mersenneforum.org  

Go Back   mersenneforum.org > Factoring Projects > GMP-ECM

Reply
 
Thread Tools
Old 2006-12-16, 11:49   #1
M0CZY
 
M0CZY's Avatar
 
May 2005
Brutal Police State, UK

112 Posts
Default GMP-ECM and the Cunningham Input List

Forgive me if I am posting in the wrong thread.

Earlier this year I ran GMP-ECM on the Cunningham input list, using B1= 11e6,
just like it says on the web page http://www.loria.fr/~zimmerma/ecmnet/
It took me 4 1/2 months, and no factors were found.

Question. Is it likely that there are still factors to be found in the list using
B1=11e6, or am I wasting my time?

Should I do another run using B1=11e6, or should I be using B1=43e6?

I would consider doing another run using B1=11e6 if it was thought that there
were still factors waiting to be discovered using this value of B1.

I don't think I can do a run using B1=43e6, as I would only be able to do one
curve per day, and a full run on the Cunningham input list would take me
over two years!
M0CZY is offline   Reply With Quote
Old 2006-12-16, 12:14   #2
akruppa
 
akruppa's Avatar
 
"Nancy"
Aug 2002
Alexandria

246710 Posts
Default

I'd say B1=11M is suboptimal now. There's a small chance that even that will find a factor, but higher B1, say B1=44M or even 110M, will perform better.

I'll ask Paul to update the page.

Alex

Edit: Note that finding any Cunningham factors with ECM is very hard these days. A lot of ECM has been done already, largely thanks to Bruce Dodson, and the easy pickings a long gone.

Last fiddled with by akruppa on 2006-12-16 at 12:17
akruppa is offline   Reply With Quote
Old 2006-12-16, 14:16   #3
M0CZY
 
M0CZY's Avatar
 
May 2005
Brutal Police State, UK

112 Posts
Default

OK, thanks for the reply.
I don't think that I can help out with the ECMNET project any further, as my
computer is too old and slow to run curves with a higher value of B1.
M0CZY is offline   Reply With Quote
Old 2006-12-16, 23:00   #4
jasong
 
jasong's Avatar
 
"Jason Goatcher"
Mar 2005

5×701 Posts
Default

Quote:
Originally Posted by M0CZY View Post
OK, thanks for the reply.
I don't think that I can help out with the ECMNET project any further, as my
computer is too old and slow to run curves with a higher value of B1.
I really hope this doesn't mean you're going to quit Distributed Computing. I know you didn't say you were, I'm just saying I hope you aren't.

If you want another project to run, may I suggest http://www.distributedcomputing.info for research? If you're still interested in ecm, there are other project possibilites, like the Odd Perfect Number search.
jasong is offline   Reply With Quote
Old 2006-12-19, 14:11   #5
M0CZY
 
M0CZY's Avatar
 
May 2005
Brutal Police State, UK

1718 Posts
Default

No, I am not giving up on distributed computing.
I am simply pointing out that many projects nowadays, like ECM, are much
better done by faster, more modern computers than my old, nearly obsolete
machine.
Just this morning I started doing a factoring assignment for GIMPS, which
should take me about a month to complete!
M0CZY is offline   Reply With Quote
Old 2006-12-19, 14:40   #6
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

22×5×373 Posts
Default

Quote:
Originally Posted by M0CZY View Post
No, I am not giving up on distributed computing.
I am simply pointing out that many projects nowadays, like ECM, are much
better done by faster, more modern computers than my old, nearly obsolete
machine.
Just this morning I started doing a factoring assignment for GIMPS, which
should take me about a month to complete!
Actually, I bemoan the fact that the bar is set very high for any newcomer
to existing factoring projects.

The Cunningham project requires a large amount of resources to run
ECM or NFS succesfully on any of the remaining composites. Indeed,
I no longer have sufficient machines to perform Cunningham NFS
factorizations.

This tends to discourage new participants. Which is one reason that I
have been promoting the homogeneous Cunningham project. These
have not been attempted with ECM to the same extent as the Cunningham
project and results are obtainable for someone with only a single PC.
R.D. Silverman is offline   Reply With Quote
Old 2006-12-19, 18:16   #7
masser
 
masser's Avatar
 
Jul 2003
wear a mask

1,483 Posts
Default

Quote:
Originally Posted by R.D. Silverman View Post
Actually, I bemoan the fact that the bar is set very high for any newcomer
to existing factoring projects.

The Cunningham project requires a large amount of resources to run
ECM or NFS succesfully on any of the remaining composites. Indeed,
I no longer have sufficient machines to perform Cunningham NFS
factorizations.

This tends to discourage new participants. Which is one reason that I
have been promoting the homogeneous Cunningham project. These
have not been attempted with ECM to the same extent as the Cunningham
project and results are obtainable for someone with only a single PC.
Here's the link to some information on the Homogeneous Cunningham project:

http://www.leyland.vispa.com/numth/f.../anbn/main.htm

trolling around...
masser
masser is offline   Reply With Quote
Old 2006-12-19, 18:39   #8
xilman
Bamboozled!
 
xilman's Avatar
 
"π’‰Ίπ’ŒŒπ’‡·π’†·π’€­"
May 2003
Down not across

101000100110012 Posts
Default

Quote:
Originally Posted by masser View Post
Here's the link to some information on the Homogeneous Cunningham project:

http://www.leyland.vispa.com/numth/f.../anbn/main.htm

trolling around...
masser
Indeed. And, as I said in another thread, as soon as the number of remaining composites on that page falls to 30 a new batch will be added. Some of those newcomers will surely have factors that are very easily found with ECM.

Paul
xilman is offline   Reply With Quote
Old 2006-12-19, 19:00   #9
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

22×5×373 Posts
Default

Quote:
Originally Posted by xilman View Post
Indeed. And, as I said in another thread, as soon as the number of remaining composites on that page falls to 30 a new batch will be added. Some of those newcomers will surely have factors that are very easily found with ECM.

Paul
I will finish 3,2,379-, 379+, and 389- by the end of this week.
This will finish 3,2 to exponent 400.

I will then do 6,5,218+ and 6,5,223+.

There still remain some numbers with exponent divisible by 5 that are
*very* easy among the remaining current composites.
R.D. Silverman is offline   Reply With Quote
Old 2006-12-21, 13:59   #10
antiroach
 
antiroach's Avatar
 
Jun 2003

22·61 Posts
Default

Is there an organized effort to factors these Homogeneous Cunningham numbers? I'd like to know at least at what B1 level factoring should start and approximately how many curves to do at that level.
antiroach is offline   Reply With Quote
Old 2006-12-21, 14:13   #11
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

22×5×373 Posts
Default

Quote:
Originally Posted by antiroach View Post
Is there an organized effort to factors these Homogeneous Cunningham numbers? I'd like to know at least at what B1 level factoring should start and approximately how many curves to do at that level.
It is not an organized effort. The numbers have been tested up to about
40 digits.
R.D. Silverman is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Recommendations for cutting down input fivemack Msieve 1 2017-12-12 17:22
GPU GMP-ECM to higher input limit wombatman Factoring 5 2014-06-14 04:44
Prime hunters: I need your input :) opyrt Prime Sierpinski Project 6 2009-12-28 17:42
ecm_factor returning the same number as input Capone GMP-ECM 17 2007-06-17 09:19
segfault with large input on Opteron sean GMP-ECM 2 2005-08-15 09:59

All times are UTC. The time now is 07:11.

Wed Dec 2 07:11:42 UTC 2020 up 83 days, 4:22, 1 user, load averages: 1.53, 1.65, 1.55

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.