Author: Thorsten Czub
Date: 00:39:30 10/23/00
Go up one level in this thread
On October 22, 2000 at 15:48:01, Ratko V Tomic wrote: >That was a nice overview of possibilities. You final conclusion agrees quite >well with mine, although I came to by different reasoning. > >> So his unit of analysis is not so much data compared by scores, >> but situations compared by his capacity to give a better chance. >... >> So I would bet that Theron has not fulfilled his task just >> replenishing his codes with theoretical lines or patterns, >> but shaping algorithms capable of steering his programs to >> situations where his tactical capabilites shine at his most. > >Yep, that was exactly the position I was justifying from various angles past >several days here on CCC, in discussion with Bob and few others who believe it >is only a trivial matter of attack parameters on steroids. exactly this is the point. and i do believe this is the new thing that brings it into a new quality. the program are suddenly capable to drive into situations they do understand. i would call this a major plan A. if chris whittingtons program BEFORE he implemented similar ideas like Theron, would have been STRONGER (that means having are more sophisticated search) the effect of CSTal would have been stronger IMO. What do we have: CSTal (~1993-1995) Gandalf (2000) Gambit-Tiger (2000) Who is coming next ? i think the group will grow. Next candidates could "the king", maybe junior (?!?) and many many others. IMO you could call this evolution. but i would call it revolution. because it needs to have a different point of view of how to use chess programs components. I think those who call it still "stressing weigthings" or "emphazising speculative terms" have not completely understood the point. IMO they need to call it evolution or "normal and gradual process" because IF they would see the difference, they would have maybe tried themselves before. i am sure if there would be a computer-chess-championship now, shredder5 would have difficulties in winning against gambit-tiger. a program that won both championships in a row, gets suddenly smashed by another program that plays wild attack-chess ?! how can this be ? i would call this a revolution. it killed shredder4 (60/60) around 11.5 - 2.5 with 3 draws! do you have an idea how different a program has to be, that got before 50% (normal rebel-tiger12.0e versus shredder4), that suddenly without many book-changes gets 11.5-2.5 ? and this is not an outsider-result. another guy in the beta team got the same stunning result on his machines with his shredder4. no - you can call it evolution. i do call it revolution. its a small step, yes. but i am sure christophe has the power to increase this effort. and this makes me happy. i know he works in a brilliant team together with jeroen, with ed and others. they will advise him to give his best. The main problem e.g. Richard Lang had, was that he did not know much about chess, but was working in a team with chess-players who did not wanted to change genius, not to change the profit-machine and system. Ossi always said to Richard: pah - richard. this knowledge, you cannot generalize, it is not important to implement this stuff. it`s to rare. ossi IMO had a negative influence on richard. he pushed richard to let anything the way it was. because he saw a risk that genius could lose strength and he cannot earn that much money anymore. this was of course wrong. this way other programs overtook genius. genius was a dinosaur, not capable to adapt the different nature.
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.