Slashdot Mirror


51st Known Mersenne Prime Number Found (mersenne.org)

chalsall (Slashdot reader #185), writes: The Great Internet Mersenne Prime Search (GIMPS) has discovered the largest known prime number, 2^82,589,933-1, having 24,862,048 digits. A computer volunteered by Patrick Laroche from Ocala, Florida made the find on December 7, 2018.

GIMPS has been on amazing lucky streak, finding triple the expected number of new Mersenne primes -- a dozen in the last fifteen years.

"This anomaly is not necessarily evidence that existing theories on the distribution of Mersenne primes is incorrect," notes GIMPS. "However, if the trend continues it may be worth further investigation. " They also report that the newly-discovered prime number "is more than one and a half million digits larger than the previous record prime number" -- and it's one of just 51 known Mersenne prime numbers ever discovered. "GIMPS, founded in 1996, has discovered the last 17..."

Patrick Laroche is one of thousands of volunteers using GIMPS' free software to hunt for prime numbers -- and is now eligible for a $3,000 "research discovery award," the group writes at mersenne.org. "GIMPS' next major goal is to win the $150,000 award administered by the Electronic Frontier Foundation offered for finding a 100 million digit prime number" -- of which $50,000 will be awarded to the discoverer, with another $50,000 going to a 501(c)(3) mathematics-related charity selected by GIMPS, and $50,000 retained by GIMPS to cover expenses and fund other awards.

65 comments

  1. Re: I love 8======D because I'm GayPK!! by Anonymous Coward · · Score: 0

    I use my computer to find Mersenne primes and other such projects when I am not writing code or shopping online. I thought Mersenne primes tended to be found in blocks like the digits of PI

  2. Prime numbers by Hognoxious · · Score: 4, Funny

    A prime number can only be divided by one, itself, and Chuck Norris.

    --
    Confucius say, "Find worm in apple - bad. Find half a worm - worse."
    1. Re:Prime numbers by Anonymous Coward · · Score: 0

      Chuck Norris jokes by definition cannot suck more than Chuck Norris movies but this one is close.

    2. Re:Prime numbers by rtb61 · · Score: 1

      Is not PI the largest possible prime number.

      --
      Chaos - everything, everywhere, everywhen
    3. Re:Prime numbers by Desler · · Score: 1

      The 90s called. They want their lame Chuck Norris jokes back.

    4. Re:Prime numbers by Anonymous Coward · · Score: 0

      2(Pi) -1? (ducks)

    5. Re: Prime numbers by phantomfive · · Score: 1

      Oh? A prime number can only be divided by itself, one, and Bruce Schneier. Better?

      --
      "First they came for the slanderers and i said nothing."
    6. Re: Prime numbers by Desler · · Score: 0

      More Carlos Mencia or Dane Cook.

    7. Re:Prime numbers by quenda · · Score: 1

      The 90s called. They want their lame Chuck Norris jokes back.

      Hey, uh, 1995 called! They want their “certain decade called wanting its 'blank back' formula back!

    8. Re:Prime numbers by 93+Escort+Wagon · · Score: 0

      The ocean called - they’re running out of shrimp!

      --
      #DeleteChrome
    9. Re: Prime numbers by 93+Escort+Wagon · · Score: 1

      Well I slept with your wife!

      --
      #DeleteChrome
    10. Re:Prime numbers by hcs_$reboot · · Score: 1

      That kind of joke reminds us we're all getting older.

      --
      Slashdot, fix the reply notifications... You won't get away with it...
    11. Re:Prime numbers by HyperQuantum · · Score: 1

      Actually, Chuck Norris can divide a prime number by any number he wants.

      --
      I am not really here right now.
    12. Re: Prime numbers by Anonymous Coward · · Score: 0

      My wife's in a coma,

  3. 51 is a PRIME number by Anonymous Coward · · Score: 0

    Area 51. Coincidence?

    1. Re:51 is a PRIME number by tepples · · Score: 1

      17 * 3 is not prime in the mathematical sense or in the free 2 business day shipping sense either.

    2. Re:51 is a PRIME number by iggymanz · · Score: 1

      that would be overpriced in the rib sense too, even at a posh restaurant

    3. Re:51 is a PRIME number by hcs_$reboot · · Score: 1

      51 is a PRIME number

      at Amazon?

      --
      Slashdot, fix the reply notifications... You won't get away with it...
  4. theory is wrong by phantomfive · · Score: 1

    Might want to look for a bug in your code first, before going on to blame mathematical theory.

