Yep, the sum of six Math.random() will start to look a bit like a normal distribution, while Math.random()*6 will stay flat.
Even crazier: the sum of six (Math.random()>.5)?0:1 will also look a bit like a normal distribution, rather than just 0 50% of the time and 6 50% of the time.
Central Limit Theorem eats distributions of all shapes and sizes and shits gaussians. That's why you see normal distributions everywhere.
Comments
Looks like a way to purposefully create a biased (center-weighted) random number generator out of multiple calls.
Mathematically is the sum of six Math.random() any different than (Math.random() * 6)?
Yep, the sum of six Math.random() will start to look a bit like a normal distribution, while Math.random()*6 will stay flat.
Even crazier: the sum of six (Math.random()>.5)?0:1 will also look a bit like a normal distribution, rather than just 0 50% of the time and 6 50% of the time.
Central Limit Theorem eats distributions of all shapes and sizes and shits gaussians. That's why you see normal distributions everywhere.
Ohh okay. I see it now.
This is like how two D6 dice roll differently than a D12?
Yeah, it will biased towards the center of the range rather than uniformly distributed
https://en.wikipedia.org/wiki/Irwin–Hall_distribution