Wed Feb 24 14:57:57 2016 by Symaw |

Hello there, Can you give me an formula example to start from that will calculate the probability of future rolls based on history of known rolls. For example 2d6: 3 4 1 5 3 6 Probability to role a 2 is higher. The same with 3d6: 1 5 3 2 5 6 3 2 3 Probability to role a 4 is higher. Thank you |

Thu Feb 25 09:44:26 2016 by Torben |

The probability of a future roll does not depend on previous rolls. Dice have no memory.The probability to roll a 4 on a six-sided die is 1/6 no matter how few or how many fours you have rolled previously. Unless the die is loaded, in which case it may have a tendency to land on one side more often than others, but that would be a positive correlation, not a negative correlation. |

Thu Feb 25 12:40:02 2016 by symaw |

I'm not talking about psyhical dice's, it's about RNG - computer genarated. What is the diference? 10x |

Thu Feb 25 15:09:41 2016 by Torben |

Computer-generated dice should not depend on (recent) history either -- otherwise, there is something wrong with them. The reason I wrote "recent" is that no random-number generator on a computer is perfect, so any sequence of numbers generated by a computer will be periodic. The period is, however, very long, so you would need to generate billions of numbers before the sequence repeats. |

Fri Feb 26 09:25:11 2016 by Torben |

I may have misunderstood you. Did you want to make dice rolls that deliberately depend on the history of previous rolls? You can do that in Troll. You just need to keep track of the history in the dice description, and that makes computation of probabilities extremely expensive. For just generating random values, there is no problem. Could you clarify what it is that you want? |

Fri Feb 26 16:08:29 2016 by symaw |

Well, no! Not to generate deliberatly. Only to obtain percentage probability. I still believe it's posible, is just that I'm not a programer, I'm new with scripts and everything "code". I usualy work with examples, but I cound't found none. What I would like to obtain is the percentage increase of hitting a 4 based on the previous numbers of rolls. For example if a 4 hasn't roll on 2d6 after 10 or 20 rolls, there is an 20%-80% probability increase that it will roll. |

Fri Feb 26 16:24:19 2016 by Torben |

You write: "For example if a 4 hasn't roll on 2d6 after 10 or 20 rolls, there is an 20%-80% probability increase that it will roll". Where did you get that idea? |

Fri Feb 26 16:41:39 2016 by symaw |

If you get 2d6 RNG for 5000 or more rolls, and if you count the maximum intervals where a 4 hasn't roll, it can be between 20 - 38 rolls. Usually the interval is less of course. |

Mon Feb 29 09:40:21 2016 by Torben |

And why do you think this implies that the probability of getting a 4 will increase if you haven't rolled one in a long time? |