Author: Ed Schröder
Date: 10:39:50 09/19/99
Go up one level in this thread
>Posted by Georg v. Zimmermann on September 19, 1999 at 05:36:13: > >Why don't u turn off the "move now" function against other prgrams (excluding >first move) ? You forget one important thing, the autoplayer is a stand-alone driver that does the job. The only thing a chess program does is sending moves to the printer and the (external) driver does the rest. Therefore the "move now" function can't be disabled (protected) as the autoplayer (driver) needs this function. >2.) > >Why isn't crashing considered as a loss for that program that crashes ? But what if the opponent PC forces the crash? It's easy to do. >3.) > >If "weired caracters" cause your program to crash, your autoplayer >implementation is not very "clean". That's a valid point. Not so easy to implement as you have take about every possibility into consideration. If you send 2000-3000 weird characters to the opponent keyboard it is likely the program on the other PC can not handle this whatever is done to protection. >4.) > >If you are afraid of another program detect who it is playing (now this is the >first real interesting point), make sure your program does _not_ move >instantly when in book , or make it (randomly) leave book earlier, or make it think >about every move and just add the book evaluation from first move on ... But shouldn't the autoplayer games not reflect the strength of a chess engine? Ed Schroder >I am 90% sure that you can create an autoplayer implementation in your program >that does not allow the other program to cheat. Testers can still cheat >though, of course. > >Regards, > >Tec--
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.