[QUOTE=R.D. Silverman;204408]Has the cofactor been checked?[/QUOTE]

I just checked. It is C4880

I also checked in the morning that the cofactor is composite. I just had to rush to give a lecture.

I will post the sigma when I get home in about 4 hours or so. The discovery was done by one of my home computers. I will also then have a look at the sites where information on this factor would like to be presented. I am aware of Wilfrid Keller's site - that is what I have used to get information on Fermat numbers. It also seems to make the top-10 ECM-factors of this year, as pointed out by Batalov.

Attacking F14 was motivated by the F19 factor found by David Bessel and the fact that it was the smallest Fermat number with no known factors. I have run some curves on other Fermat numbers but F14 has received the most cpu-hours.

[QUOTE=rajula;204413]I also checked in the morning that the cofactor is composite. I just had to rush to give a lecture.

I will post the sigma when I get home in about 4 hours or so. The discovery was done by one of my home computers. I will also then have a look at the sites where information on this factor would like to be presented. I am aware of Wilfrid Keller's site - that is what I have used to get information on Fermat numbers. It also seems to make the top-10 ECM-factors of this year, as pointed out by Batalov.

Attacking F14 was motivated by the F19 factor found by David Bessel and the fact that it was the smallest Fermat number with no known factors. I have run some curves on other Fermat numbers but F14 has received the most cpu-hours.[/QUOTE]

Come visit us at [URL="http://www.fermatsearch.org"]FermatSearch[/URL]!

Luigi

[QUOTE=ET_;204417]Come visit us at [URL="http://www.fermatsearch.org"]FermatSearch[/URL]!

Luigi[/QUOTE]

Just visited - and seen that the News hasn't made it to the homepage yet. :wink:


(Sorry, I don't want to be impatient - I just could not resist after the "come visit us..."-posting :wink: )

[QUOTE=Andi47;204419]Just visited - and seen that the News hasn't made it to the homepage yet. :wink:


(Sorry, I don't want to be impatient - I just could not resist after the "come visit us..."-posting :wink: )[/QUOTE]

Yup... I'm still at work, and have no access to my webserver from here...
Be assured that the news will have their place set before tomorrow.

When I wrote "Come visit us" I intended to show another link related to fermat factors search. :smile:

[quote=rajula;204389]

F14 has a factor
[SIZE=2]116928085873074369829035993834596371340386703423373313

I found this after approximately 750 curves ran on F14 with B1=110M.

Tapio Rajala
Department of Mathematics and Statistics
University of Jyväskylä
Finland
[/SIZE][/quote]


Congratulations!

Here is the sigma from results.txt

ECM found a factor in curve #1, stage #2
Sigma=8585974330888598, B1=110000000, B2=11000000000.
UID: rajula/desktop_1, F14 has a factor: 116928085873074369829035993834596371340386703423373313

Now that I have more time, I'll reply to some of the posts... Sorry for not doing so earlier.

[QUOTE=Raman;204396]
By the way, how long does it take up to run up one ECM curve upon F14, F20, F22, F24 with some B1, B2 values?
[/QUOTE]

One curve on F14 with B1=11e7 takes something around 10h on my Phenom II X4 905e running at 2.5GHz. I can't remember but I would guess a curve on F20 with B1=3e6 would take around 200h on the same machine?

[QUOTE=debrouxl;204401]
How much RAM do ECM curves at B1=11e7 on such a monster number eat up?
[/QUOTE]

Running ECM on numbers like F14 does not require a lot of memory. I have not tried crunching anything larger than F25, but I would guess (again just a guess) that those would merely laugh at my computers capacity.

[QUOTE=ET_;204417]
Come visit us at [URL="http://www.fermatsearch.org"]FermatSearch[/URL]!
[/QUOTE]

Actually, I had alredy stumbled on this site previously. Still, thank you for the kind invitation. :smile:

[QUOTE=rajula;204436]
Running ECM on numbers like F14 does not require a lot of memory. I have not tried crunching anything larger than F25, but I would guess (again just a guess) that those would merely laugh at my computers capacity.[/QUOTE]

I just finished one curve at F27, and am at 50% Stage2 on F28.

My PC is a i5 750 @2,67, with 8 GB of RAM and running Ubuntu 64 bit. F27 required approximately 150h with B1=50,000. Stage2 is a bit slower, even though I gave Prime95 6GB of RAM. The difference between 2GB and 6GB Stage2 is almost negligible for these monsters; and the GHz rewarded are far less than the amount of cycles used, IMO.

Luigi

Group order:
[ <2, 2>, <3, 1>, <53, 1>, <107, 1>, <3433, 1>, <37087, 1>, <110323, 1>, <128321, 1>, <1738307, 1>, <9338881, 1>, <74968979, 1>, <783277631, 1> ]

[QUOTE=rajula;204429]Here is the sigma from results.txt

ECM found a factor in curve #1, stage #2
Sigma=8585974330888598, B1=110000000, B2=11000000000.
UID: rajula/desktop_1, F14 has a factor: 116928085873074369829035993834596371340386703423373313[/QUOTE]

Interesting. This was done with Prime95? So I guess Prime95 allows for sigma values greater then 32 bits?