Author: Uri Blass
Date: 23:35:47 01/13/01
Go up one level in this thread
On January 14, 2001 at 00:39:05, Thomas Mayer wrote: >Hi Eriq, > >>Think about it what is the average rating of IM's, and have comps. not exceeded >>that !! It's simple math people. Let's say you have a 1000mhz tbird 256mb 133ram >>Use any of the top 5 programs against any IM, six game match. I would surely >>put my money on the comps. In fact an old friend of mine played just genius all >>the time and lost most of the time, yes he won some but the match was not even >>and that was on just a pII300 64mb ram. and oh yes he was an IM !! I used to >>take lessons from him. He was one of the best positional players I have ever >>seen. I have played several IM's and I feel he was strong among them. >> >> final thought : comps strength on fast hardware IM if not more. > >well, first of all, it must be said that comps play completely different then >IMs or GMs - but in last some games and tournaments against humans they have, >with something that looks like chess, achieved ratings in range of IMs and GMs. >But there are many things unclear. The machines are often prepared for the >opponent, the humand being itself could not prepare because there are not enough >games of the playing version... The humans can prepare because they can buy the programs they play against them before the match. The change in the program is small so humans have advantage in preperation. I believe that not a small amount of the >strength of an GM is its possibility to prepare to special opponents, to look >through there games, to make some conclusions out of it and to fight against >that opponent... They are not only on the board in finding strategics and >tactics much better then most of us chess blunderer, they are also in analysis >better... Against computers they lost most of there strength concerning to the >analysis - as I said, there are not enough games... When computers would be >allowded to participate in tourneys for some years or so - then we would maybe >know how strong they really are against humans I think that only not commercial programs should be allowed to play in tournaments if we wants to know the real ability of chess programs. Uri , all what we can do at the moment >are some predictions, there is nothing what could be said exactly. Also: With >bying one of the new programs, it isn't like you have an IM at home now, the >program can achieve ratings like IMs but with a complete other style of play... >So they can not really explain what they are doing, they can not plan... >Maybe once there will the time, that FIDE allows comps and then the games will >be made - and I am sure the battle is not yet lost for the humans... still I see >many people, rated around 2000, which can easylie beat every of those new >programs... They may do it in some games but I believe that they lose in most of the games. I will be surprised if an 2000 player can achieve more than 20% against top programs. In the israeli league the teams could choose the player to play against the computer but it did not help and the programs scored clearly less than 50% inspite of the fact that they played against players who were clearly stronger than 2000. if they can it, GMs can it also, when they trainee a little bit on >that... GM's had the possibility to train at home against Junior but it did not help them in the tournament and they scored only 50%. They can win some games but they can also lose. 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.