Please see the array below. The numbers and their frequencies are exactly the same as A020882.
5;
13, 17;
25, 29, 37;
41, 45, 53, 65;
61, 65, 73, 85, 101;
85, 89, 97, 109, 125, 145;
Comments:
A prime appears only once in this sequence.
A number appears more than once if and only if it’s a product of two or more distinct primes that appear in this sequence.
If R is a product of k distinct primes that appear in this sequence, then R appears at least k times.
Thank you very much, Maximilian. I really appreciate your help here, as always.
This sequence is beautiful. I spent the long weekend studying it and it was worth every second! It contains all 4k+1 primes and connects the positions and repetitions of the sum of two square numbers directly to their prime factorization. Repeated appearances reveal that a number is composite and can even be used to recover its factors. I’m trying now to find a “counterpart” for 4k+3 primes.
Best,
Ali
P.S. My first attempt to respond to this email has failed and I don't know why.
PS: to be precise, the triangle doesn't exist, but it's image/range is A339952.Le mar. 8 sept. 2026 à 12:36, M. F. Hasler <oe...@dsi972.fr> a écrit :I tentatively added, in https://oeis.org/draft/A020882:
Ali Sada observed that the terms are to a large extend the same as in the triangle T(n,k) = 1 + 4(A000217(n)+A000217(k)), n,k > 0, i.e., 1 + 4*(sums of two positive triangular numbers), but e.g., T(4,2) = 45, T(7,1) = 117, T(7,5) = 153, ... never appear in this sequence. - M. F. Hasler, Sep 08 2026
Please feel free to change or add an XREF if you plan to submit that triangle separately.(I checked that it does not yet exist.)- Maximilian