Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Arrow of Time Issue



In the context of flipping five coins:

On 03/13/2012 04:11 AM, Brian Blais wrote:

"only 32 microstates?" there are more microstates than that,

There is a separation of variables. The heads/tails part of the
system has only 32 microstates. We can understand the microstates
and the entropy of this subsystem separately from the rest of the
system.

of course I realized that you were only referring
to the heads/tails part of the state, but the position of the coin in
the box is also part of the microstate, yes?

Yes and no. It depends on what you mean by "the" system and what
you mean by "the" microstate. Again: We can factor out the
heads/tails subsystem and describe its microstates and its entropy
separately.

Separation of variables works for the entropy whenever the variables
are statistically independent. You can prove that it works, as an
easy corollary of the definition of entropy and the definition of
independent probabilities. The details are worked out at
http://www.av8n.com/physics/thermo/entropy.html#sec-s-independent