Carol / Kynea Primes

[hunson]

Hey everyone,


I am searching for smaler projects in the science/prime area to participate.
Currently I am participating at primegrid and a private GFN13/14 server.

In the last couple of days I read a lot about the sieving process with the mtsieve framework etc. here in the forum.
I choose the Carol/Kynea primes because these a relatively small and fast to check and sieve. Setting up the MySQL and prpnet-server for multiple CPU-core workloads was no problem.

Now I experience a problem with the import of the ABC-file from cksieve into prpnet.

It does not seem to recognize the fileformat "ABC (130^$a$b)^2-2".
The server is set to Carol/Kynea primes via the servertype=32. I tried the fileformat "$a*<base>^$b$c" from the abcfileformat.txt but this also does not work.

Can anyone please explain what format type I have to set in the sieved file to be compatible with the prpnet import?

I know that I did no reserve any range to be sieved, for now this is just for testing and playing around. I will do a proper reservation once I figured all out.



Thanks in advance


hunson.

[hunson]

Hey everyone,


I am interested in smaller scientific projects especially prime numbers that a fast to sieve and check.

I did some reading about the sieving prp checking process here in the forum and now I want to play around with the software. The Carol/Kynea primes (http://www.noprimeleftbehind.net/Car...ime-search.htm) is my project of choise for now.

I decided to set up a prpnet-server for fun and multi-core workloads.

Setting up the MySQL- and prpnet-server was easy enough. When I try to import the ABC-file from cksieve into prpnet I get a error message which says something about a unrecognized format. Changeing the ABC-file according to the abcfileformats.txt (included in prpnet) from ABC (130^$a$b)^2-2 to ABC $a*<base>^$b$c did not resolve the issue. The prpnet-server is set to these primenumbers (servertype=32).

Can someone please tell me which format the sievefile for Carol/Kynea primes has to be edited to to import it in prpnet ?

Furthermore I did no find a "create_tables_mysql.sql" in the prpnet package, can someone please upload it ?


I know that I do not have a reserved range for any sieveing, for now this is just for fun and curiosity. I will do so if I figure everything out.



Thanks in advance


hunson

[kar_bon]

CK290 at n=153k, continuing
CK362 no further effort but still reserved

[gd_barnes]

Originally Posted by kar_bon

CK290 at n=153k, continuing
CK362 no further effort but still reserved

Thanks.


Are you still working on CK394 ?

[kar_bon]

CK394:
not reserved, only done to n=50k after prime found as posted.

[hunson]

Finished my ranges
34: No primes where found up to 50 k
130: No primes where found up to 30 k
140: No primes where found up to 30 k
The other ranges resulted the following prime numbers (checked with -tp flag of pfgw)

(132^28285+1)^2-2
(134^11556-1)^2-2
(134^28877-1)^2-2
(136^15340+1)^2-2
(136^20581-1)^2-2
(136^21467+1)^2-2
(138^16573+1)^2-2
(138^18222-1)^2-2
(140^18327-1)^2-2
(140^24595+1)^2-2
(140^28075+1)^2-2

I am wondering, can these numbers be submitted to The Prime Pages for completeness sake? There are Carol/Kynea primes listet, but they to not have a official subcategory like twin-primes.



Releasing all reserved ranges (34, 130, 132, 134, 136, 138, 140, 142)

[paulunderwood]

Originally Posted by hunson

Finished my ranges
34: No primes where found up to 50 k
130: No primes where found up to 30 k
140: No primes where found up to 30 k
The other ranges resulted the following prime numbers (checked with -tp flag of pfgw)
I am wondering, can these numbers be submitted to The Prime Pages for completeness sake? There are Carol/Kynea primes listet, but they to not have a official subcategory like twin-primes.



Releasing all reserved ranges (34, 130, 132, 134, 136, 138, 140, 142)

They are far too small for the top5000 primes:

We at the Prime Pages attempt to keep a list of the 5000 largest known primes plus a few each of certain selected archivable forms. To make the top 5000 today a prime must have 447881 digits or meet the size requirements for it's archivable form. (Query time: 0.001901 seconds.

They are not an "archivable form"; There are no papers published on C/K primes in reputable journals.

[hunson]

I thought so, too bad.
Well at least these primes are documented now :)

[kar_bon]

(688^70718+1)^2-2 found prime, 401338 digits
CK 688 at n=74k, continuing

[kar_bon]

Found (968^79512-1)^2-2, 474826 digits

Base 968 tested until n=80k, releasing.

[paulunderwood]

Originally Posted by kar_bon

Found (968^79512-1)^2-2, 474826 digits

Base 968 tested until n=80k, releasing.

Congrats on a top5000 prime (which requires generic modular reduction).