Author: Thomas Logan
Date: 14:42:37 10/01/05
Go up one level in this thread
On October 01, 2005 at 15:27:01, Uri Blass wrote: >On October 01, 2005 at 12:59:35, Thomas Logan wrote: > >>On October 01, 2005 at 12:40:53, Uri Blass wrote: >> >>>On October 01, 2005 at 12:22:18, Thomas Logan wrote: >>> >>>>On October 01, 2005 at 10:53:02, robert flesher wrote: >>>> >>>>> I am also seeing Fruit with a positive score against Shredder 9.1, truly a huge >>>>>achievement for Fabien. We shall we if it continues over 100 games. However, >>>>>after looking at the games i have little doubt. Cheers >>>> >>>>Hi >>>> >>>>Yes it truly is huge achievement for Fabien >>>> >>>>Makes me sad for my favorite engine Shredder should trend continue >>>> >>>>Oh well perhaps SMK will be inspired to reach even greater heights >>>> >>>>My hat is off to Fabien and the Fruit team >>>> >>>>Tom >>> >>>I am sure about it but the same is going to be also for Fabien. >>> >>>Fruit still has weaknesses that Shredder does not have and it extends too much. >>> >>>Limiting fruit's extensions may lead to at least 1 elo increase for fruit(at >>>least 1 can be also 10 or 50 and I am simply not sure how much) >>> >>>Limiting fruit's qsearch so it does not need to search some minutes for mate in >>>1 is going to lead to another elo increase. >>> >>>Some important extensions(see botvinik-markof extension) can again lead to >>>another increase and tablebases may lead to another increase. >>> >>>My guess is that Fruit is going to earn more than 50 elo in the next year(more >>>than 100 is also possible but I prefer to be careful in my estimate). >>> >>>Uri >> >>Yes 100 in incredibly high >> >>even 50 is a pretty healthy increase >> >>I would like to see Fruit go deep and matched against Hydra >> >>Tom > >You may consider it as high but knowing all the disadvantages of Fruit it seems >to me clearly possible. > >I support fruit because Fabien fights with hands behind his back because he >cannot learn from code of Fritz and Shredder that is few months old when the >opponents may learn from code of Fruit2.1 > >I consider Fabien as the best active chess programmer even if Fruit will not >keep the first place and Shredder10 becomes slightly better because of the fact >that fabien is fighting for the first place in inferior conditions. > >Uri Hi Uri How long do you think it takes to analyse the code and implement the concepts contained in the code ? Do you think Fritz 9 has gotten benefit from fruit 2.1? Tom b
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.