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 2019-06-05 08:11

I've included a separate page for all remaining data for [url='https://www.rieselprime.de/ziki/Williams_prime']Williams primes[/url] (for now only MM-type).
Listed are all bases b < 2050 not yet available as own page and n-max=1000 or given range.

 sweety439 2019-06-09 00:02

In the page [URL="https://www.rieselprime.de/ziki/Proth_prime_small_bases_least_n"]Proth prime small bases least n[/URL], the data is not correct: the forms 2*512^n+1, 4*32^n+1, 4*512^n+1, 4*1024^n+1, 10*1000^n+1 and 12*12^n+1 [I]may[/I] have primes, since it is not known whether there are Fermat primes > 2^(2^4)+1, etc. (all these numbers are [B]generalized Fermat numbers[/B]). (Note that such forms do not include 8*128^n+1, 8*128^n+1 have [I]no possible[/I] primes)

Also, recently a prime 7*1004^54848+1 was found, please add it.

 Dylan14 2019-06-10 20:07

[QUOTE=WGJC3107;519023]
Is it possible for me to make an account on there to edit pages or is that just admin’s job?[/QUOTE]

You’ll have to contact Karsten directly in PM or email, or you can post in the “let me in” thread with your desired name.

 kar_bon 2019-06-14 09:14

I've inserted/created a category for [url='https://www.rieselprime.de/ziki/Williams_prime']Williams[/url] (types MM and MP so far) sequences without any prime found yet and listed the countings in the table.
I've also added all wanted seqs. for Williams type [url='https://www.rieselprime.de/ziki/Williams_prime_MP_least']MP[/url] with searched ranges not yet found anywhere.

The other types (PM and PP) will follow later.

 sweety439 2019-06-15 04:44

[QUOTE=kar_bon;519292]I've inserted/created a category for [url='https://www.rieselprime.de/ziki/Williams_prime']Williams[/url] (types MM and MP so far) sequences without any prime found yet and listed the countings in the table.
I've also added all wanted seqs. for Williams type [url='https://www.rieselprime.de/ziki/Williams_prime_MP_least']MP[/url] with searched ranges not yet found anywhere.

The other types (PM and PP) will follow later.[/QUOTE]

For Williams MP, there should be also notes:

* Base 512: Proth prime 511•2[SUP]n[/SUP]+1 (first n-value divisible by 9)
* Base 817: Conjectures 'R Us not started yet

Like the notes in Williams MM:

* Base 128: Riesel prime 127•2[SUP]n[/SUP]-1 (first n-value divisible by 7)
* Base 478: Conjectures 'R Us not started yet

 sweety439 2019-06-15 04:58

The note for the data list in the page [URL="https://www.rieselprime.de/ziki/Proth_prime_small_bases_least_n"]Proth prime small bases least n[/URL] is not correct, some forms are also generalized Fermat numbers but with known primes (such as 2*8^n+1, 4*2^n+1, 6*6^n+1, etc.), and there are numbers instead of "-" in the list for these forms, however, for generalized Fermat numbers without known primes (such as 2*512^n+1, 4*32^n+1, 12*12^n+1, etc.), there are "-" in the list, these forms have no known primes, like all current 0's in the list (such as 2*365^n+1, 4*53^n+1, 5*308^n+1, etc.), all of these forms have no known primes. Do you think the generalized Fermat numbers (2*512^n+1, 4*32^n+1, 12*12^n+1, ...) should be listed in the "Wanted values" section? They also have the Nash Weight, and there are no known primes of the form 2*512^n+1, like there are also no known primes of the form 2*365^n+1.

(I know that CRUS excludes the generalized Fermat numbers like 2*512^n+1 from searching, but some problems requires the generalized Fermat numbers, e.g. the "minimal primes problem", finding all minimal primes in given base, for base 32, the form 4{0}1 (4000...0001) is generalized Fermat numbers 4*32^n+1, and for my problem [URL="https://mersenneforum.org/showthread.php?t=21839"]A Sierpinski/Riesel-like problem[/URL], I do not exclude generalized Fermat numbers like 4*32^n+1 from searching and consider the problem S32 is not proven and with k=4 remain.

 sweety439 2019-06-17 17:18

[URL="https://www.rieselprime.de/ziki/Category:Reserved"]Category: Reserved[/URL] contains Williams MP reserved by CRUS (123, 342, 438, 487, 757, 997, 1005) but does not contain Williams PP reserved by CRUS (327 and 1017).

 sweety439 2019-06-18 13:03

Currently [URL="https://www.rieselprime.de/ziki/Category:Reserved"]Category:Reserved[/URL] does not contain Williams MM 268, which is also reserved by CRUS.

