Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Revised ECM Test Suite (some additional analysis needed)

Author: Omid David Tabibi

Date: 09:10:49 07/19/04

Go up one level in this thread


>>2bq1b2/1p3k2/1r1p1p1p/pNpP1PpP/P1P3P1/6B1/1P5Q/2K1R3 w - - bm Re6; id "ECM.396";
>
>this looks like a very correct exchange sac which takes a long time to see, but
>after Re6 Bxe6 fxe6+ i see no proper defense for black for the various white
>threats (invasion of the Q to g6, grabbing a pawn (a5 or d6 or f6)). what kind
>of defense do you suggest after Re6?

After fxe6+ simply go Kg8. White has several threats, but black manages to
defend against them. Any winning line for white after Kg8?


>
>>rq1r2k1/3nbp1p/p2p2p1/1p1Pp1P1/3B1P2/3B3Q/PPP4P/2KR2R1 w - - bm f5; id
>>"ECM.497";
>
>f5 looks good after 1.f5 exd4 2.fxg6 hxg6 3. Bxg6 fxg6 4.Qe6+ Kg7 5. Qxe7+ Kg8
>6. Qe6+ Kg7 but is there any way to continue?

Yes, 1.f5 seems to be the best move, although I can't see a clear win.


>
>>r3k1r1/pb1qpp1p/1pnp1n1B/2p2R2/3PP3/2P3PN/P1P3BP/R2Q3K b q - bm Nxe4; id
>>"ECM.561";
>i can't see anything clear here, Nxe4 doesn't seem to work

agree.

>
>>8/p5k1/2p2p2/4p1p1/1p2P3/1P4PP/1PP5/5K2 b - - bm g4; id "ECM.601";
>g4 seems to draw (1...g4 2.h4 a5 and wait - the white king cannot attack the
>black pawns because of ...f5! at the right moment; on c3 black plays bxc3 and
>continues to wait). but i see nothing for white after 1...Kf7 2.g4 (what else,
>since black would transpose back to the 1...g4 line if necessary).

agree.

>
>>8/8/4b1p1/2Bp3p/5P1P/1pK1Pk2/8/8 b - - bm g5; id "ECM.605";
>this looks like a famous position to me but i don't know where i've seen it
>before. the solution would be 1...g5! 2.fxg5 d4+! 3.Bxb4 Kg3, and black wins
>because he will take the h4 pawn and win with the two far-separated passers,
>while white's 2 pawns are useless because they are stopped by the black bishop
>which is simultaneously defending the b3 pawn. it's a "principle of one
>diagonal" vs "the pants" situation.

agree.

>
>
>>r1b2rk1/2q1bppp/p1n1p3/1p1np1P1/5P2/1NNBBQ2/PPP4P/R4RK1 w - - bm Bxh7+; id
>>"ECM.665";
>1. Bxh7+ Kxh7 2. Qh5+ Kg8 3. Rf3 looks like what this should be, 3...g6 4. Qh6
>Bd6 (what else?) 5. Rh3 f6 6. Nxd5 exd5 7. Qh8+ Kf7 8.Qxf6+ Ke8 9. Qxg6+ looks
>winning, any improvements for black?

agree.

>
>>2rqr1k1/1p1bbppp/p3p3/n7/3P2Q1/2PB1N2/P4PPP/R1B1R1K1 w - - bm Bxh7+; id
>>"ECM.673";
>1.Bxh7+ Kxh7 2. Qh5+ Kg8 3. Ne5 g6

3... Rf8 instead, and even though white has a clear advantage, I don't see any
win.

So why not play something like 1.Ne5 instead of Bxh7+ ?



>4. Nxg6 Bf6 (what else?) 5.Ne5 is winning for
>white, any improvements for black?


