Author: Jan K.
Date: 10:17:14 05/06/05
Go up one level in this thread
On May 06, 2005 at 11:41:25, Ryan B. wrote: >Using the idea of having a White attack board and Black attack board (WB and BB) >from http://members.home.nl/matador/chess840.htm I assume mobility is taken by >simply counting the squares each piece you are counting mobility for attacks. >Once you have the date in WB and BB this seems very fast and simple to implement >however I see a huge problem with bishops using this method I have not been able >to find a cost effective answer for yet. If you have 2 bishops you will need to >know how mobile each bishop is to determine individually if the bishop is weak >or strong. (I need to color square of the strong bishop for other parts of the >eval.) Simply counting the squares attacked by bishops does nothing to tell you >what color the squares are that are being attacked. Does anyone have a cost >effective way of implementing strong/weak bishops using this method of mobility >or have I simply misunderstood how mobility should be done with this data? >Thanks, >Ryan I also implement bishop mobility through attack tables. I simply keep another table(int[64]) of additional information for both sides, the number of attacks takes first 6 bits of the integer. Of course if you just want to store mobility, an array of bytes will be enough, but you can store a lot of other data into 32-bit integer. And here the slowdown isn't that bad, because you don't have to erase this table every time you generate attack tables. That consumes some time for WB and BB if I understand the concept.
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.