Author: Uri Blass
Date: 09:06:48 04/02/02
Go up one level in this thread
On April 02, 2002 at 11:03:12, Jorge Pichard wrote: >On April 02, 2002 at 09:16:06, Uri Blass wrote: > >>On April 01, 2002 at 20:35:29, Jorge Pichard wrote: >> >>>Here is a game where both programs NOT knowing what to do in the opening, since >>>they are playing without Opening Books Managed to play a decent game of the >>>Orangutan or Sokolsky Opening. Again I forced the first move 1.b4.... tomorrow I >>>will let both programs select their own first move. Uri needs to make an opening >>>Book for his program Movei. >> >>I think that there are more important things and movei >>did not get a losing position out of the opening. >> >>movei even could get a better position and win a pawn by 19.b5 >> >>I find it strange that movei searched only to depth 2 when >>it played 19.Bf3 and I will try to find the problem(if it is a bug in my program >>or maybe the celeron was slowed down for some reason). > >Yes it is strange that it only searched to depth 2; I will replay the game from >move 19.... today. > > Pichard The problem continued also at moves 20 and 21 and was only fixed later. It did not do a stupid tactical error in these moves but it is possible that it could find better moves with deeper search. I checked only move 19 and my program considers Rgf1 to be the best move at depthes 7,8,9 Another note is that I found no time control that my program play both 2.Bb2 and both 4.Bc3 and on my computer at time control that is slow enough to play 2.Bb2 my program always play 4.h3 and the only explanation that I can find for both moves is assuming that the celeron's speed is not proprtional to my program's speed. I hope that there will be no problem in the game because my tests were only from the opening position and I am afraid of a bug in the time management if my program does not know correctly the number of moves for the time control. My program usually gets this number in the beginning of the game and it is reducing it by 1 every time that the computer is to move. It can cause a wrong assumption about the number of moves when the game does not start from move 1. Uri
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.