Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Can anyone or engine solve this problem?

Author: Dieter Buerssner

Date: 16:02:03 03/24/04

Go up one level in this thread


On March 23, 2004 at 18:25:54, Anson T J wrote:

>A game begins with 1.e4 and ends in the fifth move with knight takes rook mate.
>Reconstruct the game.
>
>[d]rnbqkbnr/pppppppp/8/8/4P3/8/PPPP1PPP/RNBQKBNR b KQkq e3 0 1

The problem has exactly 2 solutions (one is a simple transposition of the
other). I already indicated the second solution (besides the one, that was
mentioned in other followups) in another post. I also indicated, how I came to
the conclusion (with brute force search). I was not 100% sure, whether another
first move of black than Nf6 could yield in another transposition (I was pretty
sure, from thinking myself about it). So, additionally to the experiments I
mentioned in my other followup, I also did a perft(9), but disallowing Nf6 this
time (used a hash table for it). No other solutions were found. Because I
already did a "perft(8)" after Nf6 without hash tables (with an algorithm, that
will find all transpositions), there cannot be another solution.

>Major kudos if you solve it yourself!

Not really myself. Brain time: < 1/2 hour (rather easy when you intend to use a
fast tool for the brute force part), CPU time: ~3 hours to come to the
conclusion, perhaps another 2 hours wasted CPU time. Coding time: 20 minutes
(really small changes to already existing perft code).

If anybody finds another solution, I will give a crate of beer (or a nice bottle
of wine, or something similar) to the first one. It would mean a horrible bug
(in my thinking and/or coding).


Regards,
Dieter



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.