Author: Mridul Muralidharan
Date: 03:44:23 07/18/05
Go up one level in this thread
I completely agree with what Thomas Mayer has posted below ... rating is just a means to an end - I dont care for them. I want to find weaknesses in my program and use ICC and similar clubs since most opponents there are on more powerful hardware with better programs. It also gives me a chance to talk to other authors (if they are also ob'ing the game) or get comments from other strong players. For the same , I dont put any formula (other than !wild I think) , or restrict based on timecontrols. - Mridul On July 17, 2005 at 20:39:14, Steven Edwards wrote: >Sharing ICC noplay lists > >Should authors and operators of computer program members on ICC (and on other >servers) share noplay lists? > >My current strategy against cheaters who hide computer play behind a human user >ID is to monitor Symbolic's log and manually enter the suspected cheater's ID >into its noplay list. But this takes time, both my time and the program's time. > I have to check the games, the ratings, the history, etc., and I suspect I an >duplicating work already done by other authors and operators. > >My proposal is to have a unified noplay list shared by all program authors. Its >entries would include all opponent IDs that had one or more of the following: > >1. A "!computer" in their formula. > >2. Repeated disconnects at a rate three or more sigma above the mean. > >3. Play 800 Elo points or more stronger against computer programs than against >humans. > >4. Exhibit obvious cheating; e.g., by playing perfect endgames in nontrivial >positions.
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.