20220809, 17:40  #100  
Jun 2012
2^{3}·3·151 Posts 
Quote:
Code:
worktodo.ini You can then generate a GNFS polynomial using the np flag, often in conjunction with other flags (like g 0) for a more efficient search. But the area of poly search is too complex for a single post. Let’s say you generate a polynomial, or build one for SNFS using some of the classic methods or Yafu. Msieve places this polynomial in a file called Code:
msieve.fb Now just invoke msieve with the command Code:
./msieve v nc target_density=120 t 8 This will eventually spit out all factors of your number. If msieve stops, you can restart it with the ncr flag. nc1 starts filtering phase nc2 starts linear algebra phase nc3 starts square root phase So if LA fails, you can restart it with the ncr and nc3 flags, thus telling msieve to finish LA and continue through square root and grind out the factors. Last fiddled with by swellman on 20220809 at 17:41 

20220809, 17:49  #101 
"Ed Hall"
Dec 2009
Adirondack Mtns
13·19^{2} Posts 
There are several steps, all with their own files that need to be done in order to make it through. One of the first steps is to find a "good" polynomial (technically, a pair). This polynomial is necessary for the rest of the steps and is located in a file commonly named msieve.fb. Although Msieve can do sieving, for anything larger than a few digits, an external program is used. This program needs the polynomial and creates a file of relations, commonly named msieve.dat. These relations are then filtered (nc1) and more files are created. Then the stage where the GPU is used can be run (nc2) and a checkpoint file is created, which can be used for a restart (ncr). Depending what files you already have, you can choose where to start/continue.
All of this can be found in the Readme.nfs documentation supplied in the same folder as msieve. Note: I see swellman has posted as well, but I'll let this one stand as is. It shouldn't conflict. 
20220809, 18:56  #102  
Random Account
Aug 2009
Not U. + S.A.
2×1,123 Posts 
Some were interested in my number. 128 digits, a.k.a. C128.
Code:
99183446604001940389822905615521048138949828629950976062436205758607664727160159612467952971325033872947039325281585706220344353 Code:
./msieve v g 0 np Quote:
I paragraphed your text above to make it easier to follow. BTW, I see mentions of "LA." Beyond the obvious, what is this? 

20220809, 19:17  #103 
Jun 2012
3624_{10} Posts 
LA = linear algebra
TD = target_density 
Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
PFGW benchmarking  carpetpool  Hardware  4  20190930 20:06 
Looking for benchmarking help with a Phenom or PhenomII X6  mrolle  Software  25  20120314 14:15 
GMP 5.0.1 vs GMP 4.1.4 benchmarking  unconnected  GMPECM  5  20110403 16:16 
Benchmarking dualCPU machines  garo  Software  2  20100927 20:33 
Benchmarking challenge!  Xyzzy  Software  17  20030826 15:43 