Author: Uri Blass
Date: 02:45:26 02/10/06
Go up one level in this thread
On February 09, 2006 at 20:12:39, Steve Maughan wrote: >Will, > >I do think Fruit had an impact but I think you are over emphasizing it. In my >view the key events for amateur chess are the following (all dates from memory): > >1996 / 1997: >Crafty - everyone saw the inner workings of a decent chess program. Null move >was there for all to see (yes I know about Chrilly's 1993 paper but this was one >of the first working examples) and null move became a standard part of virtually >every amateur chess program > >1999: >TSCP - showed the world that a chess program didn't need to be that complex to >play chess. This game amateurs a lot of confidence to start and in some cases a >starting point > >2000 - 2003: >Thanks to Christophe (Gambit Tiger) much more attention is paid to king safety >and selective search ideas. > >2001: >UCI protocol. The old school hated it but essentially it makes writing a chess >program easier. > >2002: >Ed shows some of the inner workings of Rebel. (IMO this has had as much, if not >more, of an impact on amateur chess as Fruit). For the first time everyone can >see how a veteran of 20 years has tuned a classic piece of software. We see >Rebel's innovative square control routine and the reduction that Rebel uses. > >2003 - 2005: >I'd call this the post Crafty period, where Hyatt is not the main source of new >ideas for amateurs. Many amateurs (Tord Rhomstad, Sergei Markov [even >Vincent!!] come to mind) start to share some interesting ideas. Notable, >history based reduction, Botvinick-Markov extensions and exotic move ordering >heuristics. IMO this sparked many new ways to think about the search. > >2004 - 2005: >Fruit bounds onto the scene. For me the big learning was that clean and simple >can still be strong - bugs can dramatically reduce playing strength. It's >obvious but Fruit demonstrated it. I'd also say that Fruit marked the end of >the 'bitboards are the future' movement. > >End of 2005: >Rybka - (bitboards are still alive!!) I think the main contribution of Rybka so >far is to re-awaken an interest in complex / accurate evaluations. Fruit et al >had made us think that a simple evaluation, if coupled with a good search can >play string chess - Rybka pull us back in the other direction. We do not know if the evaluation of rybka is more complex than fruit's evaluation so I disagree that rybka pull's us in the other direction. I also do not consider the evaluation of fruit to be a simple evaluation function. There are a lot of chess programs with more simple evaluation than fruit. 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.