If i take for example a poly from Kamada`s page per copy & paste i got this error message.
[URL]http://hpcgi2.nifty.com/m_kamada/f/c.cgi?q=10009_207[/URL]
[URL="http://hpcgi2.nifty.com/m_kamada/f/c.cgi?q=10009_207"][/URL]

When I run SNFS using factmsieve.py, I create a poly file with lines for only

n:
Y0:
Y1:
c0:
c....
skew:
type:

The poly file you linked to at kamada is missing the Y's and has many other lines.

Yes - i thought i can take the complete poly from Kamada`s site.
Is it necessary to delete the other lines?

Hmm but with other snfs polys from Kamada`s site it works correct.
For example this poly its working [URL]http://hpcgi2.nifty.com/m_kamada/f/c.cgi?q=52221_175[/URL]
[QUOTE]n: 278586026388018839424488101621126621862807120406538167286450533736771898645273264778821489200475270906208438824116075815743606167406550256787201567684498069231429
m: 100000000000000000000000000000000000
deg: 5
c5: 47
c0: -11
skew: 0.75
type: snfs
lss: 1
rlim: 6200000
alim: 6200000
lpbr: 28
lpba: 28
mfbr: 53
mfba: 53
rlambda: 2.5
alambda: 2.5[/QUOTE]PS: I`m using WinVista 32bit & Python 3.2 with factmsieve.py 0.84

Sadly none of these poly files fail during input for me with either 0.84 or 0.85. I do, of course, ensure that they are in native format (Windows for me).

When you use cuda for the poly select, in version 0.83-0.84 you limit the poly research field to 0-4000. When i use 0.76, the 'field' is much larger, up to 250k (for the C128). Maybe it has an influence?

In GGNFS format, if a poly file has an 'm' line, then Y0 is -m and Y1 is 1.

Is there anywhere I can get Linux 64-bit versions for CentOS? Or is it necessary to build from sourceforge source? I'm working through another person that has direct access to the target system, and would prefer to avoid the build step and any potential cycles within that if possible.

William

Actually, I wouldn't be opposed to building, as long as the steps were laid out in excruciating detail, and were correct. I don't want to have to spend the next few days trying to make it work.

[QUOTE=wblipp;267444]Is there anywhere I can get Linux 64-bit versions for CentOS?[/QUOTE]

Jeff's [URL="http://gilchrist.ca/jeff/factoring/nfs_beginners_guide.html"]Beginner's Guide[/URL] says Greg has [URL="http://www.mersenneforum.org/showpost.php?p=126546&postcount=23"]64 bit Linux builds[/URL], but that has only the sievers. Jeff's guide says I also need executables for eleven other programs (from makefb to sqrt in the list). Are these available someplace, or must they be built from sourceforge?

[QUOTE=wblipp;267987]Jeff's [URL="http://gilchrist.ca/jeff/factoring/nfs_beginners_guide.html"]Beginner's Guide[/URL] says Greg has [URL="http://www.mersenneforum.org/showpost.php?p=126546&postcount=23"]64 bit Linux builds[/URL], but that has only the sievers. Jeff's guide says I also need executables for eleven other programs (from makefb to sqrt in the list). Are these available someplace, or must they be built from sourceforge?[/QUOTE]Actually, if you use the GGNFS/msieve hybrid scripts, you do not need these programs (they are the post-processing tools for the GGNFS suite):
makefb.exe
matbuild.exe
matprune.exe
matsolve.exe
procrels.exe
sqrt.exe

Others that may not be needed: sieve.exe is the line-siever for GGNFS.

And these are the GGNFS poly-select tools for numbers over 99 digits:
pol51m0b.exe
pol51m0n.exe
pol51opt.exe

This one is the poly-select for <99 digits:
polyselect.exe

If you use msieve's poly-select, you don't need the above programs.

One caveat: if you don't include these in the proper place the work scripts will complain. Just comment out the checks for them or set this: "$CHECK_BINARIES=1;" to 0. (They may actually bail out improperly if the other stuff is not included, so it might be better to make sure you comment out the lines where they're invoked....)

hello guys, im a noob with ggnfs/msieve and used jeffs guide for the beginning ([url]http://gilchrist.ca/jeff/factoring/nfs_beginners_guide.html[/url]). now i have a simple question: how can i test own polynomials? the script always uses polynomials from msieve