There are several ways to do this. I do it by creating a poly file - "example.poly" that has lines for n, Y0, Y1, c0, c1, ... cn, skew, and type. The script will look for the poly file and use it.

ok thank u for ur answer but i already tried this and get the error msg "Error: evaluated polynomial value polyval is not a multiple of n!"

You'll get that error if your polynomial cannot correspond to the number you are factoring. Are you sure you didn't make a typo creating it?

(We prefer english to txt spk btw)

i dont see any typo. i factored rsa100 and there is a rsa100.poly file in the folder. i think this is the created file by msieve:


n: 1522605027922533360535618378132637429718068114961380688657908494580122963258952897654000350692006139
Y0: -2063478208689716601070096
Y1: 30967945534613
c0: 2575995887443956202408325707
c1: 19135305567240159659098
c2: -13968790193861201
c3: -1179863448
c4: 84
skew: 6422631.29
type: gnfs


Then i wanted to make some tests with other polynomials and changed for example c4 from 84 to 86 and then i get the error message

If you change one of the coefficients you have to change all the others to compensate for it; the whole polynomial is very specific to the number you want factored.

If you want lots of polynomials that will work, you can run msieve on the same input and look for 'msieve.dat.p' which will have many polynomials that are slightly worse but at least will work.

u are right. it was very stupid to think this will work ^^ of course we need for example irreducible polynomials...ok so will try to create test polynomials. thank you very much for your help

Lattice siever parameters.
 

Hello,

Does anyone have a complete list of parameters gnfs-lasieve4I1?e takes? I'm looking for one to say what to call the *.afb.0 files (an option to say don't save them to disk would be useful when sieving below the AFB limit).

I've read the INSTALL.and.USE file in the src/lasieve4 directory, but that's not complete.

A full list of files read and written and what should be in each would also be very helpful.

Chris K

hello guys, i wanted to test a polynomial with ggnfs. so i created a polynomial with the base-m-method for a 82 digits number. ggnfs ran for some hours (i know that its a "bad" polynomial but i wanted to test it) and then i got an error message. anyone knows why it didnt work? i paste from the command window:

-> ________________________________________________________________
-> | Running factmsieve.py, a Python driver for MSIEVE with GGNFS |
-> | sieving support. It is Copyright, 2010, Brian Gladman and is |
-> | a conversion of factmsieve.pl that is Copyright, 2004, Chris |
-> | Monico. Version 0.76 (Python 2.6 or later) 10th Nov 2010. |
-> |______________________________________________________________|
-> This is client 1 of 1
-> Running on 4 Cores with 1 hyper-thread per Core
-> Working with NAME = testzahl1a
-> Selected default factorization parameters for 80 digit level.
-> Selected lattice siever: gnfs-lasieve4I11e
-> No parameter change detected, resuming...
-> Running setup ...
-> Estimated minimum relations needed: 1e+06
-> resuming a block for q from 16780000 to 16790000
-> Running lattice siever ...
-> entering sieving loop
-> making sieve job for q = 16780000 in 16780000 .. 16782500 as file testzahl1a.
job.T0
-> making sieve job for q = 16782500 in 16782500 .. 16785000 as file testzahl1a.
job.T1
-> making sieve job for q = 16785000 in 16785000 .. 16787500 as file testzahl1a.
job.T2
-> making sieve job for q = 16787500 in 16787500 .. 16790000 as file testzahl1a.
job.T3
-> testzahl1a : Severe error!
Traceback (most recent call last):
File "D:\incoming\bachelorarbeit\ggnfs1\factmsieve.py", line 2066, in <module>

run_siever(client_id, num_clients, SV_THREADS, fact_p, lats_p)
File "D:\incoming\bachelorarbeit\ggnfs1\factmsieve.py", line 1683, in run_siev
er
.format(fact_p['q0'], 2 ** lats_p['lpba']))
ValueError: Unknown format code 's' for object of type 'int'

D:\incoming\Bachelorarbeit\ggnfs1\testzahl1a>

[QUOTE=chris2be8;270322] Does anyone have a complete list of parameters gnfs-lasieve4I1?e takes? I'm looking for one to say what to call the *.afb.0 files (an option to say don't save them to disk would be useful when sieving below the AFB limit).
[/QUOTE]

I've found the code in gnfs-lasieve4e.c that parses the parameters, which tells me what parms it expects. When I've had time to experiment I'll post a list.

Chris K

Hello,

Here's my best guess at gnfs-lasieve4I1*e's parameters. Most of it comes from reading the source and the INSTALL.and.USE file in src/lasieve4.
[code]
-F force AFB to be calculated
-J n bits for 2nd dimension of sieving region
-L prog exit if prog gives a non-zero rc
-M n first_mpqs_side ?
-N 1234 write last special q processed to foo.bar.last_spq1234
-P n first_psp_side ?
-R resume sieving
-S skew of polynomial
-a sieve on algebraic side
-b foo call AFB file foo.afb.[01]
-c n number of special q to sieve
-f n first special q to sieve
-i n selecting side ?
-k keep factor base in foo.afb.[01]
-n 1234 catch signals and write last special q processed to foo.bar.last_spq1234
-o output file name
-q n 0=sieve on algebraic side, 1=sieve on rational side. Not accepted by all sievers.
-r sieve on rational side
-t n selecting side ?
-s short output. Only some versions.
-v verbose output
-z Compress output
[/code] I'm not sure what the ones with a ? do. I don't know C so I can't follow the code far enough.

-q is only supported by the 64 bit Linux siever.
-s is not supported by the 64 bit Linux siever.

Chris K

Trouble with 64-bit compilation in Linux
 

I've run into some difficulty in setting up a "new-to-me" dual core 64-bit machine. I have almost everything running, but for a small problem with ggnfs. I have the sourceforge svn revision 414 set of files. I get the following error:
[code]
gcc: error: asm/liblasieveI15.a: No such file or directory
make[3]: *** [../../bin/gnfs-lasieve4I15e] Error 1
rm lasieve-prepn.o prho.o asm/liblasieve.a asm/liblasieveI13.a gnfs-lasieve4eI14.o real-poly-aux.o recurrence6.o mpqs.o primgen32.o asm/liblasieveI14.a redu2.o gnfs-lasieve4eI12.o input-poly.o gnfs-lasieve4eI13.o gnfs-lasieve4eI15.o asm/liblasieveI12.a
[/code]The processor is:
[code]
AMD Athlon(tm) 64 X2 Dual Core Processor 6000+
[/code]so, I used the x86_64 target.

The bin directory only shows gnfs-lasieve4I[B]12[/B]e, gnfs-lasieve4I[B]13[/B]e and gnfs-lasieve4I[B]14[/B]e. Unfortunately, Aliqueit recently called for gnfs-lasieve4I[B]11[/B]e.

I have also looked at frmky's set of 64-bit files, but they don't include gnfs-lasieve4I11e.

I see a Makefile in the lasieve4 directory that shows:
[code]
all : $(BINDIR)/gnfs-lasieve4I12e $(BINDIR)/gnfs-lasieve4I13e \
$(BINDIR)/gnfs-lasieve4I14e $(BINDIR)/gnfs-lasieve4I15e \
$(BINDIR)/gnfs-lasieve4I16e $(BINDIR)/gnfs-lasieve4I11e
[/code]Should all six of these be created?

For the moment, I am working through this by raising the gnfs_cutoff in aliqueit.ini, but I'd like to find a proper solution, if one exists.

Thanks for any comment...