6

In January this year the biggest prime number so far has been found and it is a Mersenne Prime. $$2^{57885161}-1$$.

My question: Are all prime numbers from 0 up to $2^{57885161}-1$ found, or not?
If you only look for prime numbers which are Mersenne Primes, you skip quite a few.

Stefan4024
  • 35,843
  • 5
    No. In fact the number of prime numbers up to the largest found so far, by the prime number theorem, exceeds by a very large factor the number of estimated atoms in the universe. It would be prohibitive to enumerate them. – A. Webb Oct 07 '13 at 15:01
  • The testing for Mersenne primes is especially efficient, so there are certainly many unpublished primes below the largest known Mersenne prime. It's a bit insubstantial to try and pin down the range in which all primes have been "found", but there are not enough books to hold all that exist below the largest Mersenne prime you mention. – hardmath Oct 07 '13 at 15:02
  • At first I was thinking Duh. The fact that they were just testing Mersenne numbers was so obvious that I didn't think about it. Now I see that it's a real question so I upvoted it. I realize it's not so obvious that it's so much easier to test a Mersenne number with prime base than it is to test all the other numbers of similar size. – Timothy Feb 26 '20 at 02:41

1 Answers1

6

As you may know, GIMPS is conducting a search. They give each user a different prime number to test. Then they give that number to a different user to check.
They only test numbers of the form $2^p-1$, where $p$ is also prime. So they have only checked a few million possible values of $p$, that is a few million possible values of $2^p-1$.

The link www.mersenne.org/report_milestones/ shows that they have checked all primes $2^p-1$ if $p$ is below 44,576,437, and have double-checked primes $2^p-1$ for values of $p$ up to $p=26,187,517$.

There are around $(2^{57885161})/57885161$, or around $2^{57885135}$ unknown primes below the current record.

Empy2
  • 50,853
  • I think your denominator should be adjusted for taking a natural logarithm, rather than log base 2. – hardmath Oct 07 '13 at 15:16
  • 1
    +1 but closer to $2^{57885136}$, actually (natural log versus log base 2). – Erick Wong Oct 07 '13 at 15:16
  • @Michael Yes, I am familiar with GIMPS (thanks to Numberphile) Thanks, I will have a look at their website. –  Oct 07 '13 at 15:28
  • Indeed, the exponent is actually 57885135.742... – Charles Oct 24 '13 at 16:45
  • GIMPS just finished checking all the exponents up to M57885161 as of two days ago. So now the 48th known Mersenne can just be called the 48th Mersenne prime. :-) – Danny Oct 09 '21 at 01:46