Loading ...
Sorry, an error occurred while loading the content.

Re: [PrimeNumbers] PRP Top 50 has become PRP Top 100

Expand Messages
  • mikeoakes2@aol.com
    In a message dated 20/04/2001 21:30:09 GMT Daylight Time, ... digits). Why the lower limit on size, Henri? There is a natural cutoff set by the ability of
    Message 1 of 9 , May 6, 2001
    • 0 Attachment
      In a message dated 20/04/2001 21:30:09 GMT Daylight Time,
      HLifchitz@... writes:

      >Please continue sending me new probable primes (but larger than 12000
      digits).

      Why the lower limit on size, Henri? There is a natural cutoff set by the
      ability of programs like Titanix to prove primality - currently about 2300
      digits - so people aren't going to send you smaller PRPs than that. And as
      these prgrams (and hardware speeds) improve, the smaller candidates on your
      list would gradually get removed.

      At present there is a "vacuum", with nowhere for interesting PRPs in the
      range 2300-12000 digits to go. So it would be hospitable of you to give them
      a home. Moreover, they would then be publicly visible, making it easier for
      someone to take up the challenge of proving them prime as soon as that
      becomes feasible.

      Mike Oakes
    • d.broadhurst@open.ac.uk
      ... Don t forget VFYPR. For cases where N^2-1 has good -- but not yet BLS -- factorization, VFYPR is more powerful. Bouk and I and are mounting an attack on
      Message 2 of 9 , May 6, 2001
      • 0 Attachment
        Mike Oakes wrote:

        > There is a natural cutoff set by the
        > ability of programs like Titanix to prove primality

        Don't forget VFYPR. For cases where N^2-1 has good --
        but not yet BLS -- factorization, VFYPR is more powerful.

        Bouk and I and are mounting an attack on the APRCL record,
        at 5k+ digits, after 20 days CPUtime spent on ECM,
        getting 72 digits short of 3*F1+F2=1. We have good actuarial
        reasons for believing that VFYPR will complete
        before ECM is likely to cough up another p35.

        It would be useful, for cyclotomic PrPs,
        like U, V, (2^n+1)/3, repunits, etc,
        if folk posted some percentages (as I did) in

        http://
        ourworld.compuserve.com/homepages/hlifchitz/Henri/fr-us/PrpRec.htm
        [too long for one yahoo line, sorry]

        David
      • Michael Bell
        ... Hi, I think it would be useful if there was a list of PRP s with over 15 or 20% of either N-1 or N+1 factored (or maybe 3*F1+F2 =.6 or so). This would
        Message 3 of 9 , May 6, 2001
        • 0 Attachment
          >
          > Don't forget VFYPR. For cases where N^2-1 has good --
          > but not yet BLS -- factorization, VFYPR is more powerful.
          >
          > Bouk and I and are mounting an attack on the APRCL record,
          > at 5k+ digits, after 20 days CPUtime spent on ECM,
          > getting 72 digits short of 3*F1+F2=1. We have good actuarial
          > reasons for believing that VFYPR will complete
          > before ECM is likely to cough up another p35.
          >
          > It would be useful, for cyclotomic PrPs,
          > like U, V, (2^n+1)/3, repunits, etc,
          > if folk posted some percentages (as I did) in
          >

          Hi,

          I think it would be useful if there was a list of PRP's with over 15 or 20%
          of either N-1 or N+1 factored (or maybe 3*F1+F2>=.6 or so). This would then
          make a list of primes that it may actually be practical to prove in the not
          to distant future.

          Any volunteers for hosting a page?

          Michael.
        • d.broadhurst@open.ac.uk
          ... Well Andy Steward has *oodles* of such animals... David
          Message 4 of 9 , May 6, 2001
          • 0 Attachment
            Michael Bell wrote:

            > I think it would be useful if there was a list of PRP's
            > with over 15 or 20% of either N-1 or N+1 factored
            > (or maybe 3*F1+F2>=.6 or so). This would then
            > make a list of primes that it may actually be practical
            > to prove in the not to distant future.
            > Any volunteers for hosting a page?

            Well Andy Steward has *oodles* of such animals...

