Author: Frederik Tack
Date: 15:34:57 12/22/05
Go up one level in this thread
> >I think that is a good suggestion but I will probably not implement it. >I would be surprised if the savings is anywhere near 10%. But please let >us know what you find between at the top of negamax/quiescence and >instead after the move within negamax/quiescence. > Tested it. There's no significant difference between the two. In fact, the test on top of negamax is even a tiny bit faster. This is probably because of compiler optimization. I use Delphi and results aren't always what you expect due to this. I have noticed that on some occasions after adding some code, the performance of the engine increases instead of decreasing.
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.