Author: Rolf Tueschen
Date: 12:27:52 06/21/04
Go up one level in this thread
On June 21, 2004 at 15:07:35, Cesar Contreras wrote: >Hi, thank you responding. > >>You write that you think that you can make an analogy to IQ tests ... but this >>is also wrong. > >Could you please explain me why i can't make that analogy. > >>Finally let me inform you without being arrogant that there is no lack of good >>test methodology because test theory is a perfect field > >Yes, but what i mean is about tests oriented to chess programming. Not only test >suites, but maybe about tournament decisions (players selection, time selection, >number of games) > >>you miss is a good test suite and the answer is - - - there is no such test >>suite because it's a contradiction in itself. You build up a chess engine for >>that it could play chess. But chess isn't just puzzle solving. > >You are right, it's diferent, but a chess engine plays like this: "given a >centain position, find the best move you can find (that one that gives you the >best score)", witch it's very similar to test solving, the main difference >(maybe fundamental) are the samples. > >What i can undertand it's that the test could needs to be so big that can't be >applicated, but it's scientifically proved or is just experience and opinions? Just a short answer due to time: this is also science if you want and also experience, no, not opinions, but I prefer to say - as a chessplayer - the main reason is CHESS itself. Because of that we need far more positions than just 100! Chrilly Donninger spoke of some 700 and more, it is important because of CHESS again, that we get positions of the same topics but where the solution does NOT function, so to speak the mirrored (opposite). Then you have CHESS in these positions. Otherwise you dont have real chess that you do research - this is also the answer to your IQ analogy...
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.