Main
Date: 05 Apr 2005 16:50:42
From:
Subject: Is Chess Almost "Analysed Out"?
Have the possible openings in the game of Chess been so thoroughly
analysed that Chess will soon be "analysed out"? Here are my thoughts
on the subject.

(1) Respected estimates of the number of possible board positions in
the first 40 moves of a Chess game range from 10^40 to 10^120. That's
from 1 followed by 40 zeroes to 1 followed by 120 zeroes:

http://mathworld.wolfram.com/Chess.html

To give a sense of how big the number 10^120 is, typical intelligent
estimates of the number of particles in the observable universe tend to
fall in the neighborhood of 10^80. So, if we take the number of
particles in the observable universe according to this estimate and
multiply it by 10,000,000,000,000,000,000,000,000,000,000,000,000,000
(!!!) we get a good approximation of the number of possible positions
in the first 40 moves of Chess, according to one widely-respected
estimate.

I don't think we'll be exhausting Chess any time soon :)

http://www.varatek.com/scott/bnum.html

2) Even if there were a Flawless Chessplaying Supercomputer that had
defeated every human being in every game played against it over many
centuries, that wouldn't keep human beings from playing against one
another. Consider...

An inexpensive, poorly-maintained automobile that can manage 60 miles
per hour can outrun the fastest human being living (easily). There are
many millions of such automobiles (including several in front of my
next door neighbors house), yet this does not in any way diminish the
thrill of one human being running a race against another. Even so, if
we had many millions of computer machines that could play Chess
"perfectly", that would in no way diminish the excitement and challenge
of one person playing against another.

The number of possible openings combinations in Chess is so large, that
even if we had The Definitive Book of Perfect Chess Openings compiled
by The Flawless Chessplaying Computer mentioned above, focusing ones
Chess study time on openings would still be a pretty poor way to become
a strong human player, in my humble opinion. A certain amount of time
spent on openings is fine, but I have to think that focusing on
strategy, tactics, and positional concepts will generally yield a much
better result, even at the highest levels of play.

David Brett Richardson
http://www.100bestwebsites.org/
"The 100 best sites on the Web, all in one place!





 
Date: 08 Apr 2005 11:41:45
From:
Subject: Re: Is Chess Almost "Analysed Out"?
I have heard before that there are more possible moves in a chess game
than there are atoms in the universe. But it is still a finite number,
and chess will probably one day be solved. I doubt that the solution
will demonstrate a perfect game of chess, i.e., that White can win by
force from the initial position or that Black must play a particular
sequence of moves to draw. It's more likely, imho, that the solution
will confirm what we already know: that a mistake-free game of chess
is a draw, that there are a lot of ways to get to that result and that
there are even more ways to go wrong. It's those millions and millions
of ways to go wrong that will keep chess interesting for humans.



  
Date: 10 Apr 2005 14:33:05
From: David Richerby
Subject: Re: Is Chess Almost "Analysed Out"?
<[email protected] > wrote:
> It's more likely, imho, that the solution will confirm what we already
> know: that a mistake-free game of chess is a draw, that there are a lot
> of ways to get to that result and that there are even more ways to go
> wrong.

`Know' is putting that rather strongly, even though I realise you're using
the word facetiously. Most people seem to believe that chess is drawn
with perfect play by Black but there's a sizeable minority who believe it
to be won for White and you might even find that some people believe it's
a win for Black.

There's an intriguing information-theoretic argument for this last case,
due I think to Mikhail Suba. The idea is that White has to commit himself
first so Black has an advantage from having more information about the
position. I think the argument is very elegant but I don't think the
advantage there is enough to compensate for White's initiative.

For the record, I'd guess that there's something like a 75% chance that
chess is drawn with perfect play, and, if it isn't, it's almost certainly
won for White rather than Black.


> It's those millions and millions of ways to go wrong that will keep
> chess interesting for humans.

Amen.


Another point, that nobody seems to have mentioned is that, if chess were
shown to be a forced win for White and people started to force the win,
the opening(s) in question could be banned from tournament play, in much
the same way as has been done for draughts. Players could be forced to
use those openings which, while still theoretically drawn, give best
practical chances for a mistake from either side -- I suppose that would
be called `words practical chances' in ordinary parlance!


Dave.

--
David Richerby Carnivorous Puzzle (TM): it's like
www.chiark.greenend.org.uk/~davidr/ an intriguing conundrum but it eats
flesh!


  
Date: 08 Apr 2005 18:56:05
From: Morphy's ghost
Subject: Re: Is Chess Almost "Analysed Out"?
In the year of our Lord 8 Apr 2005 11:41:45 -0700,
[email protected] wrote:

>I have heard before that there are more possible moves in a chess game
>than there are atoms in the universe. But it is still a finite number,
>and chess will probably one day be solved. I doubt that the solution
>will demonstrate a perfect game of chess, i.e., that White can win by
>force from the initial position or that Black must play a particular
>sequence of moves to draw. It's more likely, imho, that the solution
>will confirm what we already know: that a mistake-free game of chess
>is a draw, that there are a lot of ways to get to that result and that
>there are even more ways to go wrong. It's those millions and millions
>of ways to go wrong that will keep chess interesting for humans.
>

I still like to think that Weaver Adams was right and that the Vienna
is a forced win for white, though.





The fox condemns the trap, not himself. -- William Blake