Hello, ... Thank you. I'll check it up. -- With best regards, Anton Korobeynikov. Faculty of Mathematics & Mechanics, Saint Petersburg State University....
... Hi Sander, I once did a c129 just using the values for c125 from the tables (not knowing any better), but it was a very long slow job (30 days). Instead of...
I'm trying ggnfs for the first time on 10^147+19, so the polynomial x^5+19000 is ok right? Anyone with experience have suggestions for the best parameters? ...
I found a post about another snfs number of similar size and used those parameters: skew: 2, rlim: 5000000, alim: 5000000, lpbr: 27, lpba: 27, mfbr: 48, mfba:...
Hello, You can find many similar (not same) factorizations at: http://homepage2.nifty.com/m_kamada/math/factorizations.htm The SUBMIT/RESERVE pages of each...
Thanks for your answer. Is the monic polynomial x^5+19000 with m=10^30 better than 100x^5+19 with m=10^29 ? In the example you linked they used "Max relations...
Hi, I've spent some time trying to find clues to the cause of the sqrt bug, and I've found 2 things that might be helpful to someone more familiar with the...
Hello Greg, ... Is there any way to reproduce that bugs? -- With best regards, Anton Korobeynikov. Faculty of Mathematics & Mechanics, Saint Petersburg State...
Yes, it happens on most dependencies on nearly essentially every number I factor, so creating a test case should be very easy. I tried the small tests...
Hello, This email message is a notification to let you know that a file has been uploaded to the Files area of the ggnfs group. File : /source...
ggnfs@yahoogroups.com
Jul 28, 2006 9:59 am
1909
Hello everyone, Ivan Petrov recently had some ideas of speeding up current block Lanczos implementation. It's extremely important especially for 64-bit ...
mult64x64_relf2() seems to be absolute champion in all cases on x86 platform. Compiler GCC 3.4.4 (mingw special) 1. Machine-1. Pentium4, 2612 Mhz, Northwood,...
Hi, that was done on dual core opteron 280, with two ecm processes running with nice=19, and normal server processes (X, postfix, squid, apache, etc). I...
Here are results for the Microsoft and Intel 32-bit compilers for a 1.8GHz Pentium 4 and a AMD64 4800 X2 running in 32-bit mode 1.8GHz Pentium 4, Microsoft...
Hi! I see you test results, very interest result for opteron. Function RELF1 - It oldes implementation of RELF2, but it use 64 bit variables instead of MMX...
Hi, ... Opteron 144, 1.8 GHz, 1024KB cache running Linux in 64-bit mode ... gcc 3.4.6 ... I had to change the define of s32 from long to int. On this...
Hi! Thx for patch. I think for clean "C" need make inline function for GCC like instrict "_BitScanForward" M$ Compiller. I now make new implementation, and add...
Hi, Here's one more on the same computer but using a 32-bit binary rather than a 64-bit binary. ... Opteron 144, 1.8 GHz, 1024KB cache running Linux with a...
Hellow! I think now we have candidate implementation, but I not know what do with Opteron CPU, on thish architecture code work slow that *_mmx. P4 Centrino...
Hello, This email message is a notification to let you know that a file has been uploaded to the Files area of the ggnfs group. File : /test1.c Uploaded...
ggnfs@yahoogroups.com
Jul 29, 2006 10:03 am
1921
Hello, ... Opteron 144, 1.8 GHz, 1024KB cache running Linux ... gcc 3.4.6 ... 64-bit binaries: I had to disable mult64x64_new1 because of the error ...
Hello, Here are the speeds on my AMD64 ... Athlon 64 3200+, 2 GHz, 1024KB cache ... gcc 3.4.4 ... These were Cygwin binaries running in Windows XP, so they are...
Hello, One more...here are the speeds on my Athlon ... Athlon XP 2700+, 2.167 GHz, 256KB cache ... gcc 3.4.4 ... This is a Cygwin binary running in Windows XP....
Hellow all! test1.c updated. '*_new2' function now work with equal (old MMX implementation) speed on Opteron CPU, and +~10% effective than '*_new' on P4. ...
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.
