login
The OEIS is supported by the many generous donors to the OEIS Foundation.

 

Logo
Hints
(Greetings from The On-Line Encyclopedia of Integer Sequences!)
A342000 a(1) = 0, and for n > 1, a(n) = 1 if G(A329886(n)) >= G(A329886(floor(n/2))), otherwise 0, where G(n) = sigma(n) / (n*log(log(n))), where sigma is the sum of the divisors. 6
0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0 (list; graph; refs; listen; history; text; internal format)
OFFSET
1
COMMENTS
Equivalently, for n > 3, a(n) = 1 if the value of H(A329886(n)) <= H(A329886(n\2)), where H(n) = log(n)^(n/sigma(n)) [= exp(1/G(n))], otherwise 0.
The ratio G(n) = sigma(n) / (n*log(log(n))) comes from Grönwall's theorem (listed, for example, as Theorem 1 in the Caveney, Nicolas and Sondow paper; see also the other papers linked at A073751). This sequence gives the positions of those points in the A329886-tree (Primorial inflation of Doudna-tree) where this ratio doesn't decrease when going downwards (which it mostly does, but see also A342455 for a counterexample). There are 11355 such increasing cases among the first 65536 terms, and 113134 among the first 2^20, although overall the ratio of such cases on each row k (that has 2^k terms) seems to start decreasing after the seventh row. See A342020.
It seems that the tree has infinitely long leftward branches that contain either only zeros or only ones after a while: The leftmost edge (that are primorials, A002110, in A329886) appears to consist of zeros only after its single 1 at the second term. This depends on the (so far conjectural) observation that (log(A002110(n)) + log(prime(1+n)))^(prime(1+n)) > log(A002110(n))^(1+prime(1+n)) for all n >= 1.
On the other hand, the leftward branch starting from the left child of the 496th term of the tree appears to contain only ones (corresponding to A342455 from its fifth term onward, see comments there).
Note that as A329886 is a permutation of A025487, it contains also all terms of A004394 and A004490 in it (see e.g., A342013).
LINKS
G. Caveney, J.-L. Nicolas, and J. Sondow, Robin's theorem, primes, and a new elementary reformulation of the Riemann Hypothesis, Integers 11 (2011), #A33.
Michael De Vlieger, Tree illustrated down to level 10. Red indicates 1's in this sequence, and black indicates 0's. The numbers shown are those in A329886.
Michael De Vlieger, Chart of levels 0 <= j <= 16, vertically exaggerated 256X.
Michael De Vlieger, 1024-pixel square raster showing 2^20 terms, where black = 1 and white = 0.
J. C. Lagarias, An elementary problem equivalent to the Riemann hypothesis, arXiv:math/0008177 [math.NT], 2000-2001; Am. Math. Monthly 109 (#6, 2002), 534-543.
EXAMPLE
The binary tree begins as:
0
..................../ \.................
1 1
0......./ \........ 0 0 ......./ \........0
/ \ / \ / \ / \
/ \ / \ / \ / \
/ \ / \ / \ / \
/ \ / \ / \ / \
0 1 0 0 0 0 0 0
/\ /\ /\ /\ /\ /\ /\ /\
/ \ / \ / \ / \ / \ / \ / \ / \
0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0
0 1 0 1 0 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0
with 1 marked at each node of A329886 (Primorial inflation of Doudna-tree) if its G(n) = sigma(n) / (n*log(log(n))) ratio is greater than or equal to the corresponding ratio of its parent, and 0 otherwise.
For example, A329886(1) = 2, and A329886(3) = 4. The latter has ratio G(4) = 5.357674..., while for the former (the parent of 4), the ratio is G(2) = -4.0926..., which is less than 5.357674..., therefore a(3) = 1.
For 130 = 2*65, we have A329886(65) = 60060, A329886(130) = 3063060 = A283980(60060). Here G(3063060) = 1.594960... > 1.56762... = G(60060), in other words, here again the child has a larger G-ratio than its parent, and this is the first case where it is a child obtained with A283980 instead of doubling, thus n=130 is also the first such even number after 2 for which a(n) = 1.
MATHEMATICA
Block[{a, b, c, f, nn = 105}, b[0] = c[1] = 1; f[n_] := DivisorSigma[1, n]/(n Log[Log[n]]); Do[b[i] = Prime[1 + BitLength[i] - DigitCount[i, 2, 1]]*b[i - 2^Floor@ Log2@ i]; c[i + 1] = Apply[Times, Flatten@ MapIndexed[ConstantArray[Prime[First[#2]], #1] &, Table[LengthWhile[#1, # >= j &], {j, #2}] & @@ {#, Max[#]} &@ Sort[Flatten[ConstantArray[PrimePi@ #1, #2] & @@@ FactorInteger[b[i]]], Greater]]]; a[i - 1] = Boole[f[c[i]] >= f[c[Floor[(i + 1)/2]]]], {i, nn}]; Array[a, nn - 1]] (* Michael De Vlieger, Mar 07 2021 *)
PROG
(PARI)
default(realprecision, 10001);
G(n) = (sigma(n) / (n*log(log(n))));
A283980(n) = {my(f=factor(n)); prod(i=1, #f~, my(p=f[i, 1], e=f[i, 2]); if(p==2, 6, nextprime(p+1))^e)}; \\ From A283980
A329886(n) = if(n<2, 1+n, if(!(n%2), A283980(A329886(n/2)), 2*A329886(n\2)));
A342000(n) = if(1==n, 0, my(p=A329886(n\2)); if(n%2, G(2*p)>=G(p), G(A283980(p))>=G(p)));
(PARI)
\\ Alternative program, probably less vulnerable to the loss of precision:
Gie(n) = (log(n)^(n/sigma(n))); \\ = exp(1/G(n)), function H in comments
A342000(n) = if(n<=3, !!(n-1), my(p=A329886(n\2)); if(n%2, Gie(2*p)<=Gie(p), Gie(A283980(p))<=Gie(p)));
CROSSREFS
Cf. A342020 (row sums).
Cf. also A336834 for a similarly constructed sequence.
Cf. also A197638.
Sequence in context: A295304 A171386 A189816 * A194685 A182582 A125720
KEYWORD
nonn,tabf
AUTHOR
STATUS
approved

Lookup | Welcome | Wiki | Register | Music | Plot 2 | Demos | Index | Browse | More | WebCam
Contribute new seq. or comment | Format | Style Sheet | Transforms | Superseeker | Recents
The OEIS Community | Maintained by The OEIS Foundation Inc.

License Agreements, Terms of Use, Privacy Policy. .

Last modified April 23 23:26 EDT 2024. Contains 371917 sequences. (Running on oeis4.)