[go: up one dir, main page]

login
Revisions by Dimitris Valianatos (See also Dimitris Valianatos's wiki page)

(Bold, blue-underlined text is an addition; faded, red-underlined text is a deletion.)

Showing entries 1-10 | older changes
E.g.f.: -log( 1 - (exp(x) - 1)^3 / 3! ).
(history; published version)
#38 by Dimitris Valianatos at Mon Jul 19 01:19:07 EDT 2021
STATUS

editing

proposed

Discussion
Mon Jul 19
05:30
Hugo Pfoertner: I suggest submitting the name of this sequence as a comment in A006577, i.e., "2^a(n) is the largest number whose Collatz trajectory ...".  All of the rest strikes me as a verbose repetition of well known properties of the Collatz sequence and my recommendation is to reject this proposal.
08:06
Dimitris Valianatos: Dear Hugo: This sequence that I present is a different one and do not show in OEIS. This definition was my initial thought. What you are saying is a formula, that I included and I know many sequences are generated using another sequence from OEIS. This is repeated many times here.
It is different not to find the sequence interesting, so you can delete it. We as students suggest and you as teachers decide. However, this sequence does not exist in OEIS and connected directly with the Collatz conjecture. It consists of the maximum numbers which reach 1, only halving steps without tripling, in contrast to n.
18:43
Jon E. Schoenfield: I agree with Hugo. :-(
Tue Jul 20
04:11
Hugo Pfoertner: Any votes from other editors to keep the sequence?
#37 by Dimitris Valianatos at Mon Jul 19 01:18:58 EDT 2021
COMMENTS

If a number c satisfies the Collatz conjecture, this means that there exists k >= log_2(c) such that (1) c <= 2^k and each descendant c_i of c verifies satisfies c_i <= 2^(k+i), i -> {1,2,3,...,oo}, and (2) every ancestor c_j of c verifies satisfies c_j <= 2^(k-j), j -> {1,...,k}. Then a(c) = 2^k (see second PARI PROG).

Now, if c does NOT satisfy the Collatz conjecture, this means for EVERY k >= log_2(c) (as k tends to infinity) (1) c <= 2^k and each descendant c_i of c verifies satisfies c_i <= 2^(k+i), i -> {1,2,3,...,oo} but there is a j-th ancestor of c, the number c_j, that is c_j > 2^(k-j), j -> {1,...,k}. Then we can say arbitrarily that a(c) = -1 or a(c) = οο.

STATUS

proposed

editing

#34 by Dimitris Valianatos at Sun Jul 18 17:30:58 EDT 2021
STATUS

editing

proposed

#33 by Dimitris Valianatos at Sun Jul 18 17:30:51 EDT 2021
COMMENTS

2) every ancestor c_j of c verifies c_j <= 2^(k-j), j -> {1,..,k}. Then a(c) = 2^k (see second PARI PROG).

STATUS

proposed

editing

#32 by Dimitris Valianatos at Sun Jul 18 17:10:07 EDT 2021
STATUS

editing

proposed

#31 by Dimitris Valianatos at Sun Jul 18 17:09:54 EDT 2021
PROG

(PARI) for(k = 1, 37, n = k; s = 0; while(n > 1, if(n%2 = = 0, n/=2; s++, n = 3*n + 1; s++)); print1(2^s", "))

(PARI) a(nn) = kk = 2^floor(log(nn) / log(2) + .5); n = nn; k = kk; while(n < k, if(n%2 == 0, n/=2, n = 3*n + 1); k/=2; if(n > k, kk*=2; k = kk; n = nn)); print1(kk", ")

STATUS

proposed

editing

#30 by Dimitris Valianatos at Sun Jul 18 16:15:36 EDT 2021
STATUS

editing

proposed

#29 by Dimitris Valianatos at Sun Jul 18 16:15:28 EDT 2021
COMMENTS

As we see in the example the numbers 7, 2 ^ 16 = 65536 = 2 ^ 16 belong to the 16th generation of Collatz numbers, starting from 1. This means that they have a finite number of ancestors 16 and will leave an infinite number of descendants. I find it difficult to understand what would be the finite number c that would not belong to any generation, would leave infinite descendants and would come from infinite ancestors, given that the members of each previous generation are dwindling.

STATUS

proposed

editing

#28 by Dimitris Valianatos at Sun Jul 18 15:43:37 EDT 2021
STATUS

editing

proposed

#27 by Dimitris Valianatos at Sun Jul 18 15:41:52 EDT 2021
COMMENTS

As we see in the example the numbers 7, 2 ^ 16 = 65536 belong to the 16th generation of Collatz numbers, starting from 1. This means that they have a finite number of ancestors 16 and will leave an infinite number of descendants. I find it difficult to understand what would be the finite number c that would not belong to any generation, would leave infinite descendants and would come from infinite ancestors, given that the members of each previous generation are dwindling.