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

Losing relation sets

Expand Messages
  • Larry Soule
    I m running SNFS on a number of difficulty 146. The minFF is 255382. The number of relation-sets reported grew up to 238195. However, after the next set of
    Message 1 of 4 , Jun 1, 2005
    • 0 Attachment
      I'm running SNFS on a number of difficulty 146.
      The minFF is 255382. The number of relation-sets
      reported grew up to 238195. However, after the next
      set of sieving, that number went down to 80299.

      -> Found 80299 relation-sets versus minFF=255382.

      The jpg graph shows the same thing - a nice curve up
      to 238k followed by a big drop. Is this normal
      or is it a problem with my poly selection?

      thanks,
      Larry
    • gchil0
      Sounds like the code to eliminate heavy relations sets massacred your relations. I suspect that the factor base limit was a bit too small, but that s just a
      Message 2 of 4 , Jun 1, 2005
      • 0 Attachment
        Sounds like the code to eliminate heavy relations sets massacred your
        relations. I suspect that the factor base limit was a bit too small,
        but that's just a guess.

        Greg

        --- In ggnfs@yahoogroups.com, "Larry Soule" <lsoule@g...> wrote:
        > I'm running SNFS on a number of difficulty 146.
        > The minFF is 255382. The number of relation-sets
        > reported grew up to 238195. However, after the next
        > set of sieving, that number went down to 80299.
        >
        > -> Found 80299 relation-sets versus minFF=255382.
        >
        > The jpg graph shows the same thing - a nice curve up
        > to 238k followed by a big drop. Is this normal
        > or is it a problem with my poly selection?
        >
        > thanks,
        > Larry
      • trilliwig
        For those not aware, the default script-selected parameters for GGNFS assume a polynomial of degree 5 for middling to large factorizations (GNFS ~97 digits,
        Message 3 of 4 , Jun 2, 2005
        • 0 Attachment
          For those not aware, the default script-selected parameters for GGNFS
          assume a polynomial of degree 5 for middling to large factorizations
          (GNFS > ~97 digits, SNFS > ~117 digits). If a polynomial of other
          degree is used, these parameters will likely be suboptimal.

          --
          Sam

          --- In ggnfs@yahoogroups.com, "gchil0" <jgchilders@m...> wrote:
          > Sounds like the code to eliminate heavy relations sets massacred your
          > relations. I suspect that the factor base limit was a bit too small,
          > but that's just a guess.
          >
          > --- In ggnfs@yahoogroups.com, "Larry Soule" <lsoule@g...> wrote:
          > > I'm running SNFS on a number of difficulty 146.
          > > The minFF is 255382. The number of relation-sets
          > > reported grew up to 238195. However, after the next
          > > set of sieving, that number went down to 80299.
          > >
          > > -> Found 80299 relation-sets versus minFF=255382.
          > >
          > > The jpg graph shows the same thing - a nice curve up
          > > to 238k followed by a big drop. Is this normal
          > > or is it a problem with my poly selection?
        • Larry Soule
          Yep, that s exactly what I did - I was trying out a degree 6 poly. thanks, Larry ... GGNFS ... your ... small, ... poly selection?
          Message 4 of 4 , Jun 2, 2005
          • 0 Attachment
            Yep, that's exactly what I did - I was trying out
            a degree 6 poly.

            thanks,
            Larry


            --- In ggnfs@yahoogroups.com, "trilliwig" <trilliwig@a...> wrote:
            > For those not aware, the default script-selected parameters for
            GGNFS
            > assume a polynomial of degree 5 for middling to large factorizations
            > (GNFS > ~97 digits, SNFS > ~117 digits). If a polynomial of other
            > degree is used, these parameters will likely be suboptimal.
            >
            > --
            > Sam
            >
            > --- In ggnfs@yahoogroups.com, "gchil0" <jgchilders@m...> wrote:
            > > Sounds like the code to eliminate heavy relations sets massacred
            your
            > > relations. I suspect that the factor base limit was a bit too
            small,
            > > but that's just a guess.
            > >
            poly selection?
          Your message has been successfully submitted and would be delivered to recipients shortly.