Author: Uri Blass
Date: 09:20:47 08/01/03
Go up one level in this thread
On August 01, 2003 at 11:41:43, Vincent Diepeveen wrote: >On August 01, 2003 at 09:52:00, Uri Blass wrote: > >cut the crap just show your speed. period. >if you cannot do it fast then how can you scan in evaluation fast? I did not say that I cannot do it fast but the task is undefined. For example you talked about scores for order of moves. If a programmer has better order of moves but does the task slower because it takes time to get better scores for order of moves than his(her) program can be better but his(her) move generator can be slower. I believe that movei of today can do it relatively fast to most programs but I did not work much about better scores for order of moves and I use only history tables for scores for the moves that are not captures. I get today with my legal move generator an estimate of 38,000,000/1.2368 that is slightly more than 30,000,000 nodes per second on A1000 after 1.e4 e5 2.d4 d5 and plan to do it faster in the near future but it proves nothing because better scores for order of moves is not the first thing that I am going to try. I used my latest version that is not the best because there is information that I generate but still do not use. 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.