Really interesting article about his match, playing with the assistance of Rybka, against Stockfish. Lots of good analysis. He's such a good author.

http://www.chess.com/article/view/ho...r-in-the-world

from the LC thread, for the lols:

Hahaha wow. Not sure how to even address that.

Nakamura+Rybka playing Stockfish now http://www.chess.com/tv

Draw in the first game, Nakamura seems to have an easy draw in a locked KID position, maybe there's a chance the computer will self destruct.

The commentary is pretty awful unfortunately, I think these things are always better with a second commentator.

Daniel Rensch is hilarious. The first time I listened to his commentary I found it incredibly annoying, but he has grown on me.

He doesn't seem to get anything. Nakamura offers a piece trade (dark squared bishops) coming up to move 50, he is all "why would he do that?" He doesn't seem to have seen the famous Nakamura-Rybka game.

Now he's spouting on something about how a player in the audience can step in and adjudicate a draw by the 50 move rule. How can you be an IM and not know some of the basic rules?

Weren't we talking about this match in another thread? Can't find it now. Apparently the second half is Nakamura + pawn+White instead of Nakamura + Rybka.

I do think it's a pretty difficult job to be entertaining and instructive commentating this on your own, that's why it would be better with a co-commentator who can provide some serious insight and laugh at Rensch's jokes.

now computer marches the king to the queenside. This is pretty interesting because this only makes sense as a plan to open the kingside. I didn't expect the computer to find this plan (I don't know if it's objectively any good, but seems the right way to play for a win) - but bizarre that it plays it now and not earlier with the dark squared bishops on the board and the h-pawn back on h3.

I think it's likely Nakamura has prepared for this game. Maybe even before he was asked to do it, he's certainly played training games against the computer, probably more than any other top player ever. There's a chance he's had this position against Stockfish before.

hmm opening the position with f5 doesn't seem to have led to something good for Black, now I think the computer wins.

lol if this is a fortress it's quite incredible, I assume White get zugzwanged out though

Rybka is actually a huge help for checking if something is a fortress, even though it may not find it for itself. if you give it long enough you can see if it just keeps showing +5 or jumps to +10, +15 etc

Gotta be a win here by punting 2 pawns with b4 a/c5 and getting the king in, if nothing simpler. Not sure white can't just zug him off with the threat of taking on g3 and crap like Qf7+ Kh6 Qg8 threatening mate.

I do like that Rensch is not using a computer for the analysis

He just said Nakamura is getting piece odds in the last two games? Hard to believe that is true, I think I would smash Stockfish in a long game with piece odds. 75% chance this dude doesn't know a pawn from a piece.

Didn't see the finish, did anything happen or Naka just resign?

he played on a few moves in Q3 v R3 and then resigned, Stockfish didn't win a pawn but it's clear White has about 10 winning plans

Third/Fourth game is pawn and half-move (Nakamura gets White in all games) after all. I think Stockfish is a small favourite here. It started with a Nd5 Sveshnikov where I'm not sure who would be better normally, but Black doesn't really need his h-pawn. I feel like a Bg5 Sveshnikov would have been a better choice, since the main lines should lead to Black getting mated without that pawn.

lol me, queens came off and White should have the only winning chances in the ending

some 2200 streaming it here, he somehow seems more knowledgeable and more entertaining than the chess.com broadcast

http://www.twitch.tv/zugaddict

That guy is great.

"What we have here is Maurice Ashley–style chess commentary. There is a pawn on e4 and it is being menaced."

inspired by this I tried playing Stockfish with piece odds (Queens Knight, 10 15 time). I smashed it easily by just trading pieces and I think any 1500 could do the same at least most of the time. (I did think I had blundered in my White game after ...Qf8, but actually I was fine). Two pawns would be a much tougher test.

[FEN "rnbqkbnr/pppppppp/8/8/4P3/8/PPPP1PPP/R1BQKBNR b KQkq - 0 1"]

1. c3 e5 2. e4 d5 3. Nf3 dxe4 4. Nxe5 Qd5 5. f4 exf3 6. Nxf3 Nf6 7. d4 Bd6
8. Bd3 Bg4 9. O-O O-O 10. h3 Bxf3 11. Qxf3 Qxf3 12. Rxf3 Nbd7 13. Bg5 h6
14. Bxf6 Nxf6 15. Re1 Rfe8 16. Rfe3 Rxe3 17. Rxe3 Re8 18. Rxe8+ Nxe8 19.
Kf2 Nf6 20. Bf5 Kf8 21. Bc8 b6 22. Bb7 Ke7 23. b4 Bf4 24. Ke2 Kd6 25. b5
Nd5 26. Kd3 Bg5 27. Kc2 f5 28. Bc6 Bf6 29. Kd3 g5 30. Bb7 h5 31. a4 g4 32.
hxg4 fxg4 33. c4 Ne7 34. Ke4 Bg5 35. g3 Bd2 36. c5+ bxc5 37. dxc5+ Kxc5
38. Ke5 c6 39. bxc6 Nxc6+ 40. Kf5 Be1 41. Ba8 Bxg3 42. Kg5 Bc7 43. Kh4 Kb6
44. a5+ Nxa5 45. Kxh5 g3 46. Kg4 Nb7 47. Kf3 a5 *


[FEN "r1bqkbnr/pppppppp/8/8/4P3/8/PPPP1PPP/RNBQKBNR b KQkq - 0 1"]

1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 Ne7 5. a3 Bxc3+ 6. bxc3 c5 7. Bd3 c4
8. Be2 O-O 9. Bg5 Bd7 10. h4 h6 11. Bxe7 Qxe7 12. g4 Qd8 13. Nf3 Rb8 14.
Qd2 Ba4 15. g5 h5 16. g6 fxg6 17. Rg1 Qe8 18. Rc1 Qf7 19. Qe3 Be8 20. Kd2
Qe7 21. Ng5 b5 22. Rg3 a5 23. Rcg1 Kh8 24. Rf3 Rxf3 25. Bxf3 Kg8 26. Qf4
Ra8 27. Bg2 Bd7 28. Qg3 b4 29. cxb4 axb4 30. axb4 Qxb4+ 31. Qc3 Qf8 32.
Qf3 Qe7 33. Bh3 Rf8 34. Qg3 Qb4+ 35. c3 Qb6 36. Ke2 Re8 37. Qe3 Ra8 38.
Kf1 Ra7 39. Kg2 Ra8 40. Qd2 Qb3 41. Rc1 Kh8 42. Qc2 Qxc2 43. Rxc2 Bc8 44.
Bxe6 Bxe6 45. Nxe6 Rc8 46. Nf4 Rd8 47. Nxg6+ Kh7 48. Nf4 g5 49. hxg5 Rg8
50. g6+ Kg7 51. Ra2 Kh6 52. Ra6 Rg7 53. Rd6 Re7 54. Nxd5 Rg7 *