Author: Uri Blass
Date: 14:31:01 01/06/06
Go up one level in this thread
On January 06, 2006 at 16:43:37, Joseph Ciarrochi wrote: >Just as a follow up to this, i timed rykba and fruit at 11 ply and yes, it looks >like rybka takes longer. here are the total game times with ply set to 11. > > > >game Fruit Rybka > 1 57 sec 2 min >2 2 min 10 3 min 43 sec >3 1 min 54 3 min 31 > >When i saw this kind of thing previously, i always assumed that rybka took >longer to get deeper because of its "fat" evaluation function. Now, based on >what you folks have been saying in this thread, I wonder if the reported ply >depth is totally misleading. > >someone prsented compelling evidence that rybka output might be misleading, in >that rybka saw tactics that required ply of three at ply of one > >Ok, here is a really simple question (sorry not a programmer. please bear with >me:)) > >Is it possible to program knowledge of tactics, such that one can recognize a 3 >ply tactic at ply 1? > >best >Joseph I think that it may be possible to find some types of tactics by evaluation(I do not do it but it is one of my ideas to try) but finding very complicated types of tactics by evaluation is too much for me to believe that it is really the case. Uri
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.