On Feb 26, 2:41 pm, Martin Brown
wrote:

In below game Help Bot Sacrificed its Rook for a Knight.

11. Rf1-e1{36} Rf8-f3{20}
12. g2-f3{28} Nc6-d4{10}

The RxN g2xN sequence is basically OK. Black can do better than the move
played with something like 12. ... d5 12. cxd5 Nd4 14. Qd1 Qxd5
and I think equality.

Hey, there can be no "equality" he one of us
two patzers is going down!



BTW any chance you could modify Getclub so that it shows captures with
the conventional "x" notation instead of "-" it makes life easier for
humans!

Humans? Who cares about /them/?


24. Ne4-g5{26} Qh5-g5{60}

24. ... QxN was a bit optimistic and premature. BxN piles on the
pressure.

I had several chances in this game to press the
attack-- all of which I fumbled. The exchange sac'
is perfectly playable... possibly even winning.

In particular, where I played 14. ...Qe8, it would have
been much better to play ...Qd7, and the threat of
....Bg4 poses serious trouble for White.

A bit later on, I had multiple opportunities to
regain the exchange-- all of which I missed
(I was still searching for my fumbled "attack").


31. Rg5-e5{30} Re8-e5{12}

This is among the worst captures. Either of Bxe5 or bxc5 would hold onto
white advantage. And about half of all moves are better than Rxe5.

I simply overlooked an in-between move:
31. ... cxb4, 32. Rxe8 Qg6+, where I would
win a pawn (although trading Queens was
not on my to-do list).


44. Qe3-e5{24} Qd5-e5{120}

QxB or Rb1 both look drawish (without looking too hard).

49. Bb2-e5{36} Be7-g5{70}

Getclub might still be able to hold the game with the surprising
49. Rxa3 BxR 50. BxB KxP opposite coloured bishops. It would probably
lose the endgame against a human but it is close because of where the
pawns would have to queen. Any endgame experts shed some light?


That would have resulted in a win for me, since
the stupid program will not move it's King out of
a corner in the endgame. However, theoretically
speaking it is very likely a draw since White's
King can reach the Queen's side first.


51. Ra1-c1{20} Bg5-c1{12}

Teasing the program with Bc1 ? Tut Tut.


Who was "Tut"? Never heard of him. Probably
a nobody, a lowly peon.


This is truly amazing. I cannot
see why a materialistic chess engine did not snatch your arm off here.
52. RxB and white is back in the game!


Sanny's notation gave Qc2-c1, as if the pawn
is promoted first, and only THEN advances to
the promotion square. In fact, I advanced the
pawn, then answered a questionaire saying I
preferred Queens. The program then captured
the Queen with its Rook-- on the square c1,
which I in turn captured with my Bishop.

That's my story and I'm sticking to it.



What were the wrong moves played by Easy Level that it lost in 57
Moves?

See above. It was offered a chance to win as late as the 52nd move.

Not really.

The program did act as though it's QR were
glued to the square a1 for much of the early
part of the game. I wonder if it "thought" by
doing so, it maintained the option of castling
Queenside if the King ever made it back to
e1?

If I got this same opening position again, I
would play the very same sacrifice; but this
time, I would follow up a bit more thoughtfully.


-- help bot