I'm doing some factoring stuff at my college and I have a small cluster available. I've pieced together some information online, but if anyone can offer some...
2400
Jeff Gilchrist
jeff.gilchrist
May 3, 2009 10:54 am
... If the numbers are 120 digits they can be processed fairly fast. For GNFS numbers it would take about 4 days on 1 processor of a Pentium4 3 GHz machine....
2401
torbjornalm
May 6, 2009 10:00 pm
I have come across a new problem using factMsieve. I am running a gnfs factoring with a c122.In the matrix phase (nc2), the number of columns dropped from...
2402
gchil0
May 7, 2009 3:34 am
What error does msieve give when it terminates? If you can't see it on the screen, it should be in the log file....
2403
Torbjörn Alm
torbjornalm
May 7, 2009 8:04 am
Here is the final death of th factoring: Wed May 06 23:48:47 2009 Wed May 06 23:48:47 2009 Msieve v. 1.40 Wed May 06 23:48:47 2009 random seeds: 14fbdb70...
2404
Jason Papadopoulos
jasonp@...
May 7, 2009 11:28 am
... The other times I have seen this happen, it was because there were just enough relations to create a matrix but not enough to solve that matrix. Sieving a...
2405
Jeff Gilchrist
jeff.gilchrist
May 7, 2009 12:42 pm
... As Jason said you might need some more relations. What happens from factMsieve's point of view after msieve fails, does it go on to try to sieve some more...
2406
flimped
May 12, 2009 7:48 am
... Jeff- I've been using your guide as a starting point and it has been a great help. I noticed that factMsieve.pl accepts some arguments called client_id and...
2407
Luigi Morelli
et_cygni_61
May 12, 2009 7:48 am
... We at ClusterArch are working on a 16 multicore workstations small cluster, based on a rewritten Beowulf kernel and MPI. Are there open source factoring or...
2408
torbjornalm
May 12, 2009 7:58 am
It seems that the problems in the root finding code does not only hit SNFS-parameters but also gnfs parameters, generated with ggnfs. So far I hae been hit 5...
2409
jonathankk1
May 14, 2009 5:02 pm
Hi, I'm trying to set up GGNFS to factor some composite numbers with 120+ digits, but I'm having some trouble getting it working across my cluster of...
2410
torbjornalm
May 14, 2009 8:00 pm
A snfs attempt on a c164 bombed out once msieve was called. I had set minrels to 1000000. Once the number of relations reached this limit and Msieve was called...
2411
Jeff Gilchrist
jeff.gilchrist
May 14, 2009 8:13 pm
... Are you actually using msieve 1.38? That might be your problem. You should get the latest 1.41 and use that instead from: ...
2412
Torbjörn Alm
torbjornalm
May 15, 2009 7:02 am
Hi! I tried to use Msieve v 1.41. It unfortunately is hit by an other bug. It bombs out after print out : commensing number field siev(162-digit input). It is...
2413
chris2be8
Jun 6, 2009 2:09 pm
Having managed to get ggnfs compiled I compared it's speed with NFSX (ubasic/assembler NFS implementation written by Yuji Kida). Test case 10^151+1 with small...
2414
chris2be8
Jun 6, 2009 3:11 pm
Hello, I've done some tests with various ggnfs configurations. All on a Pentium 4 3.06 GHz stepping 9 with hyperthreading. The poly file was: n:...
2415
chris2be8
Jun 19, 2009 7:05 pm
Hello, I've written a patch for factMsieve.pl to automatically calculate skew if it isn't provided in the .poly file for a SNFS job. It should also work for...
2416
Serge Batalov
batalovs
Jun 20, 2009 8:09 pm
... Ok, makes sense. Except that it doesn't have to flipped. It is always (|c0|/|cn|)**(1/n) (i.e. can be less than 1). I'll post it. Thanks. --Serge...
2417
chris2be8
Jun 21, 2009 3:44 pm
... ggnfs-doc.pdf page 17 says "It should be at least 1". So I assumed I should invert the result if less than 1. At least that's was the "Draft of February 8,...
2418
Jason Papadopoulos
jasonp@...
Jun 21, 2009 10:07 pm
... The skew being larger than 1 is a good rule of thumb, and is especially useful becase the rest of the code assumes relations will have an 'a' coordinate...
2419
chris2be8
Jul 3, 2009 4:01 pm
Hello, When factoring something like a^3k-1 I first algebraically reduce it to (a^2k+a^k+1)(a^k-1). To crack the first part I need to convert it into a degree...
2420
Serge Batalov
batalovs
Jul 7, 2009 3:53 pm
... Crossover from degree 4 to 6 (when 5 is not possible) is about difficulty 145-150 digits. Note: One sieving side, algebraic or rational, will be much...
2421
chris2be8
Jul 30, 2009 3:10 pm
I've noticed that the linecount subroutine opens FILE but never closes it. Is this likely to cause problems such as a memory leak? Is there any reason not to...
2422
chris2be8
Aug 3, 2009 4:57 pm
Hello, I was reviewing how long various runs took and noticed that neither the log nor the summary say which lattice siever was selected by factMsieve.pl....
2423
billrea65
Aug 4, 2009 7:26 am
Hello, If someone is going to tweak factMsieve.pl another enhancement would be to automatically select gnfs-lasieve4I15e or gnfs-lasieve4I16e when necessary as...
2424
chris2be8
Aug 13, 2009 4:23 pm
... Replying to my own post I added close(FILE); and ran a test run which worked OK. While looking at it I noticed that linecount reads through the whole of...
2425
David Willmore
davidwillmore
Aug 13, 2009 5:09 pm
... Every little bit helps. I can't speak for anyone but myself, but I thank you for your work on this. Cheers, David...
2426
lars_dausch
Aug 16, 2009 8:11 pm
Hi, I compiled the siever with Visual Studio in the 64 Bit version and get the following output: 537183 215 8124 85497248861947451 179 6800 1471808426074512745...
2427
Jeff Gilchrist
jeff.gilchrist
Aug 17, 2009 12:57 am
... The seiver runs fine, I have not been able to find where this output happens so it is a mystery to me as well. Jeff....
2428
lars_dausch
Aug 17, 2009 4:58 pm
... I know that it works but as a programmer who has to write massive multi threaded applications it allways makes me nervous when there is an application that...
