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

ECPP and related comments

Expand Messages
  • Chris Caldwell
    ... Now there s a pot calling the kettle black ;-) I appreciate (I think) all the help in finding holes in the rules and I greatly appreciate (I know!) the
    Message 1 of 18 , Apr 9, 2001
    • 0 Attachment
      At 07:00 AM 4/9/01 -0700, Phil Carmody wrote:
      > > Below is a 3474 digit number proved prime using Titanix. However, since
      > an EC
      > > test was not required until the 25th test (which was a 140 digit number), I
      > > don't think that this really counts as the ECPP record.
      >
      >Trouble-maker.

      Now there's a pot calling the kettle black ;-)

      I appreciate (I think) all the help in finding holes in "the rules" and I
      greatly appreciate
      (I know!) the enthusiasm of all of you. I wish I was completely up to the
      task. Part of
      the problem is the current format of the lists database, it is quite
      difficult to add numbers
      that do not have a short form. Partly this is by design, I have a strong
      bias toward
      the numbers in print, and would like folks to bend thier efforts toward the
      numbers
      "the community" has declared interesting. For ECPP, this would especially
      include
      such things as the large PRP's from the Cunningham project...

      In the past we decided that "APRT-CL assisted" could not be stamped on a prime
      for which the classical tests handle all of the early steps. (This came
      from a helpful
      discussion started by Hans Rosenthal) This (I think) should extend
      to the ECPP case. So I hardily agree with Paul as he writes:

      On Mon, 09 April 2001, "Paul Jobling" wrote:
      > Below is a 3474 digit number proved prime using Titanix. However, since
      an EC
      > test was not required until the 25th test (which was a 140 digit number), I
      > don't think that this really counts as the ECPP record.

      Of course Andrey pointed out the most glaring problem when he mentioned
      what I should have noticed--each ECPP proof is usually a chain of ECPP
      primes. I am not at all sure how to handle this one. I lean towards
      Bouk's suggestion:

      At 11:38 PM 3/28/01 -0800, Bouk de wrote:
      >A solution to this problem could be to no longer have
      >a top 20 of ECPP primes with the possible exception of
      >the record holder ..

      The idea of viewing each entry as a chain had appeal, but you know yourselves,
      you would not be able to resist leveraging off one prime to create another.

      For right now I am going to do nothing with the list; I will continue just
      including the "short form" ECPP primes and listening to the discussion.
      It may be that comments which refer to the proving method are just not
      viable in the long run.

      I am also slowly getting ready to modify the list. I have MySQL/PHP
      running on one of my servers and as a test case spent much of Christmas
      vacation translating the Prime Curios! database over. It may come on-line
      before too long. Perhaps this summer I can get to the Prime
      list... But as some of you know, I am slow... to many pans on the fire...

      But thanks again for the enthusiasm, suggestions, trouble making
      and all! You have been great!

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