Supergolden ratio

From HandWiki
Short description: Number, approximately 1.46557

Template:Infobox non-integer number In mathematics, the supergolden ratio is a geometrical proportion, given by the unique real solution of the equation x3 = x2 + 1. Its decimal expansion begins with 1.465571231876768... (sequence A092526 in the OEIS).

The name supergolden ratio is by analogy with the golden ratio, the positive solution of the equation x2 = x + 1.

Definition

ψ = ⁠a/b⁠ = ⁠a+c/a⁠ = ⁠b/c⁠. For b = 1 the boxes have volumes ψ3 = ψ2 (red) + 1 (green).

Three quantities a > b > c > 0 are in the supergolden ratio if a+ca=ab=bc This common ratio is commonly denoted ψ.

Substituting b=ψc and a=ψb=ψ2c in the first fraction, ψ=c(ψ2+1)ψ2c. It follows that the supergolden ratio is the unique real solution of the cubic equation ψ3−ψ2−1=0.

The minimal polynomial for the reciprocal root is the depressed cubic x3+x−1,[1] thus the simplest solution with Cardano's formula, w1,2=(1±13313)/21/ψ=w13+w23

or, using the hyperbolic sine, 1/ψ=23sinh⁡(13arsinh⁡(332)).

A triangle with side lengths ψ, 1, and 1 ∕ ψ has an angle of exactly 120 degrees.[2]

1/ψ is the superstable fixed point of the Newton's method iteration x←(2x3+1)/(3x2+1).

Dividing the defining trinomial x3−x2−1 by x−ψ one obtains x2+x/ψ2+1/ψ, and the conjugate elements of ψ are x1,2=(−1±i4ψ2+3)/2ψ2, with x1+x2=1−ψ and x1x2=1/ψ.

The iteration x←|1+x23 results in the continued radical [lower-alpha 1] ψ3−1=ψ2=1+1+|1+⋯3/23/23/2

Better convergence is attained by using relation y3=y(1−2y)+3 with real zero ψ2−1. Divide both sides by y, and substitute ψ−3=1/(y+2) for 3/y−2y. This gives the iteration y←1+12+y, and the continued reciprocal square root ψ2−1=1+12+1+12+1+12+⋱

Properties

Rectangles in aspect ratios ψ, ψ2 and ψ3 (from left to right) tile the square.

The supergolden ratio can be expressed in terms of itself in many different ways, for example as fractions 2ψ−1=ψ2+2ψ2=ψ3+1ψ3−1=2ψ7−1ψ7, along with 1+ψ−1=ψ2+2ψ+1=2ψ3−1ψ31+ψ−3=ψ2+2ψ2+11+ψ−5=ψ+1ψ2=ψ2−1=ψ7+1ψ7−1ψ4−1=ψ2+2ψ2−1ψ6−1=ψ2+2ψ−1.

Similarly as infinite geometric series ψ3=∑n=0∞ψ−nψ2ψ2−1=∑n=0∞ψ−2nψ=∑n=0∞ψ−3nψψ2−1=∑n=0∞ψ−4n13−ψ2=∑n=0∞ψ−5n1ψ2−1+1ψ2+2=∑n=0∞ψ−6nψ22=∑n=0∞ψ−7n. Additionally, ∑n=07ψ−n=3.

For every integer n one has ψn=ψn−1+ψn−3=ψn−2+ψn−3+ψn−4=ψn−2+2ψn−4+ψn−6 from this an infinite number of further relations can be found.

Argument θ=arcsec⁡(2ψ4) satisfies the identity tan⁡(θ)−4sin⁡(θ)=33.[3]

Continued fraction pattern of a few low powers ψ−1=[0;1,2,6,1,3,5,4,22,...]≈0.6823(1319)ψ0=[1]ψ1=[1;2,6,1,3,5,4,22,1,...]≈1.4656(2215)ψ2=[2;6,1,3,5,4,22,1,1,...]≈2.1479(157)ψ3=[3;6,1,3,5,4,22,1,1,...]≈3.1479(227)ψ4=[4;1,1,1,1,2,2,1,2,2,...]≈4.6135(6013)ψ5=[6;1,3,5,4,22,1,1,4,...]≈6.7614(11517)

Notably, the continued fraction of ψ2 begins as permutation of the first six natural numbers; the next term is equal to their sum + 1.

As derived from its continued fraction expansion, the simplest rational approximations of ψ are: 32,1913,2215,8558,277189,447305,18731278,4165328421,4352629699,8517958120,…

Newton's method for p(z) = z3 − z2 − 1: ψ (right) and its complex conjugates at the nuclei of their basins of attraction. Julia set of the Newton map in orange, with unit circle and real curve for reference.

The supergolden ratio is the fourth smallest Pisot number.[4] By definition of these numbers, the absolute value 1/ψ of the algebraic conjugates is smaller than 1, thus powers of ψ generate almost integers. For example: ψ11=67.000222765... ≈67+1/4489. After eleven rotation steps the phases of the inward spiraling conjugate pair – initially close to ±13π/22 – nearly align with the imaginary axis.

