> Why can't we do first first the factorization up to n-2 bits (1/4) of the
> trial factoring time, then start the P-1 factoring up to 1/3 of the B1
> value, after this, we can complete the trial factoring process and at the
> end we complete the P-1 (using the save file od intermediate file). (the
> parameters can be optimized)
This sounds fairly sensible. However, this entails splitting the factoring part into fairly small sub-assignments, which may cause unneccessary complications with the server. Also, trial factoring and P-1 are quite different from the point of view of system requirements - trial factoring uses very little memory (in practice it runs almost entirely in the L1 cache) whereas P-1 is actually more of a memory hog than LL testing. So I suspect we want some bias towards early trial factoring rather than P-1.
> until now >210 factors are found for 10megadiginumbers and more than 280
> exponents were factored up to 68 bits. Some (about 7) 67 digit factors
> were found but none with 68 bits.
This is likely to be a statistical anomoly. A sample size of 7 is a bit small to condemn the data as biased.
> A lot of factors of exponents between 10000 and 1000000 were found using
> the new P-1 method. Is there a database which contains which exponent were
> tested using which B1 and maybe a database od the save files?
Yes - I think we need this database - with or without savefiles, it's a waste of effort to inadvertently duplicate work done before. Since P-1 is deterministic (like trial factoring, but unlike Pollard's rho or ECM) you should get the same result every if you use the same limits on the same exponent.
If anyone has any data to contribute, I'd be willing to assemble & publish the database. I also have adequate storage space on my anon ftp server for save files.
Regards Brian Beesley _________________________________________________________________ Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
