Originally Posted by GristlyKnuckle
There is no room for opinion when the correct answer is that true randomness does not exist. The more I learn about how randomness is a presumption, the more capable I am in knowing what will happen if I do one thing or the other, and the more confused the rest of people will be about why the dice rolls seem non-random. All randomness is stupid. You're replying to Wofenring. I'm Gristly-Knuckle. Well met.
I still have no idea what point you're trying to make.
I think we're all in the understanding that RNG is not true random. It's just easier to refer to Larian's RNG as such relative to weighted dice. Larian stated they were testing a form of true RNG in early access to see the players' response. Currently we have a weighted dice option and we're are reviewing how random either are. Sure nomenclature was juggled when comparing different forms of RNG, but I'm still confused by your posts.

Originally Posted by mrfuji3
Originally Posted by DragonSnooz
Yes, all the data in that post is for weighted rolls. If weighted dice only function to reduce streaks, then it should become an even distribution faster than Larian's true random.

Here is the chart from patch 3 (unweighted).
Ah. I was of the mindset that weighted dice made it more likely to roll a middling number, not reduce streaks.

Thanks for the unweighted chart. It's Chi^2 is 23.5 which is still below the 95% confidence value of 30.14.
The most extreme values in that chart are 6 (4 times) and 17 (16 times):
The chance of rolling at least one number only 4 times is 3.5%.
The chance of rolling at least one number 16 times is 0.8%.
Neither of these are too incredibly unlikely.

BUT! If we combine your unweighted chart with @Saberem's unweighted data, we get a Chi^2 of 32.26 which is GREATER than the 95% confidence value of 31.14. Thus, we can conclude that Larian's unweighted dice rng is NOT random.
The most extreme values were 6 (7 times), with a ~1.8% chance of happening.
and 17 (rolled 28 times), with a ~0.02% chance of happening. This is pretty unlikely.

Math
Chi^2 calculation is the standard Pearson's chi-squared test, using the table https://www.statology.org/how-to-read-chi-square-distribution-table/
Likelihood P of getting a single number X times calculated using https://www.omnicalculator.com/statistics/dice
We have 20 different values on a d20, so the chance of rolling at least one number X times is 1-(1-P)^20

I definitely prefer weighted dice as RNG, I was getting tired of seeing 5, 10, or 17 xD