Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Test Positions

Author: John Merlino

Date: 12:13:22 02/09/03

Go up one level in this thread


On February 09, 2003 at 09:33:03, David Rasmussen wrote:

>What would your engine play here, and if not the suggested move, then why not?
>
>[D]2r5/p4pk1/3p2p1/Pp2p3/1P1nP1P1/q2P3P/2r3B1/1R2QR1K b - - 0 33
>
>33... Rxg2

CM9000 agrees with Rxg2.

>[D]8/p4pk1/3p2p1/Pp2p3/1P1n2P1/6KP/4q3/2Q2R2 b - - 0 38
>
>38... Qc4

You already know that CM9000 prefers Qd3+.
>
>[D]8/p7/5k2/Pn1ppp2/1Pp4P/5K2/8/6R1 b - - 0 47
>
>47... Nd4+

CM9000 prefers e4+ here, but the eval difference between it and Nd4+ is pretty
small (less than 0.1). I think one of the problems with Nd4+ is that, while it
might eventually allow the Knight to protect a Black queening square, it
prevents the Knight from doing a potentially more important job, which is
defending a WHITE queening square by going to (for example) f7. Example:

47...e4+ 48.Kf2 c3 49.h5 d4 50.h6 Nd6 51.h7 Nf7

But this position is lost for Black.

After forcing 47...Nd4+ 48.Kf2, The King immediately wants to retreat the Knight
back to e6, and the eval is dropping with every iteration. The King sees both
sides queening, but White will also have the rook to Black's Knight.

So, it appears that maybe EITHER move is lost. White's two outside passers are
stronger than Black's four connected passers!

Once again, this analysis was done by a patzer on a rather slow machine, so I'm
sure somebody else could be more precise....

jm



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.