    --
    "First they came for the slanderers and i said nothing."
    1. Re: theory is wrong by Anonymous Coward · · Score: 0

      I'm pretty sure someone did a brute-force confirmation that these numbers really are primes before announcing it. Finding large primes is computationally very expensive, verifying whether a given number is a prime is much faster.

    2. Re: theory is wrong by Anonymous Coward · · Score: 0

      GIMPS uses a very old method for verifying mersenne primes. https://en.wikipedia.org/wiki/Lucas–Lehmer_primality_test

  5. To What End by Anonymous Coward · · Score: 0

    A number with 24 million digits is well beyond the estimated number of particles in the universe which less than 100 digits.

    What use is a 24 million digit number? It may as well be an imaginary number like the square root -1.

    If a number cannot used in real life under any circumstance, what is the point?

    1. Re: To What End by Anonymous Coward · · Score: 0

      Not sure why you're comparing it to imaginary numbers, which have loads of applications. In addition to pure mathematics, they're commonly used in any field that deals with waves (quantum mechanics, electromagnetism, electronics, etc.), their spin-off quaternions which have multiple imaginary numbers are perhaps the most efficient way to describe rotations (commonly used in navigation and increasingly in computer graphics), and complex analysis often gives insight into the properties of real-valued functions (e.g. via contour integrals, residue integration, and analytic continuation). I could agree with you that such large prime numbers will likely have no near-future applications, but imaginary numbers are commonly used basically everywhere but perhaps computer science :).

    2. Re: To What End by ledow · · Score: 3, Informative

      Sigh.

      You can get past a googol (10^100) with factorials of a 3 digit number. Factorials are thus "using" such numbers. Calculate the odds of N things chosen out of M and the numbers explode quickly.

      Just because a number isn't representative of a practical physical quantity doesn't mean it's useless. Want an example? Encryption. I doubt you will ever exprienve having 2^4096 "anythings"... but the fact such a number exists, can be proven to be prime, and for which the mathematics applies is still incredibly useful.

    3. Re:To What End by hcs_$reboot · · Score: 1

      A number with 24 million digits

      Funny how we can only express a number by its power (10^24), and as technology and computers make progress we'll have to find another way to talk about a number when even its power is too large to give the brain a hint on how big it is.

      --
      Slashdot, fix the reply notifications... You won't get away with it...
    4. Re:To What End by AHuxley · · Score: 1

      The NSA and GCHQ always get a LOL as they wait for the outside world to catch up with their decryption math.

      --
      Domestic spying is now "Benign Information Gathering"
    5. Re: To What End by oobayly · · Score: 1

      How about (10^4899)^4899

    6. Re: To What End by Anonymous Coward · · Score: 0

      OK, sure, but what about when the second exponent get too large? What then hmmm?

    7. Re:To What End by Pembers · · Score: 1

      Don Knuth came up with a way of representing really big integers. He observed that multiplication is repeated addition and exponentiation is repeated multiplication, and devised a way to represent repeated exponentiation.

    8. Re:To What End by hcs_$reboot · · Score: 1

      That's nice thanks.

      --
      Slashdot, fix the reply notifications... You won't get away with it...
  6. Cool by ch-chuck · · Score: 1

    Thanks - I can use that as one of the primes in my 82 billion bit private key - lets see the NSA crack that!

    --
    try { do() || do_not(); } catch (JediException err) { yoda(err); }
    1. Re:Cool by ch-chuck · · Score: 1

      Believe it or not, the free Mathematica on a Raspberry Pi can calculate p = 2^82589933 - the last line is (without the -1)

        > 6640076912114355308311969487633766457823695074037951210325217902592

      Right now it's counting the digits (and may run out of memory yet)

      In[2]:= Total[DigitCount[p]]

      nope, just got the answer:
      Out[2]= 24862048

      --
      try { do() || do_not(); } catch (JediException err) { yoda(err); }
    2. Re:Cool by Anonymous Coward · · Score: 0

      I can calculate it easily. 2^82589933 is :
            100000000000000(...)000
      2^82589933-1 is :
            1111111111111111(....)111

    3. Re:Cool by gnasher719 · · Score: 1

      Thanks - I can use that as one of the primes in my 82 billion bit private key - lets see the NSA crack that!

      They will have no problem cracking that at all, because that prime number will be _known_.

