Author: Robert Hyatt
Date: 13:37:27 06/23/01
Go up one level in this thread
On June 23, 2001 at 10:59:47, Chessfun wrote: >On June 23, 2001 at 07:19:13, Lin Harper wrote: > >> If JRLOK or anyone else finds weaknesses in a chess program >> and seizes on them to lift his rating above what it normally >> would be, then that's fine. This guy has got Crafty's number, >> and he's making the most out of it. > >I don't think that's exactly true. >Roman finds weaknesses and probes them till he wins, then he hammers >them. Bob fixes these holes, then Roman finds others. IMHO Roman could >do this with any PC program. In the case of the other programs once Roman >had found the hole he could get to 3800 as there isn't anyone to fix those >other programs holes. > This actually doesn't happen. He _does_ let me know immediately when he finds a "pattern" of bad play... he gives me specific details on what is wrong and what might help... > >> If Crafty loses regularly to him then surely that's fault in >> the program, and it's the programmer's fault (no offence). Don't >> these programs learn by their mistakes? > >It isn't the case of a simple mistake. Roman seizes a positional advantage that >he knows will win in an endgame and he simple steers the game to it. There isn't >much learning that can be done against a player such as this. > >Sarah. This is correct and is what most overlook in the computer is a GM discussions. Once he finds that weakness... and I should note that it is _not_ only crafty (scrappy) he does this against. He has shown me many games he has played against commercial programs to (a) show me where scrappy is playing with the right 'plan' relative to pawn structure and the other program did not; (b) where the commercial program is doing something right that I am not. This is pretty common too. But he has racked up _many_ programs...
This page took 0.01 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.