Author: Uri Blass
Date: 01:43:23 03/26/04
Go up one level in this thread
On March 26, 2004 at 01:59:07, Tord Romstad wrote: >On March 25, 2004 at 16:29:24, Darren Rushton wrote: > >>Think you will find that Fritz 6 came out in 1999 - you're taling about a FIVE >>year old program. > >Not important. I don't use any techniques which were not commonly known in >1999 (or 1995, for that matter). I think that you use a lot of techniques that were never tried by the top programs in 1999 or 1995. I can say that for me the motivation to start chess programming was because I considered chess programs as weak based on my observations(including the commercial programs). I am talking about search and not about evaluation(chessmaster6000 that was ssdf leader after 1995 has an option to show the line that it searches every 1/2 second and I could see for example that it searches for a long time bad lines when at some point the side to move is losing because of mate in 1 or a move that obviously win material but chessmaster simply did not search that mate but a different line,I did not buy chessmaster9000 but I suspect that chessmaster9000 suffers from the same problem). It can happen also to movei of today but I guess that it will not happen after I rewrite my search(I am in no good mood to look at my current ugly search code and this is the reason for the delay in significant improvement). I also doubt if history based pruning was used in 1995 or 1999. I also doubt if the botvinik extension was used in 1995 or 1999. You also have your ideas for reductions or extensions that were probably not used in 1995 or 1999. 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.