They tend to agree

Markov:  (1, 1, 1), (1, 1, 2), (1, 2, 5), (1, 5, 13), (2, 5, 29), (1, 13, 34)

Orbitals.

	s	p	d	f	g	h
1	1
2	2	3
3	4	5	7
4	6	8	10	13
5	9	11	14	17	21
6	12	15	18	22	26	31
7	16	19	23	27	32	37
8	20	24	28	33	38	44
9	25	29	34	39	45	51
10	30	35	40	46	52	59

First, when the physicists measure proton wobble in the single electron case they have lowered energy until the first three rows show on their wobble calculator.   So we can see their highest magnetic wobble will spin about the unit circle 17 times, from the last element in the Markov. 

iLog(i) is in play in distributing energy through any corner of the graph, and it will not bunch up.  In Quanta, that graph is accurate to the third. These are just hyperbolic surfaces, closed with an opposite magnetic moment.

What is cart and horse?  Well, the fractional approximation rules wins, there are finite approximations in a 3D system. That yields the largest integer and spreading that about the Markov tree would select the first three rows, in sequence by resolution (= energy).  One would discover the uncertainty of counting accuracy past row three is greater than the inaccuracy of N, the total count of integers. And that is the residual error in the fraction approximation theorem.

In the physics of orbitals, maximum entropy means there are more axis i for counting as energy levels increase.  But those units are based on a prior, and damned good, estimate of Avogadro, in essence.  So they have good coefficients relating cloud size and charge count. They are essentially working to spread Plank around both the axis, and the count N, it is a fuzzy constant.  But, in the end the vacuum is a quanta, and everyone discovers the best fraction, both vacuum and physicist, both want the perfect quotient ring.