mersenneforum.org  

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

Reply
 
Thread Tools
Old 2007-01-09, 05:59   #1
bdodson
 
bdodson's Avatar
 
Jun 2005
lehigh.edu

210 Posts
Exclamation Top 10 GMP-ECM for the year

Anyone that felt there wasn't a chance at making the top-10 ECM
list, as being too difficult in 2006 might consider posting an entry
on the brand new 2007 list. Well, almost new, just a p46 and
a p50 so far; with eight free slots --- a p10 makes the list!
Actually, Paul probably won't appreciate a flood of 8 p1x's, but
seriously, here's your chance to make the list with a p35-p40
from one of the servers (running gmp-ecm?) on less tested numbers.

Bruce (http://www.loria.fr/~zimmerma/records/ecmnet.html
for the top 10 of 2007, only the p50 listed, with a p46 Cunningham
report pending http://www.loria.fr/~zimmerma/cgi-bin/last.cgi?date)

Last fiddled with by akruppa on 2007-01-09 at 08:48 Reason: fixed url
bdodson is offline   Reply With Quote
Old 2007-01-10, 15:34   #2
bdodson
 
bdodson's Avatar
 
Jun 2005
lehigh.edu

40016 Posts
Default update, free slots going fast!

Quote:
Originally Posted by bdodson View Post
... consider posting an entry on the brand new 2007 list.
Well, almost new... with eight free slots ...Bruce
Thanks to Alex for fixing the url! And congratualtions for
grabbing a slot with a p43 from 823^577-1. CWI has the
new top entry, a p51, also a non-Cunningham. Today's p46
(the 2nd from our condor xp grid) takes the 3rd slot away from
the earlier p46, pushing Alex's p43 down to 5th (sorry ).
-Bruce
bdodson is offline   Reply With Quote
Old 2007-01-15, 15:12   #3
bdodson
 
bdodson's Avatar
 
Jun 2005
lehigh.edu

210 Posts
Default get'm while they're (still) cheap!

Quote:
Originally Posted by bdodson View Post
... with eight free slots --- a p10 makes the list!
Well, that went quicker than I expected. Lehigh turned out to be
unusually idle over the semester break; and, as reported in the
Cunningham Tables thread, CWI decided to join the "competition"
with "all of our machines". So it's not a surprize that CWI grabbed
the last two free spots on the top10 ecm for 2007 (that is, gmp-ecm).
So why am I (still) occupying bandwith here? Good Question!

News Flash: prelims are over!

While it is true that there are no longer free slots, there are
still (at least) four relatively inexpensive slots. The first 10 factors
reported included 3 at 50-or-more digits, and 3 with 49-digits. Next
is a single 47-digit prime factor. While Alex's 43-digit factor won't
last (having been bumped to 10th of the ten slots), a 48-digit or
large 47-digit factor has a plausible chance of staying on the list
for a while. (Depending, perhaps, on how long CWI stays interested.)
A prime factor of 50-52 digits has a good chance of staying on the
top10 list for a month or two.

Of course, the list is for the top10 _of 2007_ for which (judging by
2005 and 2006) one needs a factor of 56- or 55-digits. But one
point established by the ecm records is that persistently looking for
50-52 digit factors does occasionally pick up larger factors --- for
example, the 2nd largest of 2006, with 61-digits, was found with
(first step limit) B1 = 43M, optimal for finding 50-digit factors.
Besides which, everyone factoring anything other than Cunningham
numbers might notice that we managed to bump all non-Cunninghams
out of the top10 in 2006 [unlike 2005]; so perhaps you'll have better
luck this year! -Bruce
bdodson is offline   Reply With Quote
Old 2007-01-19, 15:31   #4
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

164448 Posts
Default

Quote:
Originally Posted by bdodson View Post
Well, that went quicker than I expected. Lehigh turned out to be

<snip>

News Flash: prelims are over!

Besides which, everyone factoring anything other than Cunningham
numbers might notice that we managed to bump all non-Cunninghams
out of the top10 in 2006 [unlike 2005]; so perhaps you'll have better
luck this year! -Bruce
CWI just found a 61 digit ECM factor from the Cunningham
extensions.
R.D. Silverman is offline   Reply With Quote
Old 2007-01-19, 15:53   #5
bdodson
 
bdodson's Avatar
 
Jun 2005
lehigh.edu

210 Posts
Default

Quote:
Originally Posted by R.D. Silverman View Post
CWI just found a 61 digit ECM factor from the Cunningham
extensions.
As I was saying; another Brent-M-teR number. Paul_Z reports
that the 2nd largest factor of the group order of the lucky curve
(random sigma value) is 201209201, so B1 > 201M, perhaps
likely B1=260M (p60 optimal); and largest factor 7140622896769,
so B2 (2nd step limit) > 7.14e12. These are larger than the primes
in the orders used for many of the previous p6x ecm factors.

