Software for finding prime constellations

[Cybertronic]

Searching up to 103'515'091'680'000'000 = 29#*16000000 is done.

No 19-tuplet found.

best was:

91466285060714489+0000 is 3-PRP! (0.000000 seconds)
91466285060714489+0008 is 3-PRP! (0.000000 seconds)
91466285060714489+0024 is 3-PRP! (0.000000 seconds)
91466285060714489+0048 is 3-PRP! (0.000000 seconds)
91466285060714489+0080 is 3-PRP! (0.000000 seconds)
91466285060714489+0120 is 3-PRP! (0.000000 seconds)
91466285060714489+0168 is 3-PRP! (0.000000 seconds)
91466285060714489+0224 is 3-PRP! (0.000000 seconds)
91466285060714489+0288 is 3-PRP! (0.000000 seconds)
91466285060714489+0360 is 3-PRP! (0.000000 seconds)
91466285060714489+0440 is 3-PRP! (0.000000 seconds)
91466285060714489+0528 is 3-PRP! (0.000000 seconds)
91466285060714489+0624 is 3-PRP! (0.000000 seconds)
91466285060714489+0728 is composite: [C99E7872E157A1] (0.000000 seconds)
91466285060714489+0840 is composite: [BCF9EFA1D3E759] (0.000000 seconds)
91466285060714489+0960 is 3-PRP! (0.000000 seconds)
91466285060714489+1088 is 3-PRP! (0.000000 seconds)
91466285060714489+1224 is 3-PRP! (0.000000 seconds)
91466285060714489+1368 is 3-PRP! (0.000000 seconds)

You can continue this job, if you like.

[CRGreathouse]

Originally Posted by Cybertronic

Download and rename into *.bi

I get:

[path]\ld.exe: cannot find -lbig_int

[Cybertronic]

Originally Posted by CRGreathouse

I get:

[path]\ld.exe: cannot find -lbig_int

You must start the command-box


compile with fbc yourprogram.bas


I hope you have no Linux...it works only for Windows 64bit. Otherwise it don't work

[CRGreathouse]

Originally Posted by Cybertronic

You must start the command-box


compile with fbc yourprogram.bas

Yes, when I do that on 64-bit Windows I get the indicated error message. (I got a different one before I added the .bi file; it is able to find that, fortunately.)

[Cybertronic]

Yes, confirm it and found the reason ... add this files to Freebasic root folder , the same like file fbc

https://pzktupel.de/transfer/additionsFB.zip


BTW, the 19tuplet sieve need a file "primes.src"... these are only stored primes on my PC, here specific primes from 200 to ~ 50000.


...you get a new single version from me for downloading

[Cybertronic]

https://pzktupel.de/transfer/additionsFB.zip


whats new:


file "primes.src" available
fixed k-interval is k=1000 -> offsetsize : 32589158477190044730000

for single core , each condition will be saved to file 01,02,03,04,....
example: 07 have number 211927298241986152462573, so the first 7 conditions are prime.

many people can now look for this unknown 19-tuplet via reservation thread


start with k=58000 -> from 1890171191677022594340000 (valid 18-tuplet 1906230835046648293290043)


note: complete k=1000 range took 4+ days.


so you can start with

58000 core1

59000 core2
60000 core 3 and so on.

[Cybertronic]

After I found the smallest 7k triplet 2nd kind, I will look for the smallest 55-digit prime 12-tuplet. This will take 1 week to reach an offset 1e21.

Possible 1st kind is found. Offset with 22 digits.
1000000000000000000000000000000002955087732304487826931 + d, d = 0, 2, 6, 8, 12, 18, 20, 26, 30, 32, 36, 42 is the smallest 55 digit prime 12-tuplet to given pattern....in 30h it is sure. [time to compute: 16d]

[Cybertronic]

There was another smaller hit in the same cycle.
1000000000000000000000000000000002743862590724468830501 + d, d = 0, 2, 6, 8, 12, 18, 20, 26, 30, 32, 36, 42. This is now the smallest.
http://factordb.com/index.php?query=...at=1&sent=Show

[Cybertronic]

Page "Patterns of prime k-tuplets & the Hardy-Littlewood constants" got an update.

All specific k-values up to prime 30-tuplet now with higher accuracy. Provided by Martin Raab !

A smallest non-trivial prime 30-tuplet is larger then 10^47
A smallest non-trivial prime 24-tuplet is larger then 10^36

https://pzktupel.de/ktpatt_hl.php

Best wishes

[MattcAnderson]

Keep up the good work Cybertronic. Your efforts are appreciated.

Matt

[storm5510]

Am I seeing references to FreeBasic here?

I don't do much programming anymore. When I need something quick and dirty, FB is where I go. Despite its name, it can be a rather complex platform when a person starts using libraries, and there are a lot of them.

As for the big integer library, I have messed around with it a little. It can do some big numbers but nothing like what the "C" based languages can do. About the library, it doesn't appear that anyone is using the proper format to use it.

#include "big_int.bi"

There is a fairly decent IDE editor for FB. Simple name: FBIDE. It can be found here. There is a sizable help/reference library included.