            David
          • d.broadhurst@open.ac.uk
            PS: My personal gigantic bete noire is Phi(2521,9926), with 10072 digits, and N-1 347 digits short of BLS. As Gollum said, we hates it.... David
            Message 5 of 9 , May 6, 2001
            • 0 Attachment
              PS: My personal gigantic bete noire is Phi(2521,9926),
              with 10072 digits, and N-1 347 digits short of BLS.
              As Gollum said, we hates it.... David
            • Mohale's Hoek TRC
              Guys, Primeform gives the following as PRP 6*k*(2^n - 1 - k ) + 2^n - 1, with n = 4497, k = 8939. digit number = 13400. A beginner in this business, I would
              Message 6 of 9 , May 18, 2001
              • 0 Attachment
                Guys,

                Primeform gives the following as PRP

                6*k*(2^n - 1 - k ) + 2^n - 1, with n = 4497, k = 8939.
                digit number = 13400.

                A beginner in this business, I would like confirmation of this result from
                someone.

                Thank you.

                Peter Lesala.

                -----Original Message-----
                From: Henri LIFCHITZ <HLifchitz@...>
                To: Primes List <PrimeNumbers@yahoogroups.com>
                Date: Friday, April 20, 2001 10:42 PM
                Subject: [PrimeNumbers] PRP Top 50 has become PRP Top 100


                >Hello all,
                >
                >By popular demand, I have decided to extend the PRP Top 50 to a PRP Top
                100.
                >(see http://www.primes.fr.st or
                http://ourworld.compuserve.com/homepages/hlifchitz/ )
                >For convenience reasons, only the first 50 records will keep a specific
                page.
                >
                >Please continue sending me new probable primes (but larger than 12000
                digits).
                >
                >Happy hunting,
                >
                >Henri
                >
                >
                >[Non-text portions of this message have been removed]
                >
                >
                >Unsubscribe by an email to: primenumbers-unsubscribe@egroups.com
                >The Prime Pages : http://www.primepages.org
                >
                >
                >
                >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
                >
                >
                >
              • mikeoakes2@aol.com
                In a message dated 18/05/2001 14:29:21 GMT Daylight Time, ... Hi Peter, Something must be wrong with your expression as (if we do the substitutions for n and
                Message 7 of 9 , May 18, 2001
                • 0 Attachment
                  In a message dated 18/05/2001 14:29:21 GMT Daylight Time,
                  mohales@... writes:

                  > Primeform gives the following as PRP
                  > 6*k*(2^n - 1 - k ) + 2^n - 1, with n = 4497, k = 8939.
                  > digit number = 13400.
                  > A beginner in this business, I would like confirmation of this result from
                  > someone.
                  > Thank you.
                  > Peter Lesala.

                  Hi Peter,
                  Something must be wrong with your expression as (if we do the substitutions
                  for n and k) PFGW says:-
                  6*8939*(2^4497-1-8939)+2^4497-1 has factors: 7
                  Want to double-check it?
                  Mike Oakes
                • Mohale's Hoek TRC
                  Hi Mike, There is an error in the value of n = 4497, the value should be n = 44497. Thanks for the quick response. Peter. ... From: Mikeoakes2@aol.com
                  Message 8 of 9 , May 20, 2001
                  • 0 Attachment
                    Hi Mike,

                    There is an error in the value of n = 4497, the value should be n = 44497.

                    Thanks for the quick response.

                    Peter.
                    -----Original Message-----
                    From: Mikeoakes2@... <Mikeoakes2@...>
                    To: mohales@... <mohales@...>
                    Cc: PrimeNumbers@yahoogroups.com <PrimeNumbers@yahoogroups.com>
                    Date: Friday, May 18, 2001 10:03 PM
                    Subject: Re: [PrimeNumbers] PRP Top 50 has become PRP Top 100


                    In a message dated 18/05/2001 14:29:21 GMT Daylight Time,
                    mohales@... writes:

                    > Primeform gives the following as PRP
                    > 6*k*(2^n - 1 - k ) + 2^n - 1, with n = 4497, k = 8939.
                    > digit number = 13400.
                    > A beginner in this business, I would like confirmation of this result from
                    > someone.
                    > Thank you.
                    > Peter Lesala.

                    Hi Peter,
                    Something must be wrong with your expression as (if we do the substitutions
                    for n and k) PFGW says:-
                    6*8939*(2^4497-1-8939)+2^4497-1 has factors: 7
                    Want to double-check it?
                    Mike Oakes
                  Your message has been successfully submitted and would be delivered to recipients shortly.