Collatz conjecture Steps pattern

[SeaN0X]

Hello, I found a weird pattern and i don't know if it's some software error or something.
I found it while I was typing some random numbers to check their steps in the formula.
But with this one, things were different.


Check out this number:

9182818428198211111111111


if I replace any number 1 on the left part by any other number, the number of steps is always 565.

I really don't know what is happening here. I've tested the number 9182818428198211111111111, replacing any number 1 on the left part by any other number
on a lot of sites and even created my own software to compute it, and the result is always 565 steps.

If you don't belive me, check yourself:
https://www.omnicalculator.com/math/collatz-conjecture

[dodomaze]

You need to be careful with software written in Javascript, where all numbers are represented as floating-point. The moment you use a number that big, you are only taking into account the most significant digits and losing the least significant ones, hence your result. Try Python next time!

A counterexample could be 9182818428198211111111113, which takes 807 steps.