Author: Robert Hyatt
Date: 19:54:11 10/24/00
Go up one level in this thread
On October 24, 2000 at 09:02:54, Jouni Uski wrote: >On October 24, 2000 at 08:44:05, Robert Hyatt wrote: > >>On October 24, 2000 at 02:26:02, Christophe Theron wrote: >> >>>On October 23, 2000 at 23:11:15, Robert Hyatt wrote: >>> >>>>On October 23, 2000 at 19:26:17, Thorsten Czub wrote: >>>> >>>>>On October 23, 2000 at 17:49:25, Mogens Larsen wrote: >>>>> >>>>>>On October 23, 2000 at 17:17:02, Robert Hyatt wrote: >>>>>> >>>>>>>I know that CST is "lonely" but why group others with it, and say that they >>>>>>>are based on common ideas? That is one giant stretch for computer chess. >>>>> >>>>>let us wait a few weeks - that you can test it too. and others have tested >>>>>it too. so that we can discuss again on the base of data we all can prove... >>>>> >>>>>the beta testers of the rebel team posted >>>>>results in this forum and got attacked here, mainly from a guy called >>>>>mogens... they were attacked because - as he said - their euphorical reaction >>>>>was not based on FACTS and not based on objective-judgement. >>>>> >>>>>now this thing won both championships very convincingly. >>>>> >>>>>i am sure you will find ways to oversee objective FACTS and judgement in the >>>>>future, if this helps you to prove your point of view. >>>>> >>>>>its easy: you only have to wait until people forget how you attacked >>>>>a bunch of people who HAD that thing, because you had NO data, and only >>>>>hear-said, but you know it better than these people. >>>>> >>>>>thats something very interesting. That OBJECTIVE and rational people, >>>>>who found their judgemant on FACTS, do have more insights in something they >>>>>never tested, than 21 other people who tested it for weeks on autoplayers >>>>>against all kind of programs. >>>>>and then jump on those people in an open forum and talking about >>>>>propaganda and subjective-meanings, campaigns and all the mud you throw... >>>>>when in the end you had nothing than an opinion. no single data that was >>>>>NOT posted by somebody else. or hear said. >>>>>you have not seen a single main-line, nor a score live, and you >>>>>felt yourself that kind of confident to jump on honest and >>>>>normal people. I would >>>>>call this a mastepiece of arrogance. a kind of mega-outing. >>>>> >>>>>>One small leap for Thorsten, one giant stretch for mankind :o)). >>>>> >>>>>i am sure you will find ways to make the people forget about your attempt >>>>>to throw mud on unguilty people, just because you did not like the facts >>>>>these people presented. its not up to you to decide what is fact and what is >>>>>fantasy, what is objective and what is subjective. >>>>> >>>>>If i have nothing, i would close my mouth, be silent and study in my >>>>>room, what others have to present. i would silently replay their games. >>>>>and when the programs comes out of the market, i would prove them wrong. >>>>>and THEN open my mouth. >>>>> >>>>>you did it other way arround. very fine. its your decision. >>>>> >>>>>but don't speculate that this will be forgotten too soon. >>>>> >>>>>>Mogens. >>>>> >>>>>gandalf plans. cstal plans. and gambit-tiger plans. all 3 mate-attacks. >>>>> >>>> >>>> >>>>Those programs couldn't tell the difference between a "plan" and a number >>>>two washtub. To imply that they "plan" is something so ridiculous as to not >>>>even warrant another comment. >>> >>> >>> >>>So Morphy did not plan either. >>> >>>You don't know how these programs work, but you say they don't plan. >> >> >>I know how _all_ programs work. And yes, I say they do not "plan" by any >>definition of the word that a _human_ would use. I will be happy to pick >>a game (when I have time) and show you why I can say this. I can tell you >>that it is quite easy to see how programs play the game... just look at the >>game and you can see why there is _no_ planning of any kind. >> >>I would hope you are not going to try to convince me that GT is now "planning" >>which means to first figure out what you are going to try to do, then only make >>moves that further that "plan". >> >> >>> >>>I'm sure you would say the same for many games played by human grandmasters if >>>you did not know the games were played by human beings. >> >>Not likely. It is not hard to see their overall plan for the most part, by >>looking at weak squares, weak pieces, etc. IE the KID is well known as to >>what black is going to try to do. Show me where GT "plans" anything there... >> >> >> >>> >>>I think the time for the Turing test applied to chess programs is coming. Take a >>>random collection of championship games from the same entity, and tell if they >>>have been played by a computer or a human chess player. I believe it is possible >>>that some chess programs can pass the test. >> >> >>I don't think it is even _close_ to happening. >> > >It has happened already many years ago! For couple of GMs were shown games from >Aegon and thyy can't identify computers... > >Jouni Maybe there was a reason? Like they were not the _right_ GMs to ask? I'd bet today's GMs would do _far_ better at this since most have now had significant experience battling computers. I have talked to _many_ on ICC and they have become quite good at recognizing 'cheaters'... I don't think you can take one game and reliably say human or computer today. But given say 4 games, it is not hard. IE 4 games by the same computer, 4 by the same human, and it can probably done with nearly 100% accuracy. And if you let _me_ sit across the board from the computer in a "turing test" I'll bet I can find the computer in minutes. It is _not_ hard if you get to feed the computer and the human several positions. It becomes trivial then.
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.