Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Nunn Test Positions

Author: Omid David Tabibi

Date: 01:22:49 05/11/03

Go up one level in this thread


On May 11, 2003 at 03:56:15, Dave Gomboc wrote:

>I seem to recall there being 20 or 30 balanced positions selected by Dr. Nunn
>(one batch of 10, then later another batch of either 10 or 20) at some point in
>the past.  Can anyone tell me a) where I can download them,

Take a look at:
http://www.kozachenko.com/dotbishop/#suite


>and b) what I should
>cite as the source (e.g. were they published in a magazine?)
>

They haven't been published in any journal, so just mentioning them in the
appendix will suffice (preferrably together with a link to your website
containing the downloadable test suite).

P.S. Nunn positions are usually used as starting points for self-play matches.
If you would like to test middlegame performance of an algorithm (by fixed-depth
or fixed-time experiments), you'd better use other test suites (e.g., WCS, WAC,
ECM, etc). Just beware that ECM has large number of errors in the suggested best
moves.

If you are looking just for a few "balanced" positions without any tactical
motives, take a look at the 20 positions Aske Plaat used in his Ph.D. thesis
(see the appendix).


>Thanks,
>Dave



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.