[go: up one dir, main page]

login
Revision History for A253074 (Bold, blue-underlined text is an addition; faded, red-underlined text is a deletion.)

Showing entries 1-10 | older changes
Lexicographically earliest sequence of distinct numbers such that a(n-1)+a(n) is not prime.
(history; published version)
#43 by Joerg Arndt at Tue Nov 26 10:34:57 EST 2019
STATUS

reviewed

approved

#42 by Michel Marcus at Tue Nov 26 03:00:11 EST 2019
STATUS

proposed

reviewed

#41 by M. F. Hasler at Tue Nov 26 00:23:16 EST 2019
STATUS

editing

proposed

Discussion
Tue Nov 26
00:26
M. F. Hasler: [offset was 0, changed to 1 in edit #12 by Reinhard Zumkeller]
#40 by M. F. Hasler at Mon Nov 25 23:44:00 EST 2019
PROG

(PARI) A253074_upto(n=99, a, u, U)={vector(n, n, for(k=u, oo, bittest(U, k-u)|| isprime(a+k)||[a=k, break]); (a>u && U+=1<<(a-u))|| U>>=-u+u+=valuation(U+2, 2); a)+if(default(debug), print([u]))} \\ additional args allow to tweak computation. If debug > 0, print least unused number at the end. - M. F. Hasler, Nov 25 2019

STATUS

proposed

editing

Discussion
Tue Nov 26
00:23
M. F. Hasler: In a strict sense, this must have offset zero to be a permutation (= bijection from a set into itself) of the nonnegative integers.
#39 by M. F. Hasler at Mon Nov 25 20:54:22 EST 2019
STATUS

editing

proposed

#38 by M. F. Hasler at Mon Nov 25 20:48:32 EST 2019
COMMENTS

A simpler Simplified version of the proof: Assume x isn't in the sequence, then eventually all terms must be of the form PRIME - x, else x would appear next. In particular, no multiple of x can appear from there on. Assume k*x is the largest multiple of x in the sequence. Take a prime p not dividing x. Then m*x can't appear in the sequence for k+1 <= m <= k+p, and all terms are eventually of the form PRIME - m*x for all m in {k+1, ..., k+p}. Take one such term N > p, i.e., N + (k+1)*x, ..., N + (k+p)*x are all prime. Consider this sequence mod p. Since gcd(x,p)=1, the p terms cover each residue class mod p, so one is a multiple of p, in contradiction with their primality. - M. F. Hasler, Nov 25 2019

Discussion
Mon Nov 25
20:54
M. F. Hasler: Maybe this "simplified version" can go in place of the "original" version? (e.g., add "simplified by M.F.H" in "[proof outline...]" and replace the (IMHO) more complicated one by my version.)  [PS: I reverted the "(because...)" I had inserted, cf 19:18.]
#37 by M. F. Hasler at Mon Nov 25 20:42:37 EST 2019
COMMENTS

In particular, this means there are only finitely many multiples of x that appear in the sequence (because . Let Y be a multiple of x can't be larger than all multiples of x appearing in the form PRIME - x)sequence.

Let Y be a multiple of x larger than all multiples of x appearing in the sequence.

A simpler proof: Assume x isn't in the sequence, then eventually all terms must be of the form PRIME - x, else x would appear next. In particular, no multiple of x can appear from there on. Assume k*x is the largest multiple of x in the sequence. Take a prime p not dividing x. Then m*x can't appear in the sequence for k+1 <= m <= k+p, and all terms are eventually of the form PRIME - m*x for all m in {k+1, ..., k+p}. Take one such term N > p, i.e., N + (k+1)*x, ..., N + (k+p)*x are all prime. Consider this sequence mod p. Since gcd(x,p)=1, the p terms cover each residue class mod p, so one is a multiple of p, in contradiction with their primality. - M. F. Hasler, Nov 25 2019

STATUS

proposed

editing

#36 by M. F. Hasler at Mon Nov 25 19:18:49 EST 2019
STATUS

editing

proposed

Discussion
Mon Nov 25
19:37
M. F. Hasler: Also, I believe it should not be "prime p" but simply "number p" (which is sufficient for the sequel... and I don't see where the primality of p would come from.)
20:21
M. F. Hasler: I have a simpler proof: Take a prime p not dividing x. Assume kx is the largest multiple of x in the sequence.Then (k+1)x, ..., (k+p)x  all don't appear in the sequence, so all terms are eventually of the form PRIME - (k+m) x  for all m in {1,...,p}.Take one such term N > p, i.e., N + (k+1)x, ..., N + (k+p)x  are all prime.
Consider this sequence mod p. Since gcd(x,p)=1, the p terms cover each residue class mod p, so one is a multiple of p, in contradiction with their primality.
#35 by M. F. Hasler at Mon Nov 25 19:16:48 EST 2019
COMMENTS

In particular, this means there are only finitely many multiples of x that appear in the sequence (because k*a multiple of x = can't be of the form PRIME - x <=> (k+1)*x = PRIME is impossible).

Discussion
Mon Nov 25
19:18
M. F. Hasler: Instead, I inserted "(because ...)" which clarifies why there can't be multiples of x beyond some point.
#34 by M. F. Hasler at Mon Nov 25 19:09:21 EST 2019
COMMENTS

In particular, this means there are only finitely many multiples of x that appear in the sequence. To make this cleaner, let Y be a multiple of (because k*x = PRIME - x larger than all multiples of <=> (k+1)*x appearing in the sequence= PRIME is impossible).

Let Y be a multiple of x larger than all multiples of x appearing in the sequence.

STATUS

approved

editing

Discussion
Mon Nov 25
19:16
M. F. Hasler: Deleted "to make this cleaner": as it stood it says that what follows would clarify what precedes, but the next phrase "Let Y be a multiple of x larger than all multiples of x appearing in the sequence" obviously *uses* "only finitely many multiples of x that appear in the sequence" and does not explain or prove it. I deduce that that is not part of the proof but was an "accompanying message".