Some interesting positions old and new

[Vinvin]

Nr.10)
8/3k1p2/8/2n3p1/p1P1p1Pp/2N1P2P/3K1P2/8 b - - 0 1
Analysis by Stockfish 091016 64 Ultim:

1. -+ (-12.14): 1...f5 2.gxf5 g4 3.Ke1 gxh3 4.Kf1 a3 5.Kg1 Nd3 6.Na2 Kd6 7.f6 Ke6 8.Kh2 Nxf2 9.c5 Kxf6 ...
2. -+ (-11.91): 1...a3 2.Na2 f5 3.gxf5 g4 4.Ke1 Nd3+ 5.Kf1 gxh3 6.f6 Ke6 7.Nc3 Nc5 8.Kg1 Kxf6 9.Kh2 Kf5 ...

So 1...f5 is not the only winning move
1...a3 is equal

I agree, it seems to be a transposition !
1) 1.f5 and a3 later or 2) 1.a3 and f5 later.

[Vinvin]

They find the move, but they still don't recognize this to be a draw. At -10.28, some GUIs have them resign.

Yes, there's 2 levels in solving a position :
1) Good move and good score
2) Good move but bad score

Sometime (2) is because the engine see the other moves are even worse. It solve it by elimination. But sometimes the engine finds the good move by luck because a lot of moves are 0.00 as the key move.

[Paloma]

When I dug in archives, I found some interesting positions (a lot from George Tsavdaris).
There are recent positions too. Key moves are not verified for all.
....
3)
Best move and move to avoid is not clear but this position gives a hard time to engines :
[d]2kr1nr1/pbp1qp2/1p2pn1p/8/3Pp2Q/P1P3N1/2PBBPPP/R3K2R w KQ - 0 1 am O-O; bm Bxh6;
Fischer-Kovacevic : http://www.chessgames.com/perl/chessgame?gid=1044313

I suspect also that the following is wrong and it needs to be am Bxh6 bm 0-0

....

It seems only 1.Nh5 is better than 1.0-0 or 1.Bxh6 here
1093: Some interesting positions - old and new
2kr1nr1/pbp1qp2/1p2pn1p/8/3Pp2Q/P1P3N1/2PBBPPP/R3K2R w KQ - 0 1

Analysis by Komodo 10.4 64-bit 4cores: (depth 40)

1. = (0.00): 1.Nh5 Rxg2 2.Qxf6 Qxf6 3.Nxf6 Rg6 4.Ng4 e3 5.Rg1 exd2+ 6.Kxd2 Nd7 7.Bd3 ...
2. =/+ (-0.51): 1.Qxh6 Ng4 2.Qh3 f5 3.Nxe4 Bxe4 4.f3 Nf6 5.fxe4 Nxe4 6.Bf4 Ng6 7.0-0 ...
3. -/+ (-0.77): 1.c4 Ng6 2.Qxh6 e3 3.Qxe3 Nh4 4.0-0-0 Ng4 5.Bxg4 Rxg4 6.Ne2 Qf6 7.h3 ...

[Vinvin]

When I dug in archives, I found some interesting positions (a lot from George Tsavdaris).
There are recent positions too. Key moves are not verified for all.
....
3)
Best move and move to avoid is not clear but this position gives a hard time to engines :
[d]2kr1nr1/pbp1qp2/1p2pn1p/8/3Pp2Q/P1P3N1/2PBBPPP/R3K2R w KQ - 0 1 am O-O; bm Bxh6;
Fischer-Kovacevic : http://www.chessgames.com/perl/chessgame?gid=1044313

I suspect also that the following is wrong and it needs to be am Bxh6 bm 0-0

....

It seems only 1.Nh5 is better than 1.0-0 or 1.Bxh6 here
1093: Some interesting positions - old and new
2kr1nr1/pbp1qp2/1p2pn1p/8/3Pp2Q/P1P3N1/2PBBPPP/R3K2R w KQ - 0 1

Analysis by Komodo 10.4 64-bit 4cores: (depth 40)

1. = (0.00): 1.Nh5 Rxg2 2.Qxf6 Qxf6 3.Nxf6 Rg6 4.Ng4 e3 5.Rg1 exd2+ 6.Kxd2 Nd7 7.Bd3 ...
2. =/+ (-0.51): 1.Qxh6 Ng4 2.Qh3 f5 3.Nxe4 Bxe4 4.f3 Nf6 5.fxe4 Nxe4 6.Bf4 Ng6 7.0-0 ...
3. -/+ (-0.77): 1.c4 Ng6 2.Qxh6 e3 3.Qxe3 Nh4 4.0-0-0 Ng4 5.Bxg4 Rxg4 6.Ne2 Qf6 7.h3 ...

Yes, same conclusion as here : http://www.talkchess.com/forum/viewtopi ... 055#692055