Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Taking a stand and a poll

Author: Robert Hyatt

Date: 18:50:23 07/07/01

Go up one level in this thread


On July 07, 2001 at 18:50:40, odell hall wrote:

>On July 07, 2001 at 18:41:55, Robert Hyatt wrote:
>
>>On July 07, 2001 at 17:01:55, Otello Gnaramori wrote:
>>
>>>On July 07, 2001 at 09:53:16, Robert Hyatt wrote:
>>>
>>>>On July 07, 2001 at 00:59:41, Jay Rinde wrote:
>>>>
>>>>>On July 06, 2001 at 23:38:06, Robert Hyatt wrote:
>>>>>
>>>>>>On July 06, 2001 at 10:47:13, Jonas Cohonas wrote:
>>>>>>
>>>>>>>On July 06, 2001 at 09:08:17, James T. Walker wrote:
>>>>>>>
>>>>>>>>It seems that some people continually come up with reasons why computers are not
>>>>>>>>GM strength.  But if you look at the whole picture it's hard to deny.  I am
>>>>>>>>constantly reading here that "a single game means nothing";"A tournament like in
>>>>>>>>Argentina means nothing";"Playing a GM who is not familiar with computers means
>>>>>>>>nothing";"Beating low rated GMs(2500) means nothing";"The GM did not play
>>>>>>>>'anti-computer chess'" etc. etc. etc.  What do all these things put together
>>>>>>>>mean?  Last year I think it was some Spanish IM's that allowed a computer in
>>>>>>>>their tournament and all were embarrased.  Now it's Argentina and the same
>>>>>>>>result.  Now a computer has to beat a 2600 GM to prove it's GM strength although
>>>>>>>>there are many 2500 level GMs who could not do this.  Why are people constantly
>>>>>>>>trying to put artificial requirements on computers that are not required of
>>>>>>>>humans?  I believe one thing is already proven.  If humans play computers just
>>>>>>>>like any other human then computers are definitely at GM strength right now.
>>>>>>>>Also if you want to set up the computer for a fall, it can be done if you have
>>>>>>>>enough control over the conditions.  Some people want computers to be "bullet
>>>>>>>>proof" before they will declare computers GM level.  Just another requirement
>>>>>>>>that humans are not subjected to.  Some point at specific computer weaknesses
>>>>>>>>and say "see that, it can't be a GM if it does that".  Rebel took on some GMs in
>>>>>>>>the GM Challenge and played them fairly even.  Can an IM do that?  If he can he
>>>>>>>>will soon be a GM.  The only difference is a human has the opportunity to play
>>>>>>>>in FIDE tournaments and qualify for the title but computers do not.  This is
>>>>>>>>done in tournaments and not matches where one prepares specifically for the
>>>>>>>>opponent.  So that's where I stand.  Given a fair chance for the title I believe
>>>>>>>>there are several programs that could achieve the GM title.  Of course it's only
>>>>>>>>my opinion and it means nothing except that I've finally taken a stand.  I've
>>>>>>>>walked into the "Computers can be GMs" camp (if given the opportunity).
>>>>>>>>Jim
>>>>>>>
>>>>>>>I second all of the above, well put Jim!
>>>>>>>
>>>>>>>Poll results so far, from my site:
>>>>>>>
>>>>>>>Are computers GM strength ? [126 votes total]
>>>>>>>
>>>>>>>Yes(88)         70%
>>>>>>>No(26)          21%
>>>>>>>Don't know(12)  10%
>>>>>>>
>>>>>>>http://www.geocities.com/vainot/BetaChess.html
>>>>>>>
>>>>>>>Regards
>>>>>>>Jonas
>>>>>>
>>>>>>
>>>>>>I guess that solves that.  :)
>>>>>>
>>>>>>BTW, another "poll" taken almost 600 years ago proved that the world was
>>>>>>flat, too.  If you are into that kind of "proof".
>>>>>>
>>>>>>:)
>>>>>
>>>>>The world isn't flat?
>>>>
>>>>
>>>>Must be.  Before Columbus set sail way back, polls said it was flat.  No need
>>>>to try to sail around the world when you know it is flat.
>>>
>>>Instead of eluding the very well put point of James , why don't you try to reply
>>>to him.
>>>
>>
>>
>>I _did_ reply to his point.  Polls mean nothing except what popular opinion
>>says.  Nothing to do with whether something is a fact or not, just an
>>opinion.  Scientists don't go around taking polls to determine if a quark has
>>mass, they do the experiments to prove or disprove it.
>>
>>Polls are interesting to know what people _think_.  But that is _all_ they
>>reveal.
>>
>>>Thanks.
>
>
>
> Sorry Bob
>
> But the experiments have been done, over, and over, with Two 2700+
>performances, the problem is that you and others are ignoring the results.


I'm not ignoring a thing.  I watch these things play humans and computers day
in and day out.  That is a _much_ larger volume of data than the rare tournament
here and there....



This page took 0.01 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.