Re: [Bug-gnubg] Bug-gnubg Digest, Vol 129, Issue 7

[M B]

Howdy Frank,

Playing yourself with random dice would definitely not produce an obvious 5:5 
outcome. Even if GNU played against itself the outcome would favor one side or 
the other because of the randomness of the dice (I'm thinking this may have 
been done many times already). A 7 point match can end 8-0 just like a 7 point 
match can go 4-3. Doubling, gammons, backgammons all matter in the statistical 
nature of the game. To accept a double, refuse a double, "Should I stay or 
should I go", these are all human issues. GNU makes the statistical move 
without emotional cloudiness or having to count or take allot of time to 
remember. GNU is not there to win or lose, its just there to play. It does not 
care what the outcome is. If it did OMG this would be a completely different 
conversation. Visions of Terminator, The Matrix and Demon Seed pop in my head :(

Here's the deal, if I have not explained it already, as human players we are 
not detached from the game. Even if we think we are, there will always be those 
moments when (even we were are playing ourselves), we favor one side or the 
other based on position, good rolls, bad rolls, perceived outcome, etc. I would 
even go further to say we flip flop during games for the same reasons as well 
as possibly to experience a particular scenario we rarely encounter, or an 
extreme comeback or blowout, the list is limitless. The brain is powerful and 
exceptional but it is also flawed when it comes to unemotional decisions. Most 
human beings simply can't perform at the level of a computer that uses years 
and years of compiled statistical data to make its next move based on the 
current roll of the dice and position of the counters on the board.

It has already been mentioned but JD's sample size is far too small and 
emotionally skewed. I just address@hidden, you stay with us.


On Aug 17, 2013, at 12:36, "Frank Berger" <address@hidden> wrote:

Hi JD,

@JD: so if I got you right, you played 10 7-pt matches against yourself and one 
side lost 9 of them. I've never heard of such a occurence in the real life. It 
is an obvious fact that with true random the result must be 5:5 and this can 
statistically surely proved. Too bad that I lack this skill, but for me it's 
obvious.

More serious: writing a good Backgammon AI is difficult. Writing a complex 
program is difficult. Writing a random number generator is far easier. Using an 
international rewarded, hundreds of times tested RNG is trivial. Do you believe 
that all that smart guys developing Gnu-BG all fail in a trivial issue and 
don't recognize it in the last 15 years? 

@all Other: How often do you have convinced one of the conspiracy theorists? 
From my experience it's < 20%

ciao
Frank
_______________________________________________
Bug-gnubg mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/bug-gnubg