A087289
a(n) = 2^(2*n+1) + 1.
Original entry on oeis.org
3, 9, 33, 129, 513, 2049, 8193, 32769, 131073, 524289, 2097153, 8388609, 33554433, 134217729, 536870913, 2147483649, 8589934593, 34359738369, 137438953473, 549755813889, 2199023255553, 8796093022209, 35184372088833, 140737488355329, 562949953421313, 2251799813685249
Offset: 0
a(0) = 3 since there are three pairs, (0,1), (1,0) and (1,1) of polynomials (f,g) in GF(2)[x] of degree at most 0 such that gcd(f,g) = 1.
-
[2^(2*n+1) + 1: n in [0..30]]; // Vincenzo Librandi, May 16 2011
-
Table[2^(2 n + 1) + 1, {n, 0, 20}] (* or *) 3 NestList[4 # - 1 &, 1, 20]
(* or *) CoefficientList[Series[(3 - 6 x)/((1 - x) (1 - 4 x)), {x, 0, 20}], x] (* Michael De Vlieger, Mar 03 2017 *)
-
a(n)=2^(2*n+1)+1 \\ Charles R Greathouse IV, Sep 24 2015
A207262
a(n) = 2^(4n - 2) + 1.
Original entry on oeis.org
5, 65, 1025, 16385, 262145, 4194305, 67108865, 1073741825, 17179869185, 274877906945, 4398046511105, 70368744177665, 1125899906842625, 18014398509481985, 288230376151711745, 4611686018427387905, 73786976294838206465, 1180591620717411303425, 18889465931478580854785, 302231454903657293676545
Offset: 1
- David Wells, Prime Numbers: The Most Mysterious Figures in Math. Hoboken, New Jersey: John Wiley & Sons (2005) p. 15
- Vincenzo Librandi, Table of n, a(n) for n = 1..200
- FactorDB, Factorizations of 2^(4*n-2)+1
- P. H. Fuss, Correspondance math. et physique, 1 (1843) p. 145.
- Primenumbers Yahoo Group, Aurifeuille and factoring, search results.
- Eric Weisstein's World of Mathematics, Aurifeuillean Factorization.
- Yahoo Groups, Aurifeuille and factoring
- Index entries for linear recurrences with constant coefficients, signature (17,-16).
A199560
a(n) = (3*9^n + 1)/2.
Original entry on oeis.org
2, 14, 122, 1094, 9842, 88574, 797162, 7174454, 64570082, 581130734, 5230176602, 47071589414, 423644304722, 3812798742494, 34315188682442, 308836698141974, 2779530283277762, 25015772549499854, 225141952945498682, 2026277576509488134, 18236498188585393202
Offset: 0
-
[(3*9^n+1)/2: n in [0..30]];
-
LinearRecurrence[{10,-9},{2,14},30] (* or *) NestList[9#-4&,2,30] (* Harvey P. Dale, May 30 2012 *)
A336913
Image of n under the 3^x+1 map, which is a variation of the 3x+1 (Collatz) map.
Original entry on oeis.org
4, 1, 28, 2, 244, 2, 2188, 3, 19684, 3, 177148, 3, 1594324, 3, 14348908, 4, 129140164, 4, 1162261468, 4, 10460353204, 4, 94143178828, 4, 847288609444, 4, 7625597484988, 4, 68630377364884, 4, 617673396283948, 5, 5559060566555524, 5, 50031545098999708, 5
Offset: 1
For n = 5, a(5) = 3^5+1 = 244, because 5 is odd.
For n = 6, a(6) = floor(log_2(6)) = 2, because 6 is even.
Cf.
A006370 (image of n under the 3x+1 map).
Cf.
A336914 (gives number of steps to reach 1).
-
from math import floor, log
def a(n): return 3**n + 1 if n % 2 else int(floor(log(n, 2)))
print([a(n) for n in range(1, 51)])
-
'''
Program that confirms that 3^x+1 trajectories end with 1.
We avoid the expensive 3^n+1 calculation based on the following:
- 3^n is not a power of two (for n >= 1).
- 3^n+1 is not a power of two (for n > 1) because of the Catalan Conjecture, which was proven in 2002.
- Thus, floor(log2(3^n+1)) == floor(log2(3^n)) == floor(n*log2(3)) for n > 1.
Thanks to Clark R. Lyons for this optimization.
'''
from math import floor, log
log2_of_3 = log(3, 2) # 16 digits after the decimal point.
max_n = 10**15 / 2 # Larger values multiplied by log2_of_3 may have rounding errors.
def check_trajectory(n):
while n > 1:
if n % 2 == 0:
n = int(floor(log(n, 2)))
else:
if n > max_n:
raise ValueError(str(n) + " is too large to be multiplied by log2_of_3")
n = int(floor(n * log2_of_3))
n = 1
while n <= 1000000000:
check_trajectory(n)
n += 1
Showing 1-4 of 4 results.
Comments