mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Prime Wiki (https://www.mersenneforum.org/forumdisplay.php?f=158)
-   -   Prime-Wiki (https://www.mersenneforum.org/showthread.php?t=24142)

kar_bon 2019-05-13 15:04

New:
[url='https://www.rieselprime.de/ziki/Williams_prime']Table[/url] for Williams primes (b±1)•b[sup]n[/sup]±1 with
- links to categories
- links to the list tables of the four types
- files for smallest n-value of bases ≤1024

Just only uploaded those ASCII files, but have to be categorized in the file category. This should be done later perhaps with a template.
Clicking the file link will lead to the file, if logged in new files can be uploaded, old files can be deleted (history see next).

ToDo:
create an own pages for any type to show remaining/open bases without any values found so far and the history.

And not yet included:
- found yesterday 362*363^142876+1 is prime (365752 digits)

kar_bon 2019-05-14 13:19

A table for all types of Williams primes with links to the category, the data tables and the files with smallest n-values are moved to the main page of [url='https://www.rieselprime.de/ziki/Williams_prime']"Williams prime"[/url] (not the category anymore), because a search will first lead to this page.

More pages to come, much to do further, statistic is growing.

LaurV 2019-05-15 03:53

Small nitpicking: on the Riesel primes table, it is mentioned "green = base reserved", etc, but actually, those are not bases, those are "k"s. The base is always 2. This is a bit confusing, when you search the web looking for Riesel primes in some base.

Also, all odd numbers can be written as k*2^n-1 for some k and n, therefore all primes are Riesel primes according with your definition. It has to be amended to say something like "where 2^n>k" or so, similar to the Proth side.

kar_bon 2019-05-15 07:47

Thanks for the hint.

I've changed the definition on Proth/Riesel prime page and replaced the 'base' on the table pages.

kar_bon 2019-05-15 13:46

Update for "Williams prime":
Instead of linking to tha file of the least n-value I created an own [url='https://www.rieselprime.de/ziki/Williams_prime_MM_least']page[/url] which shows wanted values and the data table.
This table is generated from a CSV file (had to be renamed and csv-ed first) and showing on that page using an extension.
[STRIKE]For the other three types this has to be done also,[/STRIKE] links to CRUS bases (not yet implemented) should come later.

More and more linkings to related pages will occur like Williams, Near Woodall and Riesel for example.

[COLOR="Green"]Done[/COLOR]: own pages for all 4 types of Williams primes with files created,

kar_bon 2019-05-17 11:53

The values for Williams primes of the form (b-1)•b^n-1 are now for b <=64 (& 128) online and the [url='https://www.rieselprime.de/ziki/Williams_prime_MM_table']table[/url] shows them correctly even using data from other primes:
For bases power of 2 the data will be retrieved from their Riesel primes (n-values, search range, reserved, date).

The only thing here: I have to 'teach' the table to show a color (perhaps a different for reservation and reusing) if a Riesel k-value is involved.

This reuse will minimize of doing double work here and on other prime sequences.

sweety439 2019-05-24 05:46

Found some errors in the wiki pages:

1. [URL="https://www.rieselprime.de/ziki/Williams_prime_MP_452"]Williams prime MP 452[/URL], the "Max n" is 71490, not 71940.

2. [URL="https://www.rieselprime.de/ziki/Williams_prime_MM_128"]Williams prime MM 128[/URL], the "Max n" is 1.95M (not 800000), and this sequence is reserved by Cruelty, see the post [URL="https://mersenneforum.org/showpost.php?p=509799&postcount=72"]https://mersenneforum.org/showpost.php?p=509799&postcount=72[/URL] by Cruelty.

3. [URL="https://www.rieselprime.de/ziki/Williams_prime_MP_363"]Williams prime MP 363[/URL], the "Max n" should be 142876, why 142870?

4. For the external links of the page [URL="https://www.rieselprime.de/ziki/Williams_prime"]Williams prime[/URL], the range of b of [URL="http://harvey563.tripod.com/wills.txt"]http://harvey563.tripod.com/wills.txt[/URL] is 3 ≤ b ≤ 2049, not 2 ≤ b ≤ 2048.

sweety439 2019-05-24 06:10

Also, there are errors in all four "Williams prime least" page, the range of bases b is 2 [B]≤[/B] b [B]≤[/B] 1024, not 2 ≥ b ≥ 1024.

kar_bon 2019-05-24 11:58

Thanks, I've corrected some errors.

Williams 128: not yet, waiting for response.
Gen. repunit prime: not for now, perhaps later

sweety439 2019-05-27 06:14

For [URL="https://www.rieselprime.de/ziki/Williams_prime_MM_2"]Williams prime MM 2[/URL], the "Max n" is not 83,000,000, since it is not verified whether any undiscovered Mersenne primes exist between the 47th (M43,112,609) and the 51st (M82,589,933).

Also, for [URL="https://www.rieselprime.de/ziki/Williams_prime_MP_2"]Williams prime MP 2[/URL], the "Max n" is 8589934591 = 2^33-1, not 8,590,000,000, since it is not known whether F33 = 2^(2^33)+1 is prime or not.

kar_bon 2019-05-27 10:02

I've changed the n-max for Proth prime k=1 so this value will be listed for Williams MP 2.

The n-max for Williams MM 2 = n-max for Riesel k=1 (=Mersenne primes = GIMPS) is correct here.
The GIMPS milestone lists (currently) "All exponents below 83 676 167 have been tested at least once." so this is the same value taken for all other sequences in the Wiki.
The double check (=verified) is only available for GIMPS, (mostly) no other prime seq. in the Wiki is double checked by others (hardware & software) and searched ranges or found primes (Top5000 or not) are depend on any remark in the forum or other web sources.


All times are UTC. The time now is 18:52.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.