[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Safe Harbor Games Dice

Posted By: W Womack
Date: Monday, 28 September 2009, at 2:20 a.m.

In Response To: Safe Harbor Games Dice (Timothy Chow)

Tim,

I am not sure that my basic understanding of statistics would allow me to get to the depths that would be needed to go a lot deeper. I did mention that I ran a chi squared test on the distribution I got from my data and it is extremely unlikely (p<0.0001) to come from the expect distribution.

In looking at the distribution of non-doubles and comparing the frequency to that of a population generated from random.org it did not look like the frequency of the various non doubles is an issue. I am not sure, however that it is valid to just throw out the expected doubles in the data set, and say that each non double occurs 1/15 of the time. Perhaps someone with a deeper knowledge of the statistics could comment on this.

Granting that my eyeball assessment of non-double roll distribution is correct, and that this distribution matches expectations it would suggest that the problem might not be in the prng but in how the rolls from the prng are transferred into the game environment. It almost looks like some of the doubles being generated by the prng are being discarded rather than played (and a lot more than would come from initial rolls). However, I cannot come up with a reasonable scenario that would explain why something like that would be happening.

Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.