[QUOTE=lorgix;307669]I don't have the right tools.

10000 of them would still be a pretty small file.

My #1 wish is still that the db gets better at handling numbers with algebraic factors.[/QUOTE]
Why do you want so much and of which size?
Just download some (1 - 10) factor them, upload them and download again. If you download much the probability raises that someone else factored it in the meantime.
yoyo

[QUOTE=yoyo;307670]Why do you want so much and of which size?
Just download some (1 - 10) factor them, upload them and download again. If you download much the probability raises that someone else factored it in the meantime.
yoyo[/QUOTE]
I was thinking I'd just do a little ECM, because many of the composites have small factors.

In this case use this Perl script
[url]http://www.rechenkraft.net/wiki/index.php?title=Benutzer_Diskussion:Yoyo/factordb#yafu.pl[/url]
together with the yafu, ecm and ggnfs binaries.
It fetches a random small composites, factors it and uploads the factors and fetches the next one.
yoyo

That script completely factors a number. I beleive he is talking about using a small a amount of ecm on lots of numbers to take out those with easy factors. I modified the script to do that. I should hopefully be able to post it when I get home tomorrow. It uses the gmp-ecm built into yafu. About 1/4 of numbers were factored when I first started running it. This went down to maybe 1/8 in a few hours.

[QUOTE=henryzz;307683]That script completely factors a number. I beleive he is talking about using a small a amount of ecm on lots of numbers to take out those with easy factors. I modified the script to do that. I should hopefully be able to post it when I get home tomorrow. It uses the gmp-ecm built into yafu. About 1/4 of numbers were factored when I first started running it. This went down to maybe 1/8 in a few hours.[/QUOTE]
Here is my modified script for ecm.
It has a bug somewhere and crashes/freezes early on sometimes. It only seems to effect early on so just restart it. I have run it for several hours successfully. In addtion if you get a prp*comp on the first number it will crash.

it work... getting about 10% of the composite factored

[QUOTE=firejuggler;307711]it work... getting about 10% of the composite factored[/QUOTE]
That will only drop further. A high percentage of the numbers will have had 5 or more curves at 50k run on them now. I run batches of 5 curves around 32000 times and we can only access the smallest 10000 composites. I have factored 6000 composites though so many will have been replaced by larger/new numbers.

Why does the DB still list M43112609 as "Unknown"? That prime has been known since 2008. When is the primality proof going to be added? :huh:

[URL]http://www.factordb.com/index.php?id=1100000000215238611[/URL]

[QUOTE=Stargate38;307759]Why does the DB still list M43112609 as "Unknown"? That prime has been known since 2008. When is the primality proof going to be added? :huh:

[URL]http://www.factordb.com/index.php?id=1100000000215238611[/URL][/QUOTE]
The db doesn't handle Lucas-Lehmer results.

tweaked the param a bit
[code]
open(YAFU, 'yafu ecm('.$composite.',20) -B1ecm 100000 -threads 3|')
[/code]

13.5% factoring ;p

[QUOTE=henryzz;307703]Here is my modified script for ecm.[/QUOTE]Using the windows version. It runs the factoring fine, but never submits any result. I also don't get to see the % factored.