Author: JW de Kort
Date: 01:23:13 09/11/02
Hi, i have read the Thesis of Mr. Kervonkc with much pleasure and i must say that the critism it recieved here in our message board was often too hard. I found a lot of interesting things but unfortunately the descriptions where not very specific. I like to ask two questions: About implicetely testing of the moves legality. Kervinck describes that he does not test a move found eg in the has table directly by checking if the move found is legal given the position but instead that he does this indirectly by a table look up. I understand the general idea but does not it mean that all the moves have to be generated this way? In my engine i first take a move from the hash table and if it is legal this move is tried first and no other moves are generated. As far as i understand the idea of Kervinck, this is not possible in his sceme. Is this correct? If not can anybody explain this methode more precesely? About the attack table: As a see it, this attack table will cost a lot of time to update after every move. Can anybody explain how it works in more detail and how to implement a thing like that efficiently? regards Jan Willem
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.