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

GMP-ECM with huge B1

Expand Messages
  • biwema
    I ve noticed, that there are sometimes troubles with GMP-ECM running a p-1 check with bounds that are bigger than 4e9. I ran a p-1 test with B1=1e10 on 2^933+1
    Message 1 of 2 , Mar 3, 2004
      I've noticed, that there are sometimes troubles with GMP-ECM running
      a p-1 check with bounds that are bigger than 4e9.

      I ran a p-1 test with B1=1e10 on 2^933+1 and saved the result.

      When I tried to resume that (B1=1e10 and B2=1e12),
      I always got an error that I should use a smaller B1 or B2 min.
      Only when I set B1 and B2 to a value smaller than 4 billion, it
      worked.
      But in that way it is not possible to test B2 in different pieces up
      to high bounds.
      Furthermore, I am not sure, if that might lead to incorrect results,
      if the resuming point (B1) is smaller than the 10 billion in the
      savefile.
      I tested it on the p4 and athlon binary.

      I found a patch on the web
      (http://www.loria.fr/~zimmerma/records/ecm-5.0.patch1) but I guess
      that is already included and is printing the error messages.

      Is there a way to avoid that problem? There was nothing like this
      mentioned in the readme and todo file.

      -

      If I start to ecm B1 with Prime95 on cunningham numbers where some
      factors are already known, the input of Prime95 is just the whole
      exponent, and the factors are listed in lowm.txt.
      After taking it ober to GMPECM for stage 2, is it possible, to
      eclude this factors? I think, it might cause problems, if I just
      divide it off in the input file, because on the impact of the
      intermediate result after stage 1.
    • andrew_j_walker
      While we re talking about GMP-ECM, was the trial factoring problem in version 5 ever fixed? Andrew
      Message 2 of 2 , Mar 3, 2004
        While we're talking about GMP-ECM, was the trial factoring problem in
        version 5 ever fixed?

        Andrew
      Your message has been successfully submitted and would be delivered to recipients shortly.