mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > Aliquot Sequences

Reply
 
Thread Tools
Old 2017-08-28, 15:09   #1
garambois
 
garambois's Avatar
 
Oct 2011

24·3·5 Posts
Question Factordb and aliquot sequences with useless size terms

I recently worked on aliquot sequences with record size terms.
The works are visible on this page, but sorry, in french : http://www.aliquotes.com/suite_terme...le_record.html

Brief explanations of the principle before my two questions about factordb :

It is quite easy to calculate several hundred terms of a strictly increasing aliquot sequence, each of those terms has more than 10 ^ 7 digits.

If p is a prime and M=2^p-1 is a prime too (a Mersenne prime), N = 2^(p-1) * M, and N is a perfect number.
So we have an aliquot sequence which starts on the number :

j0 = N * z0 = 2^(p-1) * M * z0.

PGCD(M,z0)=1 because z0<M and M prime.

So, j1 = σ(j0) - j0 = σ(N * z0) - N * z0 = σ(N) * σ(z0) - N * z0 = 2 * N * σ(z0) - N * z0 = N (2 σ(z0) - z0) = N * z1.

As long as zi <M we have PGCD(M,zi)=1
We note that z1 = 2 σ (z0) - z0 and that, more generally, we will have z i+1 = 2 σ (zi) - zi.
We note that z i+1> zi as long as zi<M.

If N has millions of digits, then the terms of the aliquot sequence have millions of digits.

I have 2 questions about factordb.


1) Question 1.

It is possible to enter this aliquot sequence on factordb :

0 : 2^520 * (2^521-1) * 3 = 2^520 * (2^521-1) * 3^1
1 : 2^520 * (2^521-1) * 5 = 2^520 * (2^521-1) * 5^1
2 : 2^520 * (2^521-1) * 7 = 2^520 * (2^521-1) * 7^1
3 : 2^520 * (2^521-1) * 9 = 2^520 * (2^521-1) * 3^2
4 : 2^520 * (2^521-1) * 17 = 2^520 * (2^521-1) * 17^1
5 : 2^520 * (2^521-1) * 19 = 2^520 * (2^521-1) * 19^1
6 : 2^520 * (2^521-1) * 21 = 2^520 * (2^521-1) * 3^1 * 7^1
7 : 2^520 * (2^521-1) * 43 = 2^520 * (2^521-1) * 43^1
8 : 2^520 * (2^521-1) * 45 = 2^520 * (2^521-1) * 3^2 * 5^1
9 : 2^520 * (2^521-1) * 111 = 2^520 * (2^521-1) * 3^1 * 37^1
10 : 2^520 * (2^521-1) * 193 = 2^520 * (2^521-1) * 193^1

But it is not possible to enter this aliquot sequence on factordb :

0 : 2^74207281 * (2^74207281-1) * 3 = 2^74207281 * (2^74207281-1) * 3^1
1 : 2^74207281 * (2^74207281-1) * 5 = 2^74207281 * (2^74207281-1) * 5^1
2 : 2^74207281 * (2^74207281-1) * 7 = 2^74207281 * (2^74207281-1) * 7^1
3 : 2^74207281 * (2^74207281-1) * 9 = 2^74207281 * (2^74207281-1) * 3^2
4 : 2^74207281 * (2^74207281-1) * 17 = 2^74207281 * (2^74207281-1) * 17^1
5 : 2^74207281 * (2^74207281-1) * 19 = 2^74207281 * (2^74207281-1) * 19^1
6 : 2^74207281 * (2^74207281-1) * 21 = 2^74207281 * (2^74207281-1) * 3^1 * 7^1
7 : 2^74207281 * (2^74207281-1) * 43 = 2^74207281 * (2^74207281-1) * 43^1
8 : 2^74207281 * (2^74207281-1) * 45 = 2^74207281 * (2^74207281-1) * 3^2 * 5^1
9 : 2^74207281 * (2^74207281-1) * 111 = 2^74207281 * (2^74207281-1) * 3^1 * 37^1
10 : 2^74207281 * (2^74207281-1) * 193 = 2^74207281 * (2^74207281-1) * 193^1

I suppose the numbers are too big for factordb and factordb does not seem to know the prime number of mersenne (2^74207281-1), that's right ?

Aliquot sequences with more than 700 terms available here :

http://www.aliquotes.com/parfait_521_3.txt

And with terms with more than 44 millions digits (!!!)

http://www.aliquotes.com/parfait_74207281_3.txt

And on factordb, the limit (130100 digits) :

http://factordb.com/sequences.php?se...20&fr=0&to=100


2) Question 2.

If we look at the 3 previous aliquot sequences (2^520 * (2^521-1) * 3 and 2^74207280 * (2^74207281-1) * 3 and 2^216090 * (2^216091-1) * 3), we note that the sequence of the numbers zi are identical : only the perfect number N changes.

If I enter the first aliquot sequence in factordb, (see here : http://factordb.com/sequences.php?se...20&fr=0&to=100), the database "knows" all the prime numbers of the decompositions of the cofactors zi.
Then why if I enter another starting number in factordb, like 2 ^ 606 * (2 ^ 607-1) * 3, the database does not indicate the terms of the aliquot sequence whith the same cofactors zi ?
garambois is offline   Reply With Quote
Old 2017-08-28, 17:43   #2
firejuggler
 
firejuggler's Avatar
 
Apr 2010
Over the rainbow

93016 Posts
Default

The db has no idea of what a perfect number is, or what to do with it

Last fiddled with by firejuggler on 2017-08-28 at 17:44
firejuggler is offline   Reply With Quote
Old 2017-08-29, 07:16   #3
garambois
 
garambois's Avatar
 
Oct 2011

24×3×5 Posts
Default

Quote:
Originally Posted by firejuggler View Post
The db has no idea of what a perfect number is, or what to do with it

I assumed that it would be possible for db "to know" the Mersenne prime numbers in their "condensed" form.
For example, in the form :
2 ^ 74207281 - 1
Instead of that form which is written with more than 44 million digits :
451129962706 ... 557930315776
And especially that db is not obliged to check again the primality of this huge number first already known and famous.

