Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Lies.. Damn Lies & Statistics!

Author: Jesper Antonsson

Date: 15:00:55 01/14/05

Go up one level in this thread


On January 13, 2005 at 11:12:09, Dieter Buerssner wrote:
>One other number. The mass of the earth is 6*10^24 kg. It is mainly made of iron
>(the core). But say it was out of oxygen (which is lighter, so we overestimate
>the number of atoms). The atomic mass of oxygen is 16 g/mol. Avogadro's constant
>is 6.022e23 mol^-1. The number of atoms in the earth is:
>
>  6e24 kg / 0.016 kg/mol * 6e23 mol^-1 = 5.76 * 10^46

The calculation is incorrect. (e24*e23 = e47 > e46). The real answer should be
Xe50, I guess. No matter. Some estimate that chess is only 1e40 positions, and
if we are really optimistic about the alpha-beta gain and use some tricks,
perhaps sqrt(1e40) = 1e20 positions would suffice. 1e20 oxygen atoms weighs:

(1e20 / 6.022e23 mol-1) * 16 g/mol = 0.0027 g

1e20 bits is also "only" 12,500,000 Tbyte (of course, a bit per position won't
suffice, but anyway...). I would guess that the installed world hard disk base
is somewhere close to this number, and that these sizes will actually be in
reach of very-high-end database systems within 20 years. Furthermore, if we
calculate one billion nodes a second, we may build such a solution hash table in
1e11 seconds, or 1.7e9 minutes, or 2.7e7 hours, or 1157407 days or 3170 years.
Too much, but only by a few orders of magnitude. Perhaps we can speed it up by
parallelism.

Solving chess may very well be forever unfeasible, but I think the jury is still
out there regarding how much work that has to be done to find a solution, and
while this is the case, we really don't know much about this.



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.