 sweety439 2019-06-18 13:06

[URL="https://www.rieselprime.de/ziki/Williams_prime_PM_25"]Williams PM 25[/URL] has a remark that this should be (re)used from base 5, but [URL="https://www.rieselprime.de/ziki/Williams_prime_MM_25"]Williams MM 25[/URL] and [URL="https://www.rieselprime.de/ziki/Williams_prime_MP_25"]Williams MP 25[/URL] do not have, is this true for all perfect power bases (4, 8, 9, 16, 25, 27, 32, 36, 49, 64, 81, 100, 121, 125, 128, 144, 169, 196, 216, 225, 243, 256, 289, 324, 343, 361, 400, 441, 484, 512, 529, 576, 625, 676, 729, 784, 841, 900, 961, 1000, 1024, ...) for all four kinds of Williams primes and for Carol/Kynea primes? (I know that this is not true for generalized Cullen/Woodall primes)

 kar_bon 2019-06-18 19:33

This remark is still only a reminder, so not used on other bases than powers of 2. Therefore I will not update all perfect power bases for now.

Checking the source of such pages using another base you will find the automatic generation of the n-values.
To avoid double work all sequences should be listed as their normalized form so instead of searching [url='https://www.rieselprime.de/ziki/Williams_prime_MM_4']3*4^n-1[/url] search [url='https://www.rieselprime.de/ziki/Riesel_prime_3']3*2^n-1[/url] and generate the first sequence from that.
So if a new 3*2^n-1-prime is found, only this sequence has to be edited, the other seq. will updated on the fly:
- less editing
- less type errors

For now I've used this only for Carol/Kynea or Williams like primes, the latter are using Riesel- or Proth-type lists (base 2 only).
If I will include other bases this could be extended to forms like Williams PM 25.

I've extended the Williams template to show a grey base if it uses a normalized power of 2 base sequence like as used for Carol/Kynea.

 sweety439 2019-06-23 10:45

If a prime in the list is also other classes of primes, I think it should be in the notes, like the article [URL="https://www.rieselprime.de/ziki/Riesel_prime_3"]Riesel 3[/URL], the prime 3*2^6-1 is also Sophie Germain (since 3*2^7-1 is also prime), twin (since 3*2^6+1 is also prime), and Near-Woodall (since 3*2^6-1 = (5+1)*2^5-1), however, for the article [URL="https://www.rieselprime.de/ziki/Williams_prime_MP_5"]Williams MP 5[/URL], ALL such primes are also generalized Fermat (since all are also of the form x^2+1, since all such n's are even, and thus 4*5^n is a square), but current only 20462 and 70130 has this note.

Sophie Germain: prime k*2^n-1 such that k*2^(n+1)-1 is also prime
Sophie Germain of the second kind: prime k*2^n+1 such that k*2^(n+1)+1 is also prime
Cunningham chains with length r: k*2^n-1, k*2^(n+1)-1, ..., k*2^(n+r-1)-1 are all primes
Cunningham chains of the second kind with length r: k*2^n+1, k*2^(n+1)+1, ..., k*2^(n+r-1)+1 are all primes
Sophie Germain with order b: prime k*b^n-1 such that k*b^(n+1)-1 is also prime
Sophie Germain of the second kind with order b: prime k*b^n+1 such that k*b^(n+1)+1 is also prime
Cunningham chains with length r and order b: k*b^n-1, k*b^(n+1)-1, ..., k*b^(n+r-1)-1 are all primes
Cunningham chains of the second kind with length r and order b: k*b^n+1, k*b^(n+1)+1, ..., k*b^(n+r-1)+1 are all primes
Twin: k*b^n-1 and k*b^n+1 are both primes
Cullen base b: n*b^n+1
Woodall base b: n*b^n-1
Carol base b: (b^n-1)^2-2
Kynea base b: (b^n+1)^2-2
Williams MM base b: (b-1)*b^n-1
Williams MP base b: (b-1)*b^n+1
Williams PM base b: (b+1)*b^n-1
Williams PP base b: (b+1)*b^n+1
Near-Cullen/Woodall MM base b: (n-1)*b^n-1
Near-Cullen/Woodall MP base b: (n-1)*b^n+1
Near-Cullen/Woodall PM base b: (n+1)*b^n-1
Near-Cullen/Woodall PP base b: (n+1)*b^n+1
Fermat base b: b^(2^n)+1
Repunit base b: (b^n-1)/(b-1)
Wagstaff base b: (b^n+1)/(b+1)

All times are UTC. The time now is 08:17.