Author: Eelco de Groot
Date: 04:29:45 01/27/06
Go up one level in this thread
On January 27, 2006 at 06:39:05, Yar wrote: >Hello, > >I know that a few engines can find this move, but Topalov is sure that comps >can't find it. From the article: > > - Öåííîñòü ýòîé íîâèíêè çàêëþ÷àåòñÿ â òîì, ÷òî êîìïüþòåðíûå ïðîãðàììû åå íå >ïîêàçûâàþò. Òå íîâèíêè, êîòîðûå ïîêàçûâàþò êîìïüþòåðû, íåîáõîäèìî èñïîëüçîâàòü >êàê ìîæíî áûñòðåå, îíè ñêîðîïîðòÿùèåñÿ. Îïàñíîñòü çàêëþ÷àåòñÿ â òîì, ÷òî ëþáîé >ìîæåò íàéòè ñ ïîìîùüþ êîìïüþòåðà òàêóþ íîâèíêó. Òûêíåò ïàëüöåì, âîò òåáå è >íîâèíêà. Òàêóþ íîâèíêó íàäî ïîáûñòðåå èñïîëüçîâàòü, ïîêà äðóãîé íå íàéäåò. >Íîâèíêà, êîòîðóþ íàøëè ìû - 18.Ë:å4 èç òåõ, êîòîðûå ìîãóò ëåæàòü â >íåïðèêîñíîâåííîñòè î÷åíü äîëãî, åå ìîæåò îáíàðóæèòü òîëüêî ÷åëîâåê. > >Last sentence translation: >18.Rxe4! is novelty from those,which can be undiscovered for the very long time, >because only human can find such novelties. >V.Topalov > >You can find article here: >http://www.chesspro.ru/events2/wijk06_vasiliev3.shtml > >Use altavista to translate if you are interested. > >With best regards, > >Yar Hello Yar, my computer don't know Russian! Or is that Bulgarian? It seems to be a matter of "positional taste" for computers wheter they play Rxe4 or not. I tried the position after 20.Qe1 with Glaurung 1.02, Pro Deo and Gandalf and they largely agreed on the variations with at least three possible answers but their evaluations went from 0.22 for Black to -0.67. Three moves 20..f5, 20..Bb7 and 20..Bf5 all lie closely together. Based on the computer-evaluations, it seems White has certainly enough compensation for the exchange but it is not yet a won position at move 20. Eelco
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.