20070306, 19:08  #12  
∂^{2}ω=0
Sep 2002
República de California
2^{2}×5×11×53 Posts 
Quote:
Maybe just tweak the code to check inputs below a certain size (100010000 digts sounds like roughly the right range) via prp test and issue a "composite" warning if the input is found to be so via prp. 

20070310, 08:40  #13  
Jul 2003
So Cal
891_{16} Posts 
Quote:
Greg 

20070413, 23:26  #14 
(loop (#_fork))
Feb 2006
Cambridge, England
14451_{8} Posts 
Different wall
Can you see the problem with this code:
Code:
PREFIX=parsed rm $PFX* for i in *.rels; do /home/fivemack/maths/ggnfs0.77.1/ggnfs/bin/procrels fb foo.fb prel $PREFIX newrels $i; done Fortunately, text forms of relations are fairly easy to see in a disc image Code:
for i in `seq 0 218`; do dd if=/dev/sdb2 of=block$i bs=1G skip=$i; strings n 40 block$i  grep '^[09,]*:[09afAF:,]*$') > rels$i; done Code:
[a1^5 a1^4b1 a1^3b1^2 a1^2b1^3 a1b1^4 b1^5] [c5] = [+/ algnorm1] [a2^5 a2^4b2 a2^3b2^2 a2^2b2^3 a2b2^4 b2^5] [c4] [+/ algnorm2] [a3^5 a3^4b3 a3^3b3^2 a3^2b3^3 a3b3^4 b3^5] [c3] [+/ algnorm3] ... Once you know Y1 and Y0, a little Python script can go through the rather large (for there were relations from many factorisations on the disc) set of strings of the right form, and check that the factorisation of the algebraic side is right; then you get a couple of gigabytes of text to feed to procrels, and hopefully happiness ensues. Disc space is cheap and human folly common, there's no need ever to have 'rm' in a relationprocessing script Last fiddled with by fivemack on 20070413 at 23:31 
Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
WallSunSun primes  Gandolf  Math  49  20170413 20:00 
If you could meet any Mersenne prime discoverer...  ixfd64  Lounge  24  20120820 21:00 
The Hello  I am  Nice to meet you thread....  Prime Monster  Lounge  23  20120211 11:08 
Meet the English?  wblipp  Lounge  27  20100918 13:03 
The Ladder Against The Wall  Numbers  Puzzles  27  20050702 10:19 