Computer Chess Club Archives


Search

Terms

Messages

Subject: solving chess

Author: jefkaan

Date: 11:07:13 11/21/02


1) A few weeks ago someone came with the suggestion to
change the stalemate rule into mate; ok, in such a case
i think chess can be solved (presuming stalemating the
king is a win for the side with pieces and not a loss).

2) although after a few decades i'm convinced the best
chess computers, and maybe some superKramnik humans
will play 99.9% draws, the result for hte remaining .1 % will
not be evenly divided between white and black; in fact
the winning chances for white will be probably amost
twice as high as for black, i suspect.

3) as the above fact, advantage for white, increasing
a/o as a result of improving opening theory, is a
deficiency in Tpr/Elo calculations (as they don't distinguish
between white/black) which should be improved;
or..:

4) with some other reasoning one could say the advantage
for white is not fair in chess; in Go for example, black
as first players has an advantage, but in the end-result
this is compensated with the socalled 'komi' rule.
In chess therefore white should be handicapped, eg.
playing with 3horses/bishop instead of horse/bishop,
replacing one rook by a bishop, or ... (my favorite):

5) leaving out the pawn at f3 !!!

Now if y'all hackers start programming this
variants with some nice inline assembling code for
attack tables and so forth we simple amateur
chess players can all be happy. And finally,  well yep indeed
this message was not about solving chess, but more about
some new chess variants; but the heading 'solving chess' came
up some time ago, and seemed to attract more readers :)



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.