Author: Frank Phillips
Date: 06:29:49 11/26/03
Go up one level in this thread
On November 26, 2003 at 05:35:54, Sven Reichard wrote: >On November 25, 2003 at 05:58:18, Frank Phillips wrote: > >>The machine should also have to make its own way to the venue. >> >>I believe the top human players have teams of advisors preparing openings etc. >>Why not start the game from a given position (out of theory) and each side plays >>black and then white. This would negate Kasparov opening knowledge, which often >>nets him many points. We might even get some interesting games. >> >>'Your' proposals seems to be about making it easier for humans to win (or at >>least not lose). >> >>Why stop at openings? > >The question is what purpose Man-Machine matches have. IMHO they can be used to >decide who plays better, i.e., more effectively. If this is the case, they >should be fair, and thus give the same conditions to both opponents. > >A human player certainly analyzes the opponents games prior to the match. Thus >the machine should have access to the human's games and be allowed to modify its >book and evaluation function accordingly. > >A human player has coaches and advisors who help him/her during preparation. >Thus human intervention should be allowed to the program as well. > >A human player analyzes games that occurred during the match, again with the >help of coaches, and modifies his strategy accordingly. Thus, modification of >the book, evaluation function, or program code should be allowed during the >match. > >Coaches can't intervene during a game. Thus, the program is on its own from the >first to the last move. (This doesn't exclude the use of opening books and table >bases, which can be considered part of the program, in particular if they have >been compiled by the program itself.) > >A human player is stuck with his hardware, so the machine's hardware shouldn't >be changed. A program running on different hardware should be considered a >different player (maybe this is a bit extreme). > >If the matches are held for other reasons (e.g., to decide whether the machines >"think" or are "intelligent"), different conditions apply. > >BTW, if we exclude opening knowledge by starting from weird positions, we will >play a game different from Chess. > >Just my two bits >Sven. I was going to respond, but we seem to completely agree. Unless I missed something. Frank
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.