Author: Thomas Mayer
Date: 03:35:35 07/11/03
Go up one level in this thread
Hi Russell, to keep it short: I believe that you can't order chess program development in stages at all. Afaik we have about 200 WB-compatible engines now and I am quite sure that they all have slightly bit different steps of development. E.g. Dieter told me some days ago that the first Yace could not play games, he just used it as a analyzing tools. Others may have start without any eval at all just to get that thing play. For me, it was quite important that the program starts to play soon - just for motivation... :) Like Uri I haven't implement SAN yet - and I do not plan to implement it at all. Also you work all the time everywhere - e.g. yesterday I found some big bugs in Quarks hashing scheme, now it seems to work way better. Also some search technics might not work at some state in your engine - anyway it's a good idea to try them out later - when something does not work in your engine yet it's anyway possible that it works later. Evaluation is also a field where you are never finished. I also implement from time to time some new features in the UI of the console mode, just because I have some new ideas... Also I plan to implement UCI, but by working on it, I found some other stuff which I have done first... So you might have a to do list, but it's really unsure if you follow that step by step... So it might be a good idea to create a list like you have to start with, but it's pretty unsure in which row you will follow it. At least that is, what most Amateurs do. Greets, Thomas
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.