Author: Marc Lacrosse
Date: 13:34:49 06/19/05
Go up one level in this thread
On June 19, 2005 at 16:20:21, Roger Brown wrote: >>Great result! >>from this result I can now conclude that fruit is much better than ruffian. > > > > >Hello Madhaven, > >Really? > >On this basis you are prepared to say that a program with a SSDF rating is lower >rated than this one? On the basis of 22 blitz games per engine? > >Where are the comparative Ruffian games? > >Have you analysed the games themselves? Does Fruit demonstrate some remarkable >understanding of the game that exceeds Ruffian's? > >You are of course *free* to conclude what you will but it seems a wee bit too >early to call either way. > >Oh, I am a Fabien/Fruit fan incidentally. I completely agree with you, Roger. Such a small amount of games has nearly no statistical signification (see the "+" /"-" values). I just wished to know wether Fruit was able not to be ridiculous at such timing against the best ones. That's already more than twice the number of games that Fruit played at Massy (and then Fabien told me its third place was due to the fact that Fruit was not to good at blitz). Regarding the quality, I briefly reviewed the majority of the games : they are globally very good and sometimes brilliant. But there are already lots of larger tests : in 48 hours Fruit really established itself as a true contender for the title of best program in the world, at any timing (see CEGT)! Amazing : when I published preliminary results yesterday I had comments saying that there was surely a flaw in my tests ... A star is born ... Marc
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.