Author: Uri Blass
Date: 10:44:05 01/14/01
Go up one level in this thread
On January 14, 2001 at 13:36:14, Uri Blass wrote: >On January 14, 2001 at 12:54:35, Dan Andersson wrote: > >>Hope you don't mean using both 64-bit bitboards and 64*32 bit attack board. That >>seems utterly redundant. Incremental updates can be made reasonably fast using >>some clever data structures for each piece on the board.I assume you mean 64 bit >>bitboards by 'bitboards',IMO I really think people should realise that there are >>many kinds of bitboards. If I had access to a machine with 128 or 256 bit words >>I would use that size bitboards. >> >>Regards Dan Andersson > >I think that you do not answer the questions where and how to use attackboard. > >I know almost nothing about bitboards and I am interested to know the fastest >known way to write some evaluation functions with bitboard in order to compare >between bitboard with alternative methods. > >I defined a possible mobility square table evaluation and I am interested to >know the number of nodes per second that programs with bitboard can generate and >also the number of nodes per second that programs without bitboard can generate. > >I think the right way to compare bitboard with alternative methods is to compare >the same evaluation function with bitboard and without bitboard. > >A post with a source code of an evaluation function that calculates mobility >square tables with bitboard and a source code of an evaluation function that >calculates the same without bitboard and comparison between the speed of both >evaluations will be an excellent reply to my question. I can add that the time of updating arrays after generating a move should be also considered in this comparison. In other words I ask only for a comparison between relative simple chess programs. The programs do not need even to use the alpha beta and can generate a full tree of x plies with evaluations of all the nodes. The simple question is if bitboard is faster or slower than an alternative way in genreating the nodes and the evaluations. Uri
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.