ Kg8 46. Qe8+ Kg7 47. Qe5+ Kg8 48. Qe8+ Kg7
49. Qe5+ {3-fold repetition} 1/2-1/2
[/pgn]
note the analysis is not from the 24 core that played, but my 8 core machine.
Re: Round 3 fight now
Posted: Fri Jan 08, 2016 12:24 am
by Leto
Komodo won the final game, all other games were drawn.