64 bit linux 11e siever binary?

axn · 2009-08-30, 15:56

Where can I get 11e siever binary for 64 bit linux?

fivemack · 2009-08-30, 17:28

http://www.boo.net/~jasonp/lasieveLinuxAsm64.zip

axn · 2009-08-31, 03:37

Quote:

Originally Posted by The Page

Not Found

The requested URL /~jasonp/lasieveLinuxAsm64.zip was not found on this server.
Apache/2.0.48 (Unix) FrontPage/5.0.2.2634 PHP/4.3.4 Server at www.boo.net Port 80

Batalov · 2009-08-31, 05:14

Try ftp://fivemack.dyndns.org/M877/gnfs-lasieve4I11e.zip

Arrgh. It's there, but you will not get it. (Maybe Tom will fix it for you.)

fivemack · 2009-08-31, 09:44

Try http://www.chiark.greenend.org.uk/~t...sm64latest.zip

axn · 2009-08-31, 16:48

Quote:

Originally Posted by fivemack

Try http://www.chiark.greenend.org.uk/~t...sm64latest.zip

Much obliged. I'll do some trials with it and see if it works out. Thanks.

henryzz · 2009-08-31, 18:07

Quote:

Originally Posted by axn

Much obliged. I'll do some trials with it and see if it works out. Thanks.

i did a c91 with it yesterday evening and i got a largish duplication rate

BTW i did the same c91 more than once over the last couple of days and for 25bit large primes it needed 1.5M relations and for 22bit it needed 0.5M
why? shouldn't 25 need 8x the number of relations as 22
these were both with 12e

fivemack · 2009-08-31, 22:13

Basically, if the large-prime bound is much too large, you get enough relations to match up enough of the small large primes to make a matrix before you get any relevant number of paired relations for the large large primes; effectively you're running with a smaller large-prime bound having wasted an awful lot of your relations at the singleton-removal stage.

henryzz · 2009-09-01, 07:22

Quote:

Originally Posted by fivemack

Basically, if the large-prime bound is much too large, you get enough relations to match up enough of the small large primes to make a matrix before you get any relevant number of paired relations for the large large primes; effectively you're running with a smaller large-prime bound having wasted an awful lot of your relations at the singleton-removal stage.

but the 22 took 50 minutes and the 25 took 40 minutes
maybe 22 is slightly too low even though sieving tests said i should use it
looks like its time for more runs

2009-08-30, 15:56	#1
axn Jun 2003 5082₁₀ Posts	64 bit linux 11e siever binary? Where can I get 11e siever binary for 64 bit linux?

2009-08-31, 05:14	#4
Batalov "Serge" Mar 2008 Phi(4,2^7658614+1)/2 10010100010110₂ Posts	Try ftp://fivemack.dyndns.org/M877/gnfs-lasieve4I11e.zip Arrgh. It's there, but you will not get it. (Maybe Tom will fix it for you.) Last fiddled with by Batalov on 2009-08-31 at 05:17

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
QS Lattice Siever	R.D. Silverman	Factoring	31	2018-10-08 17:17
Adventures with 16f siever	VBCurtis	Factoring	6	2018-01-24 11:06
A siever for K (b, n, c fixed)?	pepi37	Software	7	2015-07-10 04:42
GGNFS SVN 374 with 11e siever: Speedup for c85-95!	Andi47	Aliquot Sequences	0	2009-11-02 16:41
Let's fix the elusive siever bug	Batalov	Factoring	4	2008-10-18 12:43

2009-08-30, 17:28	#2
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 3×2,141 Posts	http://www.boo.net/~jasonp/lasieveLinuxAsm64.zip

2009-08-31, 09:44	#5
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 3×2,141 Posts	Try http://www.chiark.greenend.org.uk/~t...sm64latest.zip

2009-08-31, 22:13	#8
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 3·2,141 Posts	Basically, if the large-prime bound is much too large, you get enough relations to match up enough of the small large primes to make a matrix before you get any relevant number of paired relations for the large large primes; effectively you're running with a smaller large-prime bound having wasted an awful lot of your relations at the singleton-removal stage.