Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: an example how users - not programmers - use tests

Author: Gian-Carlo Pascutto

Date: 09:06:31 06/19/04

Go up one level in this thread


On June 19, 2004 at 12:02:19, Steve Glanzfeld wrote:

>So he compiles
>a test from typical minor piece ending positions (maybe from books, GM
>analysis...) where there is a specific difficult good move to be found, or a
>good looking but in fact bad move to be avoided (= solve).

[...]

>Now please don't tell me that wouldn't be ok...
>
>So what's all that "flawed" (etc.) blah??

Many/most positions do not have a single good move, but instead the
engine/player must play consistently. A testsuite does not measure the
ability to do so.

Additionally, the number of "avoid move" problems is insignificant
in most testsets. See also Mr. Cozzie's comment towards the WM test:
it's easy to get a higher score in "king attacking" by simply making
the engine more aggressive, but this does NOT make the engine stronger
in analyzing a king attack.

--
GCP



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.