>
>>r1b2r1k/ppppq1pp/2n1n3/6N1/2B2P2/4B3/PPP3PP/R2Q1RK1 w - - bm Nxh7; id "ECM.674";
>1.Nxh7 Kxh7 2. Qh5+ Kg8 3. f5 Ne5 4. fxe6 Nxc4 5. Rxf8+ Qxf8 6. Rf1 wins for
>white, improvements for black?

agree.


>
>
>>r1b2rk1/1pq2ppp/p1p1p3/2n1P3/2N2P2/3B4/PPP3PP/R2Q1RK1 w - - bm Bxh7+; id
>>"ECM.677";
>1.Bxh7+ Kxh7 2. Qh5+ Kg8 3. Rf3 Ne4 4. Rh3 f5 seems to hold the draw for black,
>as white's attacking ideas look dangerous but don't seem to work, e.g. 5. Rd1 b5
>6.Nd6 Nxd6 7. Rxd6 Qb6+ 8. Kf1 Ra7 9.Rh6 Ke8 - of course white can always take a
>perpetual, but i don't see a win.


>
>>r5k1/pn1q1rpp/2pp4/5R1N/bP6/4BQ2/P4PPP/2R3K1 w - - bm Nxg7; id "ECM.714";
>1. Nxg7 Rxg7 2. Bh6 looks great for white, any defense for black??

2... Qe7
White has an advantage in that line, but I don't see any immediate win. There
are other first moves that give white advantage.


>
>>4kbr1/pp1bpp1p/3p1p2/3N4/1PBQP3/q7/P1r2PPP/R4RK1 b - - bm Rxg2+; id "ECM.725";
>1...Rxg2+ 2.Kxg2 Qh3+ 3. Kg1 Qg4+ 4. Kh1 Qf3+ 5.Kg1 Bh3 6. Ne3 Bh6 looks like
>the idea of this puzzle, but does 7. Qd1 hang on now?

7.Qd1 and I don't see how black will proceed.

1...Qh3 looks much better than 1...Rxg2+.


>
>>r3r1k1/pbq1ppbp/1pp2np1/4N3/3P4/2P5/PPB1QPPP/R1B1R1K1 w - - bm Nxf7; id
>>"ECM.762";
>1. Nxf7 Kxf7 2. Qe6+ Kf8 3. Bb3 Nd5 4. Re3 must be the idea here, continue
>4...Bf6 5. Rh3 Kg7 6. Bh6+ Kh8 7. Bxd5 exd5 8. Qf7 +- - improvements for black?
>

right, very deep line!


>>2rqr1k1/pb2bp1p/1pn1pnpB/4N3/3P4/P1N3R1/1P3PPP/RB1Q2K1 w - - bm Rxg6+; id
>>"ECM.810";
>is this a typo and should be Nxg6? that is winning, Rxg6 looks very bad...

That is what I thought...

>
>>1q1r3k/3P1pp1/ppBR1n1p/4Q2P/P4P2/8/5PK1/8 w - - bm Rxf6; id "ECM.825";
>this is another famous one:
>1. Rxf6 Qxe5 2.fxe5 gxf6 3. exf6 easily winning. do you see an alternative
>solution? in any case, for a human this is simplest, because there is no chance
>of error after 1.Rxf6, it's completely over.

agree.


>
>>rr4k1/5p1p/n2p1npQ/q1pP4/6P1/2N2P1N/PP1R3P/2K4R b - - bm Rxb2; id "ECM.853";
>1...Rxb2 2.Kxb2 Rb8+ 3. Kc2 Nb4+ 4. Kd1 Qa3 5. Ke2 Qxc3 looks winning for black
>- objections?

agree.

>
>>6k1/1R4pp/2p5/8/P1rNp3/6Pb/4PK2/8 w - - bm a5; id "ECM.878";
>1. a5 draws easily, as in 1.a5 Rxd4 2.a6 Ra4 3. a7 Bc8 4. Rc7 Be6 5. Rb7 =,
>other first moves may still have to fight for the draw (1. Ke3, but i think that
>should also be good enough for a draw).

I thought there is a win for white I am missing...





This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.