Author: Bob Durrett
Date: 13:02:55 12/07/02
Go up one level in this thread
On December 07, 2002 at 14:32:53, Robin Smith wrote: >On December 06, 2002 at 20:11:31, Bob Durrett wrote: > >>In other words, I propose a comprehensive test which has, itself, been tested >>and verified against the SSDF [and similar] test data. > >It is trivial to desing a test that makes Fritz look smarter than all other >programs. Or a different test that makes Shredder look the smartest. Or >another that makes Tiger look smartest, etc. Your idea of making test results >correlate with SSDF results gets around this, but then what purpose does the >test serve? > >The SSDF already is a test. No set of test positions will be able to predict >SSDF results more acurately than the SSDF results themselves. So what purpose >is served by a set of test positions? The only advantage I can see for such a >test is to get quicker results for programs not tested yet by SSDF. But the >results will also always be highly suspect. > >Robin To answer your question: It would, IMHO, have very little value for testing the absolute top engines, and would not serve to identify "the very best" engine. This is because the top engines are too close to each other in strength. But it would/should be VERY useful for people in the process of developing new chess-playing programs because the test positions give information regarding the specific deficiencies of the program. This is not just a matter of "plugging holes," but also providing a comprehensive evaluation of all functions within the new program. High scores would indicate a very mature program and low scores would suggest "back to the drawing board" [or the programmer's equivalent of a drawing board]. Most importantly, the test results would point to the areas where improvement is needed. > >P.S. Like you, I am also an electrical engineer, and I don't believe your >earlier statement that "Engineers are NOT adept at intellectual sparring!" In >fact, while I have an MSEE degree, I also have a Philosophy degree (the ultimate >degree for those who like to engage in intellectual sparring). I need you to take my place in discussions with Rolf: I am neither a philosopher nor an "intellectual." Also, my IQ is much much too low for that sort of thing! In fact, according to Rolf, I am an "autonomic thinker"! : ) Bob D.
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.