Not signed in (Sign In)

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.

    • CommentAuthorgriz
    • CommentTimeJan 5th 2009
     
    -----------------
    Random numbers are far too important to be left to chance.
    • CommentAuthorgriz
    • CommentTimeJan 5th 2009
     
    my first effort.
    not really a puzzle ...
    just click run and watch.
    looks best in slow motion.

    griz-----------------
    Random numbers are far too important to be left to chance.
    •  
      CommentAuthorTapir
    • CommentTimeJan 6th 2009
     
    Can't image it always successes!-----------------
    My games: Tapir Games
    My phyards: Tapir@phyard
    • CommentAuthorgriz
    • CommentTimeJan 6th 2009
     
    yes ... was very tricky.
    any change whatsoever ... in density or placement ...
    and the whole thing is different.
    question ...
    is there a random function we can use?
    and if so ... a seed on can replicate to start the
    pseudo randomization?
    thanks-----------------
    Random numbers are far too important to be left to chance.
    •  
      CommentAuthorTapir
    • CommentTimeJan 6th 2009 edited
     
    do you want to make the simulation get the same result in each play?

    I have not study box2d ( http://box2d.org ) very deeply now. So I don't know how the random factor affect the whole simulation.

    The other factor is the simulation time step. Now the time step is the real time interval, it is not fixed, which means it also has some randomness. Maybe I will add the fixed time step feature later.-----------------
    My games: Tapir Games
    My phyards: Tapir@phyard
    • CommentAuthorgriz
    • CommentTimeJan 6th 2009
     
    yes, i noticed the time step can change the interaction.
    i was surprised that the above puzzle reproduced exactly ...
    that is GOOD! now we can have a scientific experiment. ;)

    ideally, it is great to have the simulation reproduce exactly
    depending on the starting seed so one can retain some control
    and repeatability ...
    but still introduce 'pseudo randomness' ...
    because then you can end up with very interesting results ...
    but are able to reproduce them because you are starting
    with a 'seed' for the 'randomization.'
    so if you like what you see ...
    you can repeat it.

    great forum ...
    this is exactly what i was looking for ...
    trying to find more stuff like 'Day of the Bobteds'
    do you know it?
    http://foon.co.uk/farcade/bobteds/
    this is what got me interested in flash programming ...
    [which i still know nothing about!]
    the physics one could simulate.
    his 'netshift' is pretty cool too.
    http://foon.co.uk/farcade/netshift/

    fun stuff
    thanks for this site.
    and it's just beginning ...
    who knows what we will discover? ;)-----------------
    Random numbers are far too important to be left to chance.
    •  
      CommentAuthorTapir
    • CommentTimeJan 6th 2009
     
    Seems I have played Bobteds about half year ago. Great game, also netshift. Thanks for your infomation.

    I will study the feasibility of your feature request. I also think this feature is very useful for many cases.-----------------
    My games: Tapir Games
    My phyards: Tapir@phyard