Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

Hey all. I am looking at dice apps for some upcoming Apoc action. My opponent and I have agreed to be open to the use of an app seeing as how we both will have access on a tablet. The reason for using the app is we are planning a 20k/side Orks vs Tyranids game; the kind of dice rolling we are going to see is going to be just silly.

My question is what kind of statistics in my analysis should I bother looking at while I put this thing through its paces? Don't need to look at outliers because all the results will be within set parameters. Standard deviation will be nice to see; I'll be looking at it for some wander to indicate possible problems with the RNG. Basic distribution of course. Any other suggestions?


Automatically Appended Next Post:
Also, if anyone has any suggestions on a particular app they use, I'm all ears. Mostly where I see the potential for time being sucked up by physically rolling large numbers of dice is in counting up your distribution. Something that makes the rolls and shows how many of each number came up would be the best, other functionalities would be nice but secondary.

Roll 1000 dice and see where the numbers lie you should get roughly 166 of each number.

I'm not speaking from personal experience but a couple of friends of mine ran a 15K apoc game, they had vassal's 40K modual open on a tablet and just input the numbers into it to get dice results.

They liked it enough that they intend on using it again the next time they run an apoc game.

Edit: Correction they weren't playing apoc they were running a normal 40K with triple FOC slots and extra objectives, I'm not sure that makes a diffence as I'm not familar with apoc rules.

That's a damn good idea. I was planning on bringing a tablet or two, might just bring the laptop. Thanks, dripwelquest

grizgrin wrote:Hey all. I am looking at dice apps for some upcoming Apoc action. My opponent and I have agreed to be open to the use of an app seeing as how we both will have access on a tablet. The reason for using the app is we are planning a 20k/side Orks vs Tyranids game; the kind of dice rolling we are going to see is going to be just silly.

My question is what kind of statistics in my analysis should I bother looking at while I put this thing through its paces? Don't need to look at outliers because all the results will be within set parameters. Standard deviation will be nice to see; I'll be looking at it for some wander to indicate possible problems with the RNG. Basic distribution of course. Any other suggestions?


Automatically Appended Next Post:
Also, if anyone has any suggestions on a particular app they use, I'm all ears. Mostly where I see the potential for time being sucked up by physically rolling large numbers of dice is in counting up your distribution. Something that makes the rolls and shows how many of each number came up would be the best, other functionalities would be nice but secondary.

Here is a standard "goodness of fit" test. (Full statistics test)

Roll the dice however many times you think you should, count up the number for each number you get.

These are your observed Values.

Calculate what each number should be. (Divide the number of times you rolled the dice by 6)

For each number, Subtract the observed from the predicted. Square it, and then divide by the predicted.

Add resulting number up, until you've done it for all six numbers.

Now, go to the internet and look up the Chi Squared table.

Look up the column labeled 0.95 and the row labeled 5.

Compare the number you calculated to the number retrieved from the table.

If the number you calculated is LARGER, then conclude that the app is not balanced.*


*As with all statistics, the more dice you roll, the more accurate this test is.

I'd be amazed if someone could write a a dice app which *doesn't* produce evenly distributed results. rand()%6+1 should work in most languages, or rand()*6+1 if rand() is between 0-1, or rand(1,6)...

Trasvi wrote:I'd be amazed if someone could write a a dice app which *doesn't* produce evenly distributed results. rand()%6+1 should work in most languages, or rand()*6+1 if rand() is between 0-1, or rand(1,6)...

I think you will find that the base "random" command is completely rubbish regardless of language.

Not entirely.

It's true that computers can't produce completely random numbers since they work by deterministic computation.

A suitable algorithm, however, can produce numbers that are indistinguishable from random for the purposes of simulating dice. In other words, if the people using the results cannot see the pattern, it might as well not be there.

This degree of unpredictability can be tested.