Aren't There Any Formulas for Prime Numbers?

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
7 replies · 2K views
bagasme
Messages
79
Reaction score
9
Hello all,

We know that following formulas failed to produced all prime numbers for any given whole number ##n##:

  1. ##f(n) = n^2 - n - 41##, failed for ##n = 41~(f = 1681)##
  2. ##g(n) = 2^(2^n) + 1##, failed for ##n = 5~(g = 4,294,967,297)##
  3. ##m(n) = 2^n - 1##, failed for ##n = 67~(m = 147,573,952,588,676,412,927)##
The question is: Are there any formulas that produce prime numbers for any given ##n## (without non-prime results), or aren't they?

Bagas
 
Mathematics news on Phys.org
bagasme said:
Hello all,

We know that following formulas failed to produced all prime numbers for any given whole number ##n##:

  1. ##f(n) = n^2 - n - 41##, failed for ##n = 41~(f = 1681)##
  2. ##g(n) = 2^(2^n) + 1##, failed for ##n = 5~(g = 4,294,967,297)##
  3. ##m(n) = 2^n - 1##, failed for ##n = 67~(m = 147,573,952,588,676,412,927)##
The question is: Are there any formulas that produce prime numbers for any given ##n## (without non-prime results), or aren't they?

Bagas

https://en.wikipedia.org/wiki/Formula_for_primes
 
  • Like
Likes   Reactions: Klystron, kith and fresh_42
bagasme said:
I see the floor function recursivemethod interested.
I'm not sure what this means.
bagasme said:
In that case, why ##f_1## need to be irrational?
Look at the denominators of the terms of the series expansion. What do you see?

Also, look again at the derivation of ##f_1##. Do you think this is really an interesting method of generating primes, or just a method of encoding primes that are already known?

I provide a new constant ##P_{buk} = 0.203005000700011...##. Derivation of the related prime generating formula is left to the reader.
 
  • Love
Likes   Reactions: jbriggs444
There is no useful formula for primes known, i.e. a formula that could be used to find new prime numbers more efficiently than clever trial and error.
 
Last edited:
  • Like
Likes   Reactions: Janosh89
It's always the current status of knowledge, a future proof of a nonexistence of such a formula would be really remarkable, but I added it explicitly for clarity.
 
Ignoring computational complexity for the sake of discussion, use of various 'sieves' seems to tell us that the interval between successive prime numbers should increase as we employ successively larger divisors. Then we encounter twin primes where the next higher odd number is also prime. Primes have fascinated me since I learned multiplication.

As pbuk and other posters have described, we do not so much generate prime numbers as discover them; like a miner finding a gold nugget in the next pan of gravel. As a student I attempted to compare intervals among prime numbers to square numbers, triangle, Catalan, and other sequences with instructive but inconclusive results.