Author: Peter McKenzie
Date: 20:04:31 10/22/98
Go up one level in this thread
On October 22, 1998 at 21:38:29, Fawna Bergstrom wrote: >Well everyone has their opinions on this kind of question--here are a few of >mine. Let's go back to basics: > >Level I: You search full-width to a fixed depth (alpha-beta, iterative >deepening, etc. are all assumed, of course.) Here your evaluator includes both >material and positional factors. Move ordering is critical. First expand >"killer" moves, "interesting" moves and moves that yield a higher evaluation. > >Level II: If you like you can then search beyond that looking at "interesting" >moves such as captures, threats, checks, etc. Don't bother with threats unless >the threatened piece is hanging and/or more valuable than the threatening piece. > You should limit the depth of this second phase or you can skip it altogether >and go straight to level III--it's your call. In level II the evaluator >adjusts >for material only. Personally I wouldn't waste too many plies on Level II. I'm not 100% sure what you mean here, but if you start returning scores from the quiescence search that don't take into account changes in evaluation due to captures effecting pawn structure, and these scores can find their way into your PV, then you're likely to get killed positionally. > >Level III: Beyond that you MUST look at checks by a side that isn't winning, >moves that get out of check by a side that isn't losing, and captures of a >piece >larger than the current material defecit, by a side that isn't winning. (For >example, if it's White's move and she is a Knight down, look at a check or a >rook capture but don't bother with any pawn captures. On the other hand, if >it's White's move and she is already a rook up, then who cares if she can also >capture the queen.) Examine large captures before smaller ones and checks last >of all. This third phase has NO DEPTH LIMIT and adjusts for material only. >Simple threats and other less-interesting moves are ignored in Level III. Its debatable whether its worthwhile looking at checks in the quiescence search. Some do, some don't. My program looks at some checks early in the q-search only. > >Of course, none of these issues will do you any good if your program is not >super-efficient. When you hit a "wall" and can't make any more conceptual >improvements, then throw away every line of code and write a new super-optimal >version. Examine public-domain source code for ideas to improve efficiency. > >I liked your question--very interesting topic.
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.