Author: Robert Hyatt
Date: 09:57:34 09/28/99
Go up one level in this thread
On September 28, 1999 at 11:06:56, James B. Shearer wrote: >On September 28, 1999 at 09:25:58, Robert Hyatt wrote: > >>On September 28, 1999 at 08:13:47, Bas Hamstra wrote: >> >>>Hello, >>> >>>A while ago it was posted here that movesorting near the root was much more >>>important than near the leafs. >>> >>>- Is that true? >>>- If so, how much more important? >> >> >>Think about this for a minute. If you get the wrong move first near the leaves, >>how much work does it take to search the wrong move and then get the right one >>to get that cutoff? Compare this to positions near the root. So yes, it is >>_exponentially_ more important to get the move ordering right nearer to the root >>than nearer to the leaves. > > Yes but there are exponentially many more nodes near the leaves so this >is not completely convincing. > James B. Shearer maybe or maybe not. ordering near the leaves, if wrong only one time in the whole search, causes hardly any problems. Ordering near the root, done wrong only one time, causes huge problems. If you have a choice of improving the ordering slightly, and could improve it near the root or near the tips, near the root is the place to be. One simple test is to take a program and not let it follow the PV from the N-1 ply search when it does the N ply search. The search blows completely up, yet we are only wrong at N nodes where N is the number of plies we are searching. Fortunately, things like the history heuristic and hashing tend to 'favor' positions near to the root, so that our ordering is better near then root than near the tips, simply because we use old information to sort near the root, while at the tips, all of that search is pretty much brand new information and we have little to help us sort out there..
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.