mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > Aliquot Sequences

Reply
 
Thread Tools
Old 2016-05-02, 10:33   #298
Drdmitry
 
Drdmitry's Avatar
 
Nov 2011

2×32×13 Posts
Default

I took over the last range which was not finished: 30000000 - 30500000 and now it is completed!

Now I can say that all odd aliquot cycles with the smallest term of 16 digits are known! Thanks very much to everyone involved. Without your help this project will never be finished.


To be absolutely rigorous, an independent double check is needed. I leave it to enthusiastic ancestors.
Attached Files
File Type: txt output_odd_10e16_30000001-30500000.txt (937 Bytes, 55 views)

Last fiddled with by Drdmitry on 2016-05-02 at 10:34
Drdmitry is offline   Reply With Quote
Old 2016-05-02, 15:45   #299
henryzz
Just call me Henry
 
henryzz's Avatar
 
"David"
Sep 2007
Cambridge (GMT)

2×5×569 Posts
Default

I would suggest thinking about how we can boincify 17 digits and possibly double check 16 digits with it.
henryzz is offline   Reply With Quote
Old 2016-05-03, 08:44   #300
AndrewWalker
 
AndrewWalker's Avatar
 
Mar 2015
Australia

10100102 Posts
Default

Well done! Do you think anything would be achievable for even numbers, such as restricting it to numbers with small factors? Would probably be too hard to do all of them at this stage.
AndrewWalker is offline   Reply With Quote
Old 2016-05-03, 12:56   #301
Drdmitry
 
Drdmitry's Avatar
 
Nov 2011

2×32×13 Posts
Default What is next?

In principle, there are two natural extensions of the project.
  • To look for all aliquot cycles such that the number, preceding the biggest number in the cycle is even, up to 14 digits long.
  • To look for all aliquot cycles with the minimal element odd and up to 17 digits.

The first project will by slightly less time consuming then the current project So it can definitely be accomplished in the reasonable time (less than a year). However it will interfere with EdH search. He mostly looked for cycles of length four, and our potential project will cover cycles of any length, so it still makes sense to perform it. I expect additional 2 - 3 cycles here.

The second project will be much longer. I am not sure that it can be accomplished in a reasonable time (at least with the resources used for the current project). Additional problem here is that the amount of sequences which need to be dealt individually will rapidly grow. In the current project we had about 15 of them, but for 17d sequences there will be thousands of them. This is because 17d numbers are quite close to 64 bit numbers capacity,

Last fiddled with by Drdmitry on 2016-05-03 at 14:49
Drdmitry is offline   Reply With Quote
Old 2016-05-03, 14:15   #302
henryzz
Just call me Henry
 
henryzz's Avatar
 
"David"
Sep 2007
Cambridge (GMT)

10110001110102 Posts
Default

It should be possible to add something like gmp to the program which would allow extension beyond 64-bits.
The even search is going to need exhaustedly extending at some point. How is now different to later?

I would hope that we can add resources if we boincify the project.
henryzz is offline   Reply With Quote
Old 2016-05-11, 02:17   #303
EdH
 
EdH's Avatar
 
"Ed Hall"
Dec 2009
Adirondack Mtns

62648 Posts
Default

Quote:
Originally Posted by Drdmitry View Post
In principle, there are two natural extensions of the project.
  • To look for all aliquot cycles such that the number, preceding the biggest number in the cycle is even, up to 14 digits long.
  • To look for all aliquot cycles with the minimal element odd and up to 17 digits.

The first project will by slightly less time consuming then the current project So it can definitely be accomplished in the reasonable time (less than a year). However it will interfere with EdH search. He mostly looked for cycles of length four, and our potential project will cover cycles of any length, so it still makes sense to perform it. I expect additional 2 - 3 cycles here.

The second project will be much longer. I am not sure that it can be accomplished in a reasonable time (at least with the resources used for the current project). Additional problem here is that the amount of sequences which need to be dealt individually will rapidly grow. In the current project we had about 15 of them, but for 17d sequences there will be thousands of them. This is because 17d numbers are quite close to 64 bit numbers capacity,
My machines are currently shut down and will be for at least several more weeks. I'm not sure what direction I will take when I start them back up. I might move to some other factoring projects. Please feel free to run that area. I found quite a few 4-cycles and am happy with those, but I also realize that only searching for 4-cycles is very inefficient overall.
EdH is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Aliquot cycles search, the state of the art. Drdmitry Aliquot Sequences 124 2017-07-02 02:48
Search of all even-15-digit Aliquot cycles Drdmitry Aliquot Sequences 25 2016-12-16 15:26
Is a search for aliquot 3-cycles feasible? schickel Aliquot Sequences 7 2013-02-08 01:33
BOINCify Aliquot searching? schickel Aliquot Sequences 23 2011-05-16 23:13
Jan Munch Pedersen's Tables of Aliquot Cycles R. Gerbicz Math 0 2010-07-01 12:30

All times are UTC. The time now is 17:44.

Mon Aug 3 17:44:41 UTC 2020 up 17 days, 13:31, 1 user, load averages: 1.25, 1.31, 1.44

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.