Author: Pete R.
Date: 16:02:26 04/03/00
Go up one level in this thread
On April 03, 2000 at 18:54:25, Marc van Hal wrote: >It is treu that most computer program's hang for a long time on one move and >later drop it many times this was the best move it is always intresting to look >at the moves where the program looked for that time. >I have to say in the past this was almost always the fact but not anymore. >One question wich should be made is wich programmer knows what development means >cause this is something wich most programs still don't know and the calculation >that a tempo has a valeu of a pawn is simply a fact so today if you ply the >queensgambit with against a computerprogram with no book then it will try to >play for keeping the c4 pawn wich is a mistake only Nimzo will play d4 c4 with >no book all other programs only see the los of the pawn and choose the Rusian >defense as mainline Not true. Fritz 5.32 without book will play the Ruy Lopez, and if you give it 1. d4 to start it will respond 1...d5 and then 2. c4 at 12 or 13 ply. So by itself it plays Ruy Lopez and Queen's Gambit, both very strong. Of course programs have no clue that the Sicilian is black's best try against 1.e4. Perhaps someday they can prove why or why not.
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.