But it must be very difficult to program things like that.


garambois is offline   Reply With Quote
Old 2017-08-29, 07:28   #4
Dubslow
Basketry That Evening!
 
Dubslow's Avatar
 
"Bunslow the Bold"
Jun 2011
40<A<43 -89<O<-88

160658 Posts
Default

The factordb doesn't know anything about the theory behind aliquot sequences, and in particular doesn't know anything about how perfect numbers with "small" cofactors all have the same cofactor sequence. All it knows is that there's a whole bunch of big numbers, and it doesn't try to factor or otherwise test the big numbers.
Dubslow is offline   Reply With Quote
Old 2017-08-30, 08:04   #5
garambois
 
garambois's Avatar
 
Oct 2011

24·3·5 Posts
Default

OK, thank you for your answers !
garambois is offline   Reply With Quote
Old 2017-08-30, 08:11   #6
fivemack
(loop (#_fork))
 
fivemack's Avatar
 
Feb 2006
Cambridge, England

2·52·127 Posts
Default

Quote:
Originally Posted by garambois View Post
I recently worked on aliquot sequences with record size terms.
Given that the cofactors are the same whatever perfect number you're looking at (at least until you start getting the Mersenne prime as an accidental cofactor), why not compute the series of cofactors explicitly and just label the perfect number something symbolic?
fivemack is offline   Reply With Quote
Old 2017-08-30, 11:03   #7
garambois
 
garambois's Avatar
 
Oct 2011

24×3×5 Posts
Default

This is exactly what I did to get for example this aliquot sequence whose terms have more than 44677235 digits in base 10 :

http://www.aliquotes.com/parfait_74207281_3.txt

My problem is that I can not inform the database factordb with this aliquot sequence because factordb does not know the "symbolic" representation of the large prime numbers of Mersenne.
garambois is offline   Reply With Quote
Old 2017-08-30, 11:47   #8
science_man_88
 
science_man_88's Avatar
 
"Forget I exist"
Jul 2009
Dumbassville

26×131 Posts
Default

Quote:
Originally Posted by garambois View Post
This is exactly what I did to get for example this aliquot sequence whose terms have more than 44677235 digits in base 10 :

http://www.aliquotes.com/parfait_74207281_3.txt

My problem is that I can not inform the database factordb with this aliquot sequence because factordb does not know the "symbolic" representation of the large prime numbers of Mersenne.
the real problem, is that it expands the numbers. it technically knows the mersenne primes up to the 35th but can't learn any more because the rest are over the expanded limit of 10 million characters. you can type any form into factordb search and it will find number of that form in theory. just not over a certain length.
science_man_88 is offline   Reply With Quote
Old 2017-08-31, 19:03   #9
garambois
 
garambois's Avatar
 
Oct 2011

24×3×5 Posts
Default

Quote:
Originally Posted by science_man_88 View Post
the real problem, is that it expands the numbers. it technically knows the mersenne primes up to the 35th but can't learn any more because the rest are over the expanded limit of 10 million characters. you can type any form into factordb search and it will find number of that form in theory. just not over a certain length.

OK, thank you !

You are right, I tried with success for the 35th Mersenne prime.
But no success for the 36th Mersenne prime !

I didn't know.

2^1398268*(2^1398269-1)*3 : SUCCESS

2^2976220*(2^2976221-1) * 3 : NO SUCCESS
garambois is offline   Reply With Quote
Old 2017-09-01, 00:06   #10
Batalov
 
Batalov's Avatar
 
"Serge"
Mar 2008
Phi(3,3^1118781+1)/3

100011010111112 Posts
Exclamation

Quote:
Originally Posted by garambois View Post
If we look at the 3 previous aliquot sequences (2^520 * (2^521-1) * 3 and 2^74207280 * (2^74207281-1) * 3 and 2^216090 * (2^216091-1) * 3), we note that the sequence of the numbers zi are identical : only the perfect number N changes.
Of course they are identical. It is a bloody trivial result.

By uploading uselessly large algebraically trivial results into factordb, you are sabotaging users who use factordb for proper reasons.
Batalov is offline   Reply With Quote
Old 2017-09-02, 00:21   #11
garambois
 
garambois's Avatar
 
Oct 2011

F016 Posts
Default

Quote:
Originally Posted by Batalov View Post
Of course they are identical. It is a bloody trivial result.

By uploading uselessly large algebraically trivial results into factordb, you are sabotaging users who use factordb for proper reasons.
Sorry, that was not my intention at all : I was not aware of it !
garambois is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Broken aliquot sequences fivemack FactorDB 45 2020-05-16 15:22
Graphs broken for aliquot sequences on factordb fivemack Aliquot Sequences 4 2018-03-23 11:54
Broken aliquot sequences schickel FactorDB 18 2013-06-12 16:09
Aliquot sequence worker for factordb yoyo FactorDB 6 2012-01-12 20:58
New article on aliquot sequences schickel mersennewiki 0 2008-12-30 07:07

All times are UTC. The time now is 17:56.

Fri Jul 3 17:56:52 UTC 2020 up 100 days, 15:29, 2 users, load averages: 0.79, 1.23, 1.42

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.