This was supposed to be a good news story, showing how a little algebra can reduce a 63 hour factorisation to only 18 hours. This was using the very latest...
Hi Bob, I had a similar problem myself a few days ago - I believe that in the last release (perhaps even earlier), I broke something in the square root code. I...
Hi Steve, The problem is probably not the 8 megs itself - rather, the memory allocations which came before that used up the available memory leaving not enough...
... Indeed - this is a known problem that I will try to correct in an upcoming release. If the linear polynomial is not monic, the score reported by polyselect...
[This didn't get posted the first two times I tried, so I'm trying again. I apologize for any duplications.] Hi Jason and other GGNFSers, I've been thinking...
Hi everybody! I was playing around with the XYYX SNFS thing and couldn't get it to work. I tried looking at n:=127^18+18^127. I don't know if it has been...
Hi all! I'm playing with the def-par file to see if I can tune the standard parameters, when I noticed a funny thing. I have just kicked off a GNFS c110, and I...
Hello all, I know alot of people have had this problem now, so let me explain the situation, as I understand it: The first time a new group member posts a...
Hello all, Just a follow-up to my earlier off-topic post. I just created another handful of moderators. Hopefully, I set all the flags right so that you can do...
Hi! I have seen the same things with several XYYX SNFS sets. I have just now at least 4 that do like that. First, I would recommend to switch C5/C0 and Y0/Y1,...
... Hi, and welcome! ... told it to? ... If the starting q-value for the lattice siever is less than alim, then the factLat script automatically adjusts alim...
Hello all, I believe I fixed the square root problem, though to be precise, I never actually found the problem itelsef (see the Changelog for more info). Also,...
... Hm... you know what, I was mistaken. I had thought the required form was 4*a^2 + b^2, but of course it is 4*a^4 + b^4, which is a bit harder to finagle....
... In early 2002 I used three main types of algebraic factorizations: 1) If x=ap, y=bp for some odd p then x^y+y^x is divisible by x^b+y^a 2) If x^y=a^2,...
... form. /bonks self. Sheesh, I'm really not keeping track of exponents too well these days. That should be: 27a^6 + b^6 = (3a^2 + b^2) (3a^2 - 3ab + b^2)...
... Hi Sam, Thanks for finding that little 4th degree polynomial! I have been recently completely convinced of how good 6th degree polynomials are for very...
... Hello k5, Aurifeuillian factors are everywhere! The xyyx number with x=129 and y=64 is indeed another convenient number to factorise using some simple...
Hi, ... Actually, this is not how it works. (Although i think it should be). Even if posts are approved the status of the user is still moderated. I've set the...
... <snip> And I don't anticipate any more ... sqrt ... anymore). ... Hi ! I can confirm this with the latest ggnfs I have done. Sqrt phase was failing with...
Hi (again)! Resuming a failing run, I have had a problem with the v0.72.12 factLat.pl script at the end of the sqrt phase : "can't call method bfloor" (I...
Thanks Chris. I should have realized that was the problem. I reduced maxRelsInFF to 26 and it is now chugging away at the matrix. It looks like it will take...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