  7. No use for for cryptography but still interesting by JoshuaZ · · Score: 4, Informative

    And before anyone asks, no these large Mersenne are much too large to be used in practical cryptography. There is a random number generator called a Mersenne twister which does use a Mersenne prime, but that uses much smaller ones to be feasible, and in any event is not sufficiently random to be safe for serious cryptographic purposes.

    The primary interest in these primes is two-fold: First they have a very efficient primality test, the Lucas-Lehmer test https://en.wikipedia.org/wiki/Lucas-Lehmer_primality_test and so if one is interested in simply finding very big primes, these are the ones to look for. For most of the last 100 years the largest known prime has beena Mersenne prime.

    Second, there's a connection with perfect numbers. A number is said to be perfect if the sum of all its positive divisors which are less than the number add up to the number. For example, 6 is perfect because 1,2 and 3 divide 6 and 1+2+3=6. But 8 is not perfect because 1+2+4=7 which is not perfect. The two oldest unsolved problems in all of math are a) are there any odd numbers which are perfect and b) are there infinitely many even numbers which are perfect? About 2000 years ago, Euclid recorded a proof (which may or may not have been due to him) that every Mersenne prime allows you to construct an even perfect number. In the 1700s, Euler proved that any even perfect number must arise from Euler's construction. So if one cares about answering this question about even perfect numbers, then one wants to investigate Mersenne primes.

  8. Re:No use for for cryptography but still interesti by Kjella · · Score: 3, Insightful

    About 2000 years ago, Euclid recorded a proof (which may or may not have been due to him) that every Mersenne prime allows you to construct an even perfect number.

    I "rediscovered" that proof as a teenager, and thought I was breaking new ground. Then I found it was actually discovered 2000+ years ago. Mathematics has a special way of putting your hubris in perspective.

    --
    Live today, because you never know what tomorrow brings
  9. Re: No use for for cryptography but still interest by mapkinase · · Score: 1

    What is practical application of this if not for cryptography

    --
    I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
  10. Re: No use for for cryptography but still interest by JoshuaZ · · Score: 1

    No direct practical applications. But practical and interesting are not synonyms.

  11. GIMPS has discovered the largest prime number by hcs_$reboot · · Score: 1

    Did they use a computer for that?

    --
    Slashdot, fix the reply notifications... You won't get away with it...
  12. Now that cryptocurrency mining doesn't pay by mnemotronic · · Score: 1

    With the price of crypto-currencies on what looks like a downward trend, this past week being an exception, could all that GPU and crypto-mining HW be put to use fishing for gimps?

    --
    The Russians have won. They have made the world a cesspool of distrust, greed, fear and hate.
    1. Re:Now that cryptocurrency mining doesn't pay by MarkRose · · Score: 1

      Yes. The most effective GIMPS use of GPUs is currently trial factoring. That is basically brute force division of the candidate number by small factors, which can eliminate candidates faster than doing a full Lucas-Lehmer or probable primality test. The new prime was trial factored by numbers up to 75 bits long. The optimal bit depth to trial factor to depends on the GPU hardware, especially the double-precision to single-precision ratio. Trial factoring can leverage SP, while the current LL and PRP implementations on GPUs leverage DP. There is an active investigation of the feasibility and performance of a SP LL/PRP algorithm on GPUs. GPU trial factoring for GIMPS is largely coordinated through GPU72.

      As for the non-GPU crypto hardware, someone could write an algorithm for FPGAs. The ASICs have no GIMPS utility that I'm aware of.

      --
      Be relentless!
    2. Re:Now that cryptocurrency mining doesn't pay by mnemotronic · · Score: 1

      Most of that is way over my head. I have always assumed that prime number searching would use integer math, think that was faster and that floating point would have rounding errors.

      --
      The Russians have won. They have made the world a cesspool of distrust, greed, fear and hate.
  13. GIMP by darkain · · Score: 1

    Wait, so the GIMP photo editor all along was just a ploy to search for Prime Numbers !?

    1. Re:GIMP by fibonacci8 · · Score: 1

      no

      --
      Inheritance is the sincerest form of nepotism.
    2. Re:GIMP by UnknownSoldier · · Score: 1

      /Oblg. Autistic Hat

      GIMPS, not GIMP.

  14. Re:No use for for cryptography but still interesti by grep+-v+'.*'+* · · Score: 1

    I "rediscovered" that proof as a teenager, ... Then I found it was actually discovered 2000+ years ago.

