20120320, 15:22  #1 
"Daniel Jackson"
May 2011
14285714285714285714
2^{2}·151 Posts 
Anyone have an ETA for M1061?
I want to know when they will be done with M1061. It's bothering me. I thought it would only take until 3/16 to factor (taking into account the huge effort of NFS@Home).

20120320, 15:37  #2 
(loop (#_fork))
Feb 2006
Cambridge, England
6322_{10} Posts 
Restrain your curiosity. You have asked this question several times and it has not been answered. Accept that it probably won't be answered except in the form of a product of two or three integers equal to 2^10611, sometime in the fullness of time.

20120320, 15:37  #3 
Tribal Bullet
Oct 2004
3×1,163 Posts 
(You don't need to ask twice)
Greg's time is extremely limited right now, and M1061 is proving to be slightly too large to handle with the 16e sieving tools. The sieving long ago reached the point where most new relations are duplicates of existing relations, so that he'll have no choice but to start the postprocessing with less oversieving than everyone is comfortable with for jobs of this size. I'll let Greg provide more details beyond that. 
20120320, 15:43  #4  
Nov 2003
2^{6}·113 Posts 
Quote:
What about trying specialq that reside inside the factor base? 

20120320, 18:24  #5 
Jul 2003
So Cal
3773_{8} Posts 
2,1061 is actually on hold for a short while. Sieving on both sides of the congruence to the maximum supported by lasieve4 yielded just under the required number of relations. It was frustratingly close with around 595 million unique relations, and I suspect another 30 million or so will do. lasieve5 will allow me to extend the specialq range and grab those relations, but I simply haven't had time to modify the source to accept GGNFSstyle input and output. Next week is spring break, so I should have time then. So the answer remains soon...
Edit: And yes, special q inside the FB were used. Last fiddled with by frmky on 20120320 at 18:25 
20120320, 23:32  #6 
Just call me Henry
"David"
Sep 2007
Cambridge (GMT/BST)
2^{2}·1,433 Posts 

20120330, 09:59  #7 
Basketry That Evening!
"Bunslow the Bold"
Jun 2011
40<A<43 89<O<88
3×2,399 Posts 
Trying to add to my understanding...
1) Sieving 2) lin alg (finding the nullspace of a hugeass matrix) 3) some sort of sqrt What happens if you try to move past the sieving with too few relations? 
20120330, 10:10  #8 
Bamboozled!
"πΊππ·π·π"
May 2003
Down not across
2^{2}·3^{2}·281 Posts 

20120330, 10:15  #9 
Basketry That Evening!
"Bunslow the Bold"
Jun 2011
40<A<43 89<O<88
1C1D_{16} Posts 
... And what's F_2? :P

20120330, 11:58  #10 
Bamboozled!
"πΊππ·π·π"
May 2003
Down not across
2^{2}×3^{2}×281 Posts 

20120330, 12:27  #11 
Tribal Bullet
Oct 2004
6641_{8} Posts 
You should visit the wikipedia page on the quadratic sieve; it should add a fair amount to your understanding.

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Predict the number of digits from within the factor for M1061  Raman  Cunningham Tables  12  20130617 21:21 
M1061 factored!!!  lycorn  NFS@Home  28  20120830 04:40 
M1061  t60  Andi47  Factoring  122  20111125 09:18 
P1 on M1061 and HP49.99  ATH  Factoring  21  20091013 13:16 
M1061...  Xyzzy  Factoring  261  20071102 05:11 