The minimal polynomial of the supergolden ratio m(x)=x3−x2−1 has discriminant Δ=−31. The Hilbert class field of imaginary quadratic field K=ℚ(Δ) can be formed by adjoining ψ. With argument τ=(1+Δ)/2 a generator for the ring of integers of K, one has the special value of Dedekind eta quotient ψ=eπi/24η(τ)2η(2τ).

Expressed in terms of the Weber-Ramanujan class invariant Gn [lower-alpha 2] ψ=𝔣(Δ)2=G3124.

Properties of the related Klein j-invariant j(τ) result in near identity eπ−Δ≈(2ψ)24−24. The difference is < 1/143092.

The elliptic integral singular value [5] kr=λ*(r) for r=31 has closed form expression λ*(31)=sin⁡(arcsin⁡((24ψ)−12)/2) (which is less than 1/10 the eccentricity of the orbit of Venus).

Narayana sequence

A Rauzy fractal associated with the supergolden ratio-cubed. The central tile and its three subtiles have areas in the ratios ψ4 : ψ2 : ψ : 1.
A Rauzy fractal associated with the supergolden ratio-squared, with areas as above.

Narayana's cows is a recurrence sequence originating from a problem proposed by the 14th century Indian mathematician Narayana Pandita.[6] He asked for the number of cows and calves in a herd after 20 years, beginning with one cow in the first year, where each cow gives birth to one calf each year from the age of three onwards.

The Narayana sequence has a close connection to the Fibonacci and Padovan sequences and plays an important role in data coding, cryptography and combinatorics. The number of compositions of n into parts 1 and 3 is counted by the nth Narayana number.

The Narayana sequence is defined by the third-order recurrence relation Nn=Nn−1+Nn−3 for n>2, with initial values N0=N1=N2=1.

The first few terms are 1, 1, 1, 2, 3, 4, 6, 9, 13, 19, 28, 41, 60, 88,... (sequence A000930 in the OEIS). The limit ratio between consecutive terms is the supergolden ratio:limn→∞Nn+1/Nn=ψ.

The first 11 indices n for which Nn is prime are n = 3, 4, 8, 9, 11, 16, 21, 25, 81, 6241, 25747 (sequence A170954 in the OEIS). The last number has 4274 decimal digits.

The sequence is extended to negative indices using Nn=Nn+3−Nn+2, obtaining (1), 0, 0, 1, 0,−1, 1, 1,−2, 0, 3,−2,−3, 5,...

Powers of the supergolden ratio can be written with Narayana numbers as quadratic coefficients ψn=ψ2Nn−2+ψNn−4+Nn−3, which is proved by mathematical induction on n. This relation also holds for n<0. The order of the coefficients corresponds to the bottom row of matrix Q below.

The generating function of the Narayana sequence for non-negative n is given by 11−x−x3=∑n=0∞Nnxn for x<1ψ

The Narayana numbers are related to sums of binomial coefficients by Nn=∑k=0⌊n/3⌋(n−2kk)

The characteristic equation of the recurrence is x3−x2−1=0. If the three solutions are real root ψ and conjugate pair β and γ, the Narayana numbers are given by the Binet formula [7] Nn−1=aψn+bβn+cγn, with real a and conjugates b and c the roots of 31x3−3x−1=0.

Since |bβn+cγn|<ψ−n/2, the number Nn is the nearest integer to aψn+1, with n≥0 and coefficient a=ψ2/(ψ3+2)= 0.4172379879262187776214755...

Coefficients a=b=c=1 result in the Binet formula for the related sequence An=Nn+2Nn−3.

The first few terms are 3, 1, 1, 4, 5, 6, 10, 15, 21, 31, 46, 67, 98, 144,... (sequence A001609 in the OEIS).

This anonymous sequence has the Fermat property: if p is prime, Ap≡A1modp. The converse does not hold, but the small number of odd pseudoprimes n∣(An−1) makes the sequence special.[8] The 8 odd composite numbers below 108 to pass the test are n = 1155, 552599, 2722611, 4822081, 10479787, 10620331, 16910355, 66342673.

A supergolden Rauzy fractal of type a ↦ ab, with areas as above. The fractal boundary has box-counting dimension 1.50

The Narayana numbers are obtained as integral powers n > 3 of a matrix with real eigenvalue ψ [6] Q=(101100010),

Qn=(NnNn−2Nn−1Nn−1Nn−3Nn−2Nn−2Nn−4Nn−3)

The trace of Qn gives the above An.

Alternatively, Q can be interpreted as incidence matrix for a D0L Lindenmayer system on the alphabet {a,b,c} with corresponding substitution rule {a↦abb↦cc↦a and initiator w0=b. The series of words wn produced by iterating the substitution have lengths l(wn)=Nn. The number of c, b and a's in each word is equal to successive Narayana numbers. Associated to this string rewriting process is a compact set composed of self-similar tiles called the Rauzy fractal, that visualizes the combinatorial information contained in a multiple-generation three-letter sequence.[9]

The ratio between the number of digits in successive strings for the binary version of the look-and-say sequence converges to ψ.[10][11]

Geometry

Psi-quad substitution tiling

Psi-quad substitution tiling.
Supergolden fractal triangle.

