buggnubg - Bugs: bug #45552, UNEVEN DISTRIBUTION OF DOUBLES



Submitted by:  None
Submitted on:  Wed 15 Jul 2015 06:20:27 AM UTC  
Originator Name:  Ian White Originator Email:  -email is unavailable-
Item Group:  Other bug Category:  Other
Release:  1.04-Release Privacy:  Public
Operating System:  Windows 7 OS Type:  64Bit
Processor Type:  Intel Severity:  3 - Normal
Open/Closed:  Closed Assigned to:  None
Priority:  5 - Normal Status:  Works For Me

Add a New Comment(Rich Markup)

You are not logged in

Please log in, so followups can be emailed to you.


Wed 15 Jul 2015 06:42:06 AM UTC, comment #1:

Since you recorded it does that mean you saved an SGF file? (Saved the match) What source of randomness do you use (Mersenne/Random.org etc). Are you using GNUbg 1.05.000 (I noticed that isn't an option in the bug tracker oops).

Beyond that anything can happen. Randomness doesn't guarantee evenness of the rolls.

I'm willing to review the match, but beyond that I have done dice studies of 20 million rolls of Bot vs Bot and there isn't an issue with the source of randomness (default Mersenne Twister)

GNUbg doesn't roll in its favor. It gets the same type of dice as the human. It just plays better than most and its skill level often makes it appear like it is luckier (the bot makes the best possible play according to the AI so it minimizes the number of potential bad results on future rolls).

To do a proper study one would need to record (save) a large number of matches and then do a study. You can't hand pick the matches either. I'd be willing to analyse a large set of matches. (They can be emailed to me at -email is unavailable- )

I ask about the random source being used (it can be changed in options). If using random.org you need to upgrade to the latest version (1.05 or higher) because that site made changes that screwed up GNUbg and the rolls it retrieved.

Michael Petch <mdpetch>
Project Administrator
Wed 15 Jul 2015 06:20:27 AM UTC, original submission:

Dear Sir,
I started using this program about a week ago and have found the dice 'behaviour' increasingly erratic. This morning I started to record one of the oddities.

Playing in the gnu v human mode the ratio of doubles, over the last hour of play, has been 33 to 7!!


Thank you,

Ian White



(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:

No files currently attached


Depends on the following items: None found

Items that depend on this one: None found


Carbon-Copy List
  • -email is unavailable- added by plm (Updated the item)
  • -email is unavailable- added by mdpetch (Posted a comment)
  • -email is unavailable- added by None (Submitted the item)

    Do you think this task is very important?
    If so, you can add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.




    Follow 2 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2018-08-11 plm Open/ClosedOpen => Closed
        StatusNone => Works For Me

    Back to the top

    Powered by Savane 3.4