Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: bm in Test suites in general

Author: Renze Steenhuisen

Date: 03:42:48 08/24/03

Go up one level in this thread


>The convention used for testing a chess playing program is that for every EPD
>record processed, the program gets one point if its predicted move (pm) matches
>any of the best moves (bm) while the program loses one point if its predicted
>move (pm) matches any of the avoid moves (am) and gets zero points for any other
>response.
>
>Since the vast majority of chess positions are unsolved, test suites may have
>theoretically incorrect values for "bm" and "am".  When it becomes obvious that
>a particular EPD record in a test suite is flawed in this way, that record in
>edited (correctly, one hopes) and the so the suite is revised.

I had a look at my collected "test suites":
   "Bratko/Kopec"
   "Bednorz/Schumacher - v.2830"
   "Bednorz/Toenez - v.2450"
   "Bednorz/Toenez - v.2630"
   "1001 Brilliant Ways to Checkmate"
   "Encyclopedia of Chess Middlegames"
   "LCT-II v1.21"
   "MATS: MAastricht Test Suite"
   "The 11 Nolot positions"
   "Nunn positions"
   "World's Greatest Chess Games"
   "Win at Chess"
   "1001 Winning Chess Sacrifices and Combinations"
and sad but true, but I only have 2 "am" values set!

>What may be needed are persons with appropriate skills and resources to take
>upon themselves the responsibilites to be long term "Keepers of the Test >Suites" and provide periodic authoritative review and revisions.

Well, I don't mind keeping them on my ftp-site, but I don't think it is a good
idea that I also maintain them. I just put my collected test suites online.

Renze

PS:
Any other ideas of test suites to put online?



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.