    But did Euclid bother to register it with the corresponding Copyright/Patent Offices? No? Then IT'S STILL UP FOR GRABS, DO IT NOW. Add "ON A COMPUTER" and you're golden.

    Or just say you identify as Euclid today and don't even bother. It didn't work for this guy but he was arguing about age, which is math. And we all know that "Math Is Hard" from that great sage: Barbie.

    --
    If the universe is someone's simulation -- does that mean the stars are just stuck pixels?
  15. Why the EFF? by mentil · · Score: 3, Interesting

    "GIMPS' next major goal is to win the $150,000 award administered by the Electronic Frontier Foundation offered for finding a 100 million digit prime number"

    Wait a minute, why is a civil liberties group funding a contest to solve mathematical problems? Following the link, they make it pretty clear that their ordinary funding doesn't go toward this, and apparently one interested party gave them the funds for this specific purpose. Still doesn't answer why they approached the EFF instead of, say, CERN.

    --
    Corruption is convincing someone that the selfless ideal is the same as their selfish ideal.
    1. Re:Why the EFF? by TeknoHog · · Score: 1

      Prime numbers are essential to modern cryptography, so it makes sense for the EFF to fund research into primes. While the newest Mersenne primes aren't readily necessary for cryptography, there's a lot more to the research. For instance, someone might discover a better way of predicting the distribution of prime numbers, which would make prime-based encryption weaker.

      --
      Escher was the first MC and Giger invented the HR department.
    2. Re:Why the EFF? by swillden · · Score: 1

      Prime numbers are essential to modern cryptography

      Not really. Prime numbers and the ability to generate them quickly and the difficulty of factoring products of large primes are essential to RSA, but that's on its way out anyway (too slow, too easy to implement incorrectly). Primes are relevant to ECC, but less so. Some ECC curves are in prime fields, others are not -- and the prime values used are all published, so prime distribution doesn't matter. The next, post-quantum, generation of asymmetric cipher and key agreement algorithms may or may not make use of primes at all, and none of the leading candidates depend on prime distribution. And, of course, few symmetric ciphers or message digests -- the real workhorses -- make any use of prime numbers.

      I have no problem with the EFF managing this contest, but it really doesn't have much more to do with cryptography than any other sort of fundamental research into number theory.

      --
      Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
    3. Re:Why the EFF? by retchdog · · Score: 1

      i suspect that the real answer to this involves John Perry Barlow getting really high with Whitfield Diffie once.

      --
      "They were pure niggers." – Noam Chomsky
  16. Re:No use for for cryptography but still interesti by complete+loony · · Score: 0

    Print it out and you could use it as a paper weight. Or just talk about it for a while I guess.

    --
    09F91102 no, 455FE104 nope, F190A1E8 uh-uh, 7A5F8A09 that's not it, C87294CE no. Ah! 452F6E403CDF10714E41DFAA257D313F.
  17. Re:No use for for cryptography but still interesti by bzipitidoo · · Score: 1

    I did a ballpark estimate of the odds of finding a new Mersenne prime. It's very roughly 1 in 100,000 that an untested number that has no small factors (less than 2^76) will be prime.

    Just one Mersenne prime has been found between 2^44 million and 2^74 million, and then there are no less than 3 in the next 10 million powers of 2 (minus 1 of course)? That's a lot.

    It adds to our knowledge. Maybe we can see more patterns, or figure out something else interesting. The more data we have...

    --
    Intellectual Property is a monopolistic, selfish, and defective concept. It is "tyranny over the mind of man"
  18. Re: No use for for cryptography but still interest by mapkinase · · Score: 1

    It is for me. Life is too short to spend on personal amusement.

    --
    I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
  19. Re:No use for for cryptography but still interesti by swillden · · Score: 1

    About 2000 years ago, Euclid recorded a proof (which may or may not have been due to him) that every Mersenne prime allows you to construct an even perfect number.

    I "rediscovered" that proof as a teenager, and thought I was breaking new ground. Then I found it was actually discovered 2000+ years ago. Mathematics has a special way of putting your hubris in perspective.

    No need for the scare quotes. If you found it on your own then you legitimately rediscovered it.

    --
    Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
  20. Re: No use for for cryptography but still interest by fibonacci8 · · Score: 1

    It is for me. Life is too short to spend on personal amusement.

    ... he says by posting on Slashdot.

    --
    Inheritance is the sincerest form of nepotism.
  21. Open Source FTW by Anonymous Coward · · Score: 0

    GIMPS has discovered the largest known prime number

    Let's see Photoshop do that!