Author: Drexel,Michael
Date: 06:47:50 09/25/04
Go up one level in this thread
On September 25, 2004 at 04:03:26, Uri Blass wrote: >On September 25, 2004 at 03:13:30, Drexel,Michael wrote: > >>On September 25, 2004 at 02:36:28, Uri Blass wrote: >> >>>On September 25, 2004 at 02:09:20, Drexel,Michael wrote: >>> >>>>On September 24, 2004 at 07:43:57, Joachim Rang wrote: >>>> >>>>>On September 24, 2004 at 05:07:52, stuart taylor wrote: >>>>> >>>>>>On September 24, 2004 at 04:44:23, jim r uselton wrote: >>>>>> >>>>>>>Hello, I'm a newbie at computer chess and I have a question. How strong are the >>>>>>>programs you buy right off the shelf. The Fritz, the Shredder, etc., etc. >>>>>>> >>>>>>>Will they play at GM strength or do you need a strong player guiding and >>>>>>>controlling move selection? >>>>>> >>>>>>The top programs on normal PC's of today, will often beat a GM, perhaps even >>>>>>more often than not. If even a GM wants to have a fighting chance, he has to be >>>>>>very familiar with computer style chess. >>>>>> >>>>>>This does not mean that a GM doesn't understand better than a computer. Any GM >>>>>>SHOULD beat a computer in almost any game, but that is now very hard to >>>>>>actualize any more, so you might as well say that computers are equal to a >>>>>>strong GM, but in a certain way. >>>>>> >>>>>>I think that if any GM would analyze absolutely determined to win, like in >>>>>>correspondents chess, then he could win (or draw) almost any game off any >>>>>>computer, even if the computer was also left analyzing for that same amount of >>>>>>time. But the GM would have to work very very hard. >>>>>> >>>>>>S.Taylor >>>>> >>>>>exactly that kind of experiment is happening right now: >>>> >>>>GM = Fide-GM >>>>Arno Nickel is not a Fide-GM. >>>>So that experiment does not happen right now. >>>> >>>>Michael >>>> >>>>> >>>>>http://www.chessfriend.com >>>>> >>>>>Select "GM Nickel - Engines" in the Navigation. >>>>> >>>>>so far it seems an engine with a dedicated and decent (not top) computer >>>>>analyzing at CC-time-controls (several days per move) is playing on GM-Level in >>>>>Correspondence Chess too. >>>>> >>>>>Of course Arno Nickel did some minor mistakes but the match indicates that even >>>>>on CC it is nowadays very hard for a human to beat a computer. >>>> >>>>With the help of a computer it´s very easy. >>>> >>>>Michael >>> >>>If it is very easy then you could expect nickel to get 6-0 result or at least >>>5-1. >>>He is not a fide GM but he is also not a weak chess player. >>> >>>There is a difference between claiming that very strong players can do it with >>>the help of computer(and even this was not proved) and claiming that it is very >>>easy. >> >>It would be very easy for a very strong player. I would expect Ulf Andersson to >>get 6-0 or at least 5-1 for example. >> >>> >>>Note also that playing against a computer does not mean that you know the name >>>of your opponent and I doubt if you can beat a strong private program like >>>falcon at correspondence time control easily even if falcon is slightly weaker >>>than shredder. >> >>I would _always_ get it out of book before move 10 somehow and beat it easily >>although I´m not a very strong player. :) >> >>Michael > >You assume that it is easy to beat a program when you are out of book. Not generally In certain type of positions. Don´t have to be necessarily closed positions but they have to be strategical challenging. Michael >I do not think that you are right. > >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.