Author: Peter Fendrich
Date: 07:28:47 09/11/98
Go up one level in this thread
On September 11, 1998 at 09:07:13, Robert Hyatt wrote: > >I won't argue, but an autoplayer that doesn't allow under-promotion is also >useless. I once introduced a bug into xboard that broke underpromotion, and I >lost two games in a week as a result, one a drawn KNN vs KP where winboard >refused to let me underpromote to a knight which was a check. > >Also, an autoplayer that is so touchy (timing-wise) is also less than useful. > >And finally, one where the protocol is highly unusual and not well-defined >is just as bad. > >Fortunately the winboard/xboard engine protocol is accepted by almost all the >freeware programs. Which is a larger number of programs than those supporting >the archaic auto232 interface. Perhaps time and numbers will settle this. >But clearly, auto232 is a poorly designed interface... It sure is! I think the key to success is easyness... If it's easy to use, all freware and shareware will use it. Probably some of the commercials as well. IMO the design should be a 'black box' that is called from both programs. Instead of _kbhit(), PeekThis(), PeekThat() the program will call the autoplayer , sending commands or polling for commands. The two 'black box' programs are communicating with each other over whatever comunication that is desired. LAN, IP, Null Modem ICC or whatever. The chess programs doesn't have to know... Next step could be to make the 'black box' configurable to some extent, like pairing strings with actual meaning. When all this is working, one could continue with a tournament manager! This would be an exciting project. As usual I am in too many of these and we only have 24 hours between "the sun is rising again" event... :) But I'm willing to participate in some way! //Peter
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.