Author: Vasik Rajlich
Date: 07:02:00 01/08/06
Go up one level in this thread
On January 07, 2006 at 10:51:22, Vincent Lejeune wrote: >On January 07, 2006 at 04:05:39, Vasik Rajlich wrote: > >>On January 06, 2006 at 10:35:46, Jonas Cohonas wrote: >> >>>>In the next Rybka release, there is a new default "moderately positional" >>>>setting which is about 75% SP and 25% VP. More importantly, a bunch of other >>>>things were changed and some new settings were added. >>>> >>>>Best regards, >>>>Vas >>> >>>This is very interesting and thanks for the additional info on future changes. >>> >>>What about a tool much like the new seperate Rybka WinFinder 1.0, adding a >>>similar program wich is purely positional in nature, for example Rybka >>>MoveFinder 1.0? >>> >>>And maybe even a way to make both programs work together like the triple brain >>>function in shredder, just where winfinder, movefinder and rybka (the normal >>>engine) work together and equally share processor power, but at the same time >>>can be used seperately, just an idea :) >>> >>>Regards >>>Jonas >> >>Re. "MoveFinder" - while it makes some sense to go all the way to the tactical >>extreme as in WinFinder, it makes much less sense to do the same to the >>positional extreme. There are all sorts of tactics which must be resolved for >>good positional play. >> >>Of course, a very positional style of search could be done, but it might be >>twice as fast, whereas WinFinder is (or at least can be, with enough work) >>_much_ faster than a normal engine. >> >>Re. a triple-brain (or double-brain) feature, a user can certainly run both >>engines at once. I am still trying to understand what users want. >> >>Vas > >I think users want is : >1) Advantage of computers : accuracy, speed, tactically strong,... >2) not the disavantage of computers : null moves how hide simple good moves,... >3) Advantage of humans : understanding of some particular positions (fortress, >trapped pieces, continue to search until the eval is very sure, sometimes >analyse wide tree with small depth and sometimes narrow tree with big depth = >depth of the tree could be VERY unequal for each moves) >4) not the disavantage of humans : (I don't know yet what to put here :) ) > >Thanks, >Vincent Vincent, humans have many disadvantages: 1) Limited endurance 2) Not obedient 3) Weak nps 4) Ultra-expensive :) Best regards, Vas
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.