Author: James T. Walker
Date: 13:38:48 02/19/01
Go up one level in this thread
On February 19, 2001 at 13:32:13, Mike S. wrote: >On February 19, 2001 at 13:21:18, Joseph Merolle wrote: > >>(...) Does it mean that there are 233 positions >>that a capture can lead to. In other words if it says tb=60 does that mean that >>from this position it could end up in one of the 60 positons in the tb. > >Yes... but not necessarily end up, it just means that 60 of the (millons of) >positions in the search tree have been found in the tablebases, which means that >these 60 positions had database material only (i.e. KR-KN or whatever, depending >on which tb's are installed). > >>and if >>this is true how come the engine does not kick off and get the eval from the tb >>instead of the engine calulations? Thanks for the help > >It gets the eval, but only for these positions somewhere in the tree. It "kicks >off" (if I understand you correctly), when a database material is reached in the >game and is at the board already. Then, no tree search is necessary. > >(Is this correct Prof. Bob?) > >Regards, >M.Scheidl I think the "hits" result when the search leads to a tablebase position. If Fritz could force the game to this position then it already knows the outcome of the game and may announce mate/draw. If these positions are good for Fritz then it will try to move the game in that direction. If the positions are bad for Fritz it will try to avoid going in that direction. I have seen Fritz get tablebase hits in under 1 minute with 15 men on the board.(5 man tablebases) Jim
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.