Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

How to RNG

ButeoRegalisButeoRegalis Member UncommonPosts: 594

I'm trying to come up with a way to generate random numbers, ideally with something like dice, that would be trending more to one side of the scale than the other, say in a range of 1-10, 1-20.

E.g. on a 6-sided die, you have the same chance to roll any of the six numbers.

If you roll 2 dice, take the sum of both die you are more likely to roll a 7 than a 1.

I'm looking for something that if you roll two dice you are more likely to roll a 12 (one side of the scale) than a 2 (the other side of the scale), with values in the middle of the scale having a "medium" probability. Obviously, this doesn't work with two dice, but that's the idea.

 

Any thoughts?

image

Comments

  • FireBatsFireBats Member Posts: 2

    Doing coding back then I had to do something similar, the trick I used was just use % based instead, let say an evenly distrubuted roll of 1 to 12 is about 8% each, you can make 1 to 4 5% each and similar

    a) seed a 1 to 1000 random

    b) divide by 10

    c) with your XY.Z% distribute your 1 to 12 random

  • AtaakaAtaaka Member UncommonPosts: 213

    Back when I cared...

    Ran 1-1000 

    Are people still using that?

     

     

     

     

  • ButeoRegalisButeoRegalis Member UncommonPosts: 594
    Originally posted by FireBats

    Doing coding back then I had to do something similar, the trick I used was just use % based instead, let say an evenly distrubuted roll of 1 to 12 is about 8% each, you can make 1 to 4 5% each and similar

    a) seed a 1 to 1000 random

    b) divide by 10

    c) with your XY.Z% distribute your 1 to 12 random

    Yep, that's what I'm after, though I was hoping to achieve this with dice, or a similar implement usable in a board game.

    I could get there in a coarse manner by just mapping multiple roll results to the same outcome.

    If I wanted to generate numbers 1 through 6 with different distributions, I could use 2 dice, add them and use the sum like so:

    sum1, sum2, etc (maps to) result, probability

    2, 3, 4, 5 -> 1,  10/36

    6, 7  -> 2,  11/36

    8, 9-> 3,  9/36

    10 -> 4,  3/36

    11 -> 5,  2/36

    12 -> 6,  1/36

    The different probabilities to roll a certain sum with 2 dice govern the splits, plus keeping the consecutive numbers together. I could have 2, 3, and 6 map to the same result, but that would make things more complicated.

    This mapping isn't quite as nice what I had hoped for. Maybe I'll get a d12 or d20 instead, then I don't have to worry about the sum 7 being so much more likely than a 2 or a 12.

    Hmm, maybe if I use different dice, like a d4 and a d6...

    image

Sign In or Register to comment.