2 ms·
If you change the multiplier from 3 to 5 or 7, then you get sequences which diverge to infinity, whereas your argument would apply equally well, suggesting that
by jcranmer 7y ago
If you change the multiplier from 3 to 5 or 7, then you get sequences which diverge to infinity, whereas your argument would apply equally well, suggesting that the intuition you're leveraging is wrong.
A better intuition that usually comes up is to turn the problem into steps of (kn + 1)/2^a, and ask is the next number larger or smaller. For k = 3, the resulting map is usually smaller, but for k > 5, it is usually larger, so rather than orbiting, numbers are more likely to spiral off to infinity.