Cool. Guess you finally admit this is also how you wrote that code.

I have like 99 basic principles regarding the theory of systems, observability, controllability, that easily counterfeit that non-sense. None of it written by an AI.

But there is no point in bothering. You wouldn't understand a word anyway.

Just like you don't understand that testing a fixed model yields fixed results. What you have proven is, that for the distribution of hashes of numbers from 1 to 100.000, the prefix theory is better.

However, this does not prove anything, I repeat, anything - as in it has ZERO VALUE when it comes to whichever other distribution.

Because it does not use a uniform distribution as the thing to be tested. It is using a fixed distribution.

Same distribution = same model = same results. I know this is tough to swallow for you, but do we need to call Bernstein on speed-dial to confirm, or let's just shit another 10 pages to explain why again and again?