In the complex plane, if ψ is a non-real root of ψ3−ψ2−1=0, then the vertices of a self-similar quadrilateral tiling (the psi-quad substitution tiling) can be taken as the six points ψ0,ψ1,ψ2,ψ3,ψ4,−ψ2. The iterated boundary is the supergolden fractal triangle, which has Hausdorff dimension d∂=2log⁡ζlog⁡ψ, where ζ is the real root of ζ7−2ζ2−1=0, so numerically d∂≈1.0295240599.

Supergolden rectangle

Nested supergolden rectangles with perpendicular diagonals and side lengths in powers of ψ.

A supergolden rectangle is a rectangle whose side lengths are in a ψ:1 ratio. Compared to the golden rectangle, the supergolden rectangle has one more degree of self-similarity.

Given a rectangle of height 1, length ψ and diagonal length ψ3 (according to 1+ψ2=ψ3). The triangles on the diagonal have altitudes 1/ψ; each perpendicular foot divides the diagonal in ratio ψ2.

On the left-hand side, cut off a square of side length 1 and mark the intersection with the falling diagonal. The remaining rectangle now has aspect ratio ψ2:1 (according to ψ−1=ψ−2). Divide the original rectangle into four parts by a second, horizontal cut passing through the intersection point.[12][13]

The rectangle below the diagonal has aspect ratio ψ3, the other three are all supergolden rectangles, with a fourth one between the feet of the altitudes. The parent rectangle and the four scaled copies have linear sizes in the ratios ψ3:ψ2:ψ:ψ2−1:1. It follows from the theorem of the gnomon that the areas of the two rectangles opposite the diagonal are equal.

In the supergolden rectangle above the diagonal, the process is repeated at a scale of 1:ψ2.

Supergolden spiral

Supergolden spirals with different initial radii on a ψ− rectangle.

A supergolden spiral is a logarithmic spiral that gets wider by a factor of ψ for every quarter turn. It is described by the polar equation r(θ)=aexp⁡(kθ), with initial radius a and parameter k=2πln⁡(ψ). If drawn on a supergolden rectangle, the spiral has its pole at the foot of altitude of a triangle on the diagonal and passes through vertices of rectangles with aspect ratio ψ which are perpendicularly aligned and successively scaled by a factor ψ−1.


See also

Solutions of equations similar to x3=x2+1:

Notes

  1. ↑ m/n√x = xn/m
  2. ↑ German Wikipedia has a table of analytical values of the Ramanujan G-function (de) for odd arguments below 47.

References

  1. ↑ (sequence A263719 in the OEIS)
  2. ↑ (sequence A092526 in the OEIS)
  3. ↑ Piezas III, Tito (Dec 18, 2022). "On the tribonacci constant with cos(2πk/11), plastic constant with cos(2πk/23), and others". https://math.stackexchange.com/questions/4600807/. 
  4. ↑ Panju, Maysum (2011). "A systematic construction of almost integers". The Waterloo Mathematics Review 1 (2): 35–43. https://mathreview.uwaterloo.ca/archive/voli/2/panju.pdf. 
  5. ↑ Weisstein, Eric W.. "Elliptic integral singular value". http://mathworld.wolfram.com/EllipticIntegralSingularValue.html. 
  6. ↑ 6.0 6.1 (sequence A000930 in the OEIS)
  7. ↑ Lin, Xin (2021). "On the recurrence properties of Narayana's cows sequence" (in en). Symmetry 13 (1): 1–12. doi:10.3390/sym13010149. Bibcode: 2021Symm...13..149L. 
  8. ↑ Studied together with the Perrin sequence in: Adams, William; Shanks, Daniel (1982). "Strong primality tests that are not sufficient". Math. Comp. (AMS) 39 (159): 255–300. doi:10.2307/2007637. 
  9. ↑ Siegel, Anne; Thuswaldner, Jörg M. (2009). "Topological properties of Rauzy fractals". Mémoires de la Société Mathématique de France. 2 118: 1–140. doi:10.24033/msmf.430. http://numdam.org/item/MSMF_2009_2_118__1_0/. 
  10. ↑ (sequence A049194 in the OEIS)
  11. ↑ Sillke, Torsten (April 7, 1993). "The binary form of Conway's sequence". Faculty of Mathematics: Universität Bielefeld. https://www.math.uni-bielefeld.de/~sillke/SEQUENCES/series001. 
  12. ↑ Crilly, Tony (1994). "A supergolden rectangle" (in en). The Mathematical Gazette 78 (483): 320–325. doi:10.2307/3620208. 
  13. ↑ Koshy, Thomas (2017) (in en). Fibonacci and Lucas numbers with applications (2 ed.). John Wiley & Sons. doi:10.1002/9781118033067. ISBN 978-0-471-39969-8.