Author: Severi Salminen
Date: 03:39:54 01/30/01
Go up one level in this thread
>I've never done a full implementation of rotated bitboards, just enough on >my first pass at it to decide not to do it on the second go round :). It >is on my list of things I'd like to try though. I feel fairly sure that >it will slow me down, but I can't really know without trying it... Well, try it. I used non-rotated for a while. I wrote bishop, rook and queen move generators in assembler so they were quite fast. But using rotated you can make them even faster - no looping, just a few MOVs and SHLs and ANDs. You'll end up doing more in makemove() but less in other places and that is allways a good thing - you make less moves than you generate. It took a few hours to make then implementation but it was worth maybe 10-20%. Severi
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.