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

Re: Fermat or Rabin/Miller PRP test uing YEAFFT

Expand Messages
  • j_chrtn
    ... multiplication routines. ... routines, and it will use the YEAFFT multiplications. ... Hi Phil, If I can plug YEAFFT multiplication routines into GMP
    Message 1 of 6 , Sep 2, 2008
    • 0 Attachment
      --- In primenumbers@yahoogroups.com, Phil Carmody <thefatphil@...>
      wrote:
      >
      > > Does anyone of you knows if there exists a C program to
      > > perform PRP
      > > tests (basic Fermat test or Rabin/Miller test) using the
      > > YEAFFT library
      > > that comes with glucas package ?
      > >
      > > As far as I know, glucas can only check for primality of
      > > Mersenne's
      > > numbers and cannot test other numbers for probable
      > > primality.
      >
      > I seem to remember it comes with the ability to replace GMP
      multiplication routines.
      > Get that working, and then just use the usual GMP exponentiation
      routines, and it will use the YEAFFT multiplications.
      >
      > Phil
      >

      Hi Phil,

      If I can plug YEAFFT multiplication routines into GMP library, that
      great.
      However, in this case, I wonder how difficult it is to do and most of
      all how I can then take advantage of the ability of YEAFTT to
      parallelize computation on SMP computers from the GMP exponentiation
      routines ?

      Actually, this is mostly this ability of YEAFTT to fit well on SMP
      machines that interested me to test PRP of the form (n+1)^p-n^p and
      especially 138^p-137^p. This later case is very tough : I've checked
      all prime exponents up to 168000 and found no PRP :-(

      JL
    • j_chrtn
      ... there ... PRP ... Mersenne s ... Hello Mark, As mentioned in my previous reply to Phil, the numbers I want to check are (n+1)^p-n^p for which I don t know
      Message 2 of 6 , Sep 2, 2008
      • 0 Attachment
        --- In primenumbers@yahoogroups.com, Mark Rodenkirch <mgrogue@...>
        wrote:
        >
        > Are there numbers of specific form that you want to test? For
        > example, if you want to test numbers of the form k*b^n+/-1, then
        there
        > are options that don't use YEAFFT.
        >
        > On Sep 2, 2008, at 8:04 AM, j_chrtn wrote:
        >
        > > Hello group,
        > >
        > > Does anyone of you knows if there exists a C program to perform
        PRP
        > > tests (basic Fermat test or Rabin/Miller test) using the YEAFFT
        > > library
        > > that comes with glucas package ?
        > >
        > > As far as I know, glucas can only check for primality of
        Mersenne's
        > > numbers and cannot test other numbers for probable primality.
        > >
        > > Regards,
        > >
        > > JL
        >
        >
        >
        > [Non-text portions of this message have been removed]
        >

        Hello Mark,

        As mentioned in my previous reply to Phil, the numbers I want to
        check are (n+1)^p-n^p for which I don't know any
        specialized/optimized program. So I used to use pfgw for this
        purpose. Pfgw is great but as far as I know cannot take advantage of
        SMP hardware...

        Do you have any idea of a program I could use for this form of
        numbers on SMP ?

        Regards,

        JL
      • Mark Rodenkirch
        pfgw is your best bet. I don t know if YEAFFT will be faster than it for numbers of that form. ... [Non-text portions of this message have been removed]
        Message 3 of 6 , Sep 2, 2008
        • 0 Attachment
          pfgw is your best bet. I don't know if YEAFFT will be faster than it
          for numbers of that form.

          On Sep 2, 2008, at 6:20 PM, j_chrtn wrote:

          > Hello Mark,
          >
          > As mentioned in my previous reply to Phil, the numbers I want to
          > check are (n+1)^p-n^p for which I don't know any
          > specialized/optimized program. So I used to use pfgw for this
          > purpose. Pfgw is great but as far as I know cannot take advantage of
          > SMP hardware...
          >
          > Do you have any idea of a program I could use for this form of
          > numbers on SMP ?
          >
          > Regards,
          >
          > JL



          [Non-text portions of this message have been removed]
        Your message has been successfully submitted and would be delivered to recipients shortly.