Author: Tony Werten
Date: 22:36:46 09/16/04
Go up one level in this thread
On September 16, 2004 at 13:18:48, Russell Reagan wrote: >On September 16, 2004 at 03:12:33, Tony Werten wrote: > >>Yes. I needed a rewrite anyway, and Borland doesn't seem willing to produce a >>64bit compiler in the near future, wich is a big disadvantage since I wanted to >>use the Kogge-Stone stuff rather than the 0x88 I used until now. ( actually, it >>will be a kind of a mixture) > >How has the Kogge-Stone stuff been working for you? I was never able to get it >to work efficiently enough (rotated bitboards were at least 2x faster). Of >course, I didn't write MMX assembly like Gerd, so obviously it won't be as fast >as his approach. For simple stuff, it might not be so fast but it has nice extras. It works best when doing things parallel. So fe kingsafety (the 8 squares around the king) works very nice since you do all 8 of them at the same time, well, for the sliders that is. At this moment, rotated BB might still be faster, since the fillstuff needs 4 registers, and most of the time actually only uses 2 (eax and edx) and pushes and pops them on/from the stack. With a 64bit processor ( and 16 registers ) it will only need 2 (and have lots off them left) so then it will really fly. So for the next 6 months it will be less optimal, but since this is my first rewrite in 6 years, I have taken a design wich will work also/better in the future. But it is a choice, not a well tested conclusion. Tony
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.