That brings the stats to CWI: 7, bdodson: 3. As I was predicting
upon hearing Peter's declaration that CWI was joining the competition.
A certain amount of rivalry is good, productive, yes? -bdodson
bdodson is offline   Reply With Quote
Old 2007-01-19, 16:03   #6
Andi47
 
Andi47's Avatar
 
Oct 2004
Austria

2·17·73 Posts
Default

Quote:
Originally Posted by R.D. Silverman View Post
CWI just found a 61 digit ECM factor from the Cunningham
extensions.
Andi47 is offline   Reply With Quote
Old 2007-01-20, 15:07   #7
bdodson
 
bdodson's Avatar
 
Jun 2005
lehigh.edu

210 Posts
Default Update

Quote:
Originally Posted by bdodson View Post
While it is true that there are no longer free slots, there are
still (at least) four relatively inexpensive slots. ... a 48-digit or
large 47-digit factor has a plausible chance of staying on the list
for a while. (Depending, perhaps, on how long CWI stays interested.)
A prime factor of 50-52 digits has a good chance of staying on the
top10 list for a month or two.
As the number of views in this thread is higher than I expected,
perhaps some "reflective narative" on the current stats are in order
(or, at least, not-too-far out-of-order!). Things have moved rather
quickly, as CWI, has in fact remained interested. That p47 is now
in the 10th slot, and seems unlike to last much past the weekend
(if it isn't already gone, pending Willemien's reading of overnight
factors). The best that can be said towards keeping this competition
from entirely turning primarily into a spectator sport is that the
49-digit prime factors aren't too large, so that a p49 above 4.0*10^48
will remain above the four p49's on the current page (actually, 3.6 would
be above 3 of the 4).

Quote:
... everyone factoring anything other than Cunningham
numbers might notice that we managed to bump all non-Cunninghams
out of the top10 in 2006 [unlike 2005]; so perhaps you'll have better
luck this year! -Bruce
Just in case Bob's post was too subtle ( ... Uh, am I permitted a
grin here? ... ) , the above jab spectacularly missed the mark,
unless anyone imagines that the new p61, coming from 77^103-1,
won't last the year. Actually, the new p6x coming so early in the
new year is Really, Really good news, seems to me. We now have
eight, and Richard Brent's champs list for all-time, all-program ecm
factors has just two p59's as the last p5x remaining (this list being
restricted to prime factors from large composites, not easier by
Msieve, et. al.). As we get nearer to 10 p6x's, past experience would
suggest that a p7x would no longer be premature. On to 256-bit
ecm factors! -Bruce
bdodson is offline   Reply With Quote
Old 2007-01-24, 16:06   #8
bdodson
 
bdodson's Avatar
 
Jun 2005
lehigh.edu

102410 Posts
Default

Quote:
Originally Posted by bdodson View Post
As I was saying; another Brent-M-teR number.

That brings the stats to CWI: 7, bdodson: 3. -bdodson
Pending an update on today's CWI factors, the stats (as of what
I know, at the moment) are

bdodson: 5, CWI: 4, R.Hooft: 1. Of course, I'd gladly trade a couple
small p5x's for a p6x; so perhaps we're tied, or so. At last we have
a top10 factor that's neither Cunningham nor on the Brent extension;
Hooft's p50 has base 337, beyond the Cunningham limit of 12, and
the Brent limit of 99. -Bruce
bdodson is offline   Reply With Quote
Old 2007-01-24, 16:31   #9
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

22×5×373 Posts
Default

Quote:
Originally Posted by bdodson View Post
Pending an update on today's CWI factors, the stats (as of what
I know, at the moment) are

bdodson: 5, CWI: 4, R.Hooft: 1. Of course, I'd gladly trade a couple
small p5x's for a p6x; so perhaps we're tied, or so. At last we have
a top10 factor that's neither Cunningham nor on the Brent extension;
Hooft's p50 has base 337, beyond the Cunningham limit of 12, and
the Brent limit of 99. -Bruce
Are these stats specifically restricted to GMP-ECM, or do they comprise
any ECM results?

From the Fibonacci/Lucas tables, Blair Kelly has found a 51 digit factor of
L1853, a 51 digit factor of F1605, and a 52 digit factor of F1506.....

Should these not be included????
R.D. Silverman is offline   Reply With Quote
Old 2007-01-24, 16:34   #10
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

22×5×373 Posts
Default

Quote:
Originally Posted by R.D. Silverman View Post
Are these stats specifically restricted to GMP-ECM, or do they comprise
any ECM results?

From the Fibonacci/Lucas tables, Blair Kelly has found a 51 digit factor of
L1853, a 51 digit factor of F1605, and a 52 digit factor of F1506.....

Should these not be included????
Update:

Blair's web pages states that the results are from GMP-ECM.....
R.D. Silverman is offline   Reply With Quote
Old 2007-01-24, 16:36   #11
xilman
Bamboozled!
 
xilman's Avatar
 
"π’‰Ίπ’ŒŒπ’‡·π’†·π’€­"
May 2003
Down not across

22×3×11×83 Posts
Default

Quote:
Originally Posted by R.D. Silverman View Post
Update:

Blair's web pages states that the results are from GMP-ECM.....
Paul Z. only reports factors of which he's been told, preferably by the finder.


Paul
xilman is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
It's been a year tcharron Information & Answers 8 2014-01-29 20:16
Year Over Year TF Progress petrw1 Factoring 3 2013-03-20 19:34
New Year's Resolutions davar55 Lounge 19 2010-12-27 19:10
What year is it? E_tron Lounge 3 2004-12-31 13:43
1 Year QuintLeo Lounge 14 2003-11-14 07:56

All times are UTC. The time now is 04:45.


Mon Oct 25 04:45:44 UTC 2021 up 93 days, 23:14, 0 users, load averages: 1.23, 1.11, 1.09

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, 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.