Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: ECM errata (201 thru 300)

Author: Howard Exner

Date: 11:20:07 01/23/98

Go up one level in this thread


On January 22, 1998 at 18:51:06, Amir Ban wrote:

>
>Here's some more work for everybody:
>
>No. 206 (Nxc7):
>
>Looks incorrect: Nxc7 Qxc7 Qd5+ Kh8 Bxd6 and now white has a nasty
>surpise with Nb6.
>
>Recommend: Throw it out.

Try Re8, instead of Bxd6? ie Nf6 2.Rxf8 Bxf8 3.Bb2 Bg7 4.Bc4 ...
>
>
>No. 208 (Ndxf4):
>
>This just wins a pawn, but so does Nhxf4.
>
>Recommend: Side solution Nhxf4.

Either move looks too easy also.

>No. 215 (Rd7):
>
>This doesn't work: Rd7 Nxd7 Rxd7 and now I suppose Qc8 is expected and
>crushed by Rxe7+, but instead black plays Rag8!

>Recommend: Throw it out.

Agree

>No. 227 (Bxg7+):
>
>I find Qd7, and it seems as good as Bxg7+. Qe7 seems a bit inferior, but
>that may be an illusion.

>Recommend: Side solution Qd7.

This one seems too easy if both moves are allowed.
Bxg7 looks much better than Qd7. It results in the QN vs RR in
a position where that QN combo is awesome.

>No. 228 (Bg1+):
>
>Bg1+ indeed draws, but so does Bxg3+ even easier.
>
>Recommend: Side solution Bxg3+

Either way this looks too easy for most programs.
>
>
>No. 231 (Rxd6):
>
>The point of the key is not obvious to me. I find Qh4+ which wins
>material.
>
>Recommend: Change key to Qh4+. Check this out.

1.Rxd6! Qxd6 2.Rd1 Qb8 3.Rxd7 1-0. This looks quicker than Qh4.

>
>
>No. 237 (Bf2+):
>
>Junior's recommendation for black here is to resign. Bf2+ seems like a
>particularly bad move. Anyone can say what the point is here ?
>
>Recommend: Throw it out.

Rebel plays Rd1+ immediately with a big plus. This should be the key
move.
Involves a Knight under promotion.

>
>
>No. 247 (Bh7+):
>
>This is a difficult one. I have no doubt the key is correct, but I can't
>see better than a draw.
>
>Recommend: No change.

 Agree as Bh7+ forces a nice win.
>
>
>No. 271 (Nxc6):
>
>Qd8 is the right move here (score = 1235). White threatens mate in 2 and
>picks up a queen and a few other pieces.
>
>Recommend: Change key to Qd8.

Yes but now it is probably too easy.
>
>
>No. 274 (Qxf3+):
>
>One of those where the cook is prettier than the original. The idea is
>the discovered check d4, but Qxf3+ is sloppy execution. Nd2! is right.
>
>Recommend: Change key to Nd2.

Agree and also the simple Ng5 should transpose as white's reply, Nd4
seems forced in both lines. Again, it is too easy.
>
>
>No. 277 (Rh8+):
>
>I find Qg6 for a +1 score. Rh8+ may well win but I don't see how.
>
>Recommend: Investigation necessary.

This is a keeper as white will eventually force Qxe6+ Kf8, then
Qf5+!
>
>
>No. 280 (Nxg7):
>
>Not an easy one. Nxg7 Kxg7 Bh6+ and white wins.
>
>Recommend: No change.

Agree
>
>
>No. 286 (Qh6+):
>
>I find mate in 11.
>
>Recommend: No change.

Ditto for Rebel 8 in about 13 seconds
>
>
>No. 287 (Qb5+):
>
>Looking at the position, the assignment "white wins" seems impossible. I
>only see that Qb5+ draws, and so does Rc8+.
>
>Recommend: Investigation necessary.

Both win after lots of forced checks and a Knight under promotion.
The lines seemed forced for white and I think programs will stumble
onto the win either initially or by accident rather quickly.
>
>
>No. 291 (Rxb2):
>
>This one may be correct, but I can't find a win.
>
>Recommend: Investigation necessary.

1. .. Rxb2 2.Kxb2 Ba3+ 3.Kxa3 Rb8! is the winning line given in the
book.
Looks like it works.
>
>No. 293 (Qh2+):
>
>There are several ways to score +5 or more, like Rxf1+ and Qxh2+, but I
>find Qh3 for +7.
>
>Recommend: Add Qh3 and Rxf1 as side solutions.

Or it could be classified as too easy.

>
>
>No. 297 (b4):
>
>I don't solve this one, but I think it's correct. b4 Bxb4 Ne4 and white
>gets an advantage which may be only positional. This needs verification.
>
>Recommend: No change.

Agree
>
>
>No. 298 (Re6):
>
>I find Rf6 first, and then settle on h4. I score Re6 lower, but it
>forces a KPKP endgame with a pawn up, so may be the simplest win.
>
>Recommend: Add Rf6 and h4 as side solutions.

Re6 is probably best. GM's would handle this resulting position
easily:

5k2/5q1p/p1p2Qp1/1pP5/4P3/8/PP4PP/6K1 w - - id=Pos 298 - ECM.EPD;
bm=f6f7;

The exchange to the endgame is best followed by b4! giving a tempo with
an eventual a2-a3.
>
>
>No. 300 (Kxd5):
>
>I find Bh2 Ke6 f5+ and this looks like a comfortable win. I don't see a
>win after Kxd5 and it looks dubious.
>
>Recommend: Since Bh2 is trivial, throw it out.

Yes it is easy but Kxd5 I think wins more quickly. The resulting N&pawns
vs N&pawns ending is elementary. The other way still poses the problem
of the two passed a pawns with the wrong colored bishop. It would still
be an elementary win for strong humans. Computers would have to be
careful
not to lose the kingside pawns for a piece.



This page took 0.01 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.