| |
BGonline.org Forums
Brett Meyer Responds to Dice Tube Posts
Posted By: Tom Keith In Response To: Brett Meyer Responds to Dice Tube Posts (Brett Meyer)
Date: Friday, 10 September 2010, at 12:47 p.m.
Hi Brett,
I have never used a dice tube, but the idea sounds great and I admire you for taking the time and effort to develop it. I have no problem believing it produces random numbers, however you wrote ...
In January, 2009, my wife and I spent several days recording data from a trial of 3,000 flips using a tube WITH NO RODS containing two 9/16" precision dice. Even with NO rods installed, randomness was achieved! (Doubles resulted 16.63% with an average pip of 8.171)
This is more a test of how well the dice perform rather than how well the tube performs. If you use good dice, you will get an even distribution of numbers.
More interesting would be to know if there is a correlation between successive rolls. Imagine a tube that isn't good at randomization (maybe it's only one inch long instead of nine inches). Suppose the die at the bottom of the tube shows a 6 and you turn the tube over. If the die falls straight down without tumbling either during the fall or after it lands, then you end up with a 1 after your "toss". Similarly, starting with a 5 gives a 2 and starting with a 4 gives a 3, etc.
Could this happen with a real tube? Probably not. My guess is there would be no measurable correlation between successive numbers, but it is the sort thing people will be concerned about. You want to know that future numbers are independent of past numbers.
Tom
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.