Author: Robin Smith
Date: 21:03:48 08/07/04
Go up one level in this thread
On August 07, 2004 at 23:49:26, Stuart Cracraft wrote: >I liked Zobrist and Kalme's research many years ago into this subject. >Kalme would provide a series of patterns to guide the program. Not specific >positions. Another example is David Wilkins program for tactical problems. > >Good to hear this kind of interest is still alive in this search-exhausted >brute-force selective-days. I am pro-patterns and pro-learning! But >the implementation is bedeviling. > >Stuart I can well imagine that "implementation is bedeviling" for "patterns to guide the program". That is why I am suggesting being able to enter "specific positions". Any while I have no doubt that for playing games this would be of no value what-so-ever, I believe that for anaysis it could be very powerful. Any time the analyst can confirm that a position the program says is +3.0, when it really should be 0.0, the analyst can plug the programs blind spot. In other cases, if the analyst wants to avoid some murky looking complications, he could enter a position evaluation of -10.0 for the position after entering the complications and see if there are any other clearer paths to an advantage. I can imagine many possible uses, but only for analyisis, not for improving engine play. P.S. I noticed in you profile that you mention John Stanback. I met him at his home several years back, he lives in the same town as me (or at least he used to). A very pleasant fellow. Do you know if he still does any chess programming? -Robin
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.