Author: Stuart Cracraft
Date: 13:29:07 07/28/04
Hi -- I have an extend on check in quiescence where it hands back to the main search with a depth of 1, without an extend on check in the main search. Without this, problem solution rate suffers. If I add extend on check in main search in all cases except where it just was entered from the quiescence search above with no additional limits besides this, it blows up and problem solution rate suffers. So I implemented a limitation of not extending more than a certain number of times at any given depth (tried 2, 3, etc.) for the check extension in the main search -- no limit in the quiescence search but the trees didn't grow (or lessen) and problem solution results remained the same as in the extend-on-check-in-quiescence-only case. So I am wondering if either my extend on check in the main search is not as useful in light of the already present quiescence check extension or that my implementation of the extend on check in the main search is non-optimal or maybe my implementation of the quiescence check extension is non-optimal. (Bottom line: without at least the quiescence check on extension problem solution rate suffers but there is no additional benefit gleaned from main search check on extension given cu Stuart
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.