Dear Alberto, thank you for the information about your work. This looks very interesting.
In particular, the way you break down into "Decoupling", "remove left entries", "remove right entries", "Left triangulation", etc, is very interesting. I look forward to seeing the paper when it appears, and Marco's code. Of course we currently already use your toom 7 sequences in MPIR. I suspect there is no way for us to tell how good the new toom 8 will be except wait for the mpn level implementation or maybe have a try ourselves. Once again, thank you for the information. Bill. 2009/10/7 Alberto <alberto.zan...@gmail.com>: > > Dear developers, > I implemented a GMP-mpz version of Toom-8 > algorithm > for multiplication. On my architecture I obtained someway interesting > results with comparison to GMP 4.3.0. You can find two graphics, > for multiplication and squaring, at > > http://bodrato.it/papers/zanoni.html#SYNASC2009 > > (under "Toom-8 way for long integers multiplication" - thanks Marco !) > > As I'm waiting for the publication of the proceedings of the SYNASC > 2009 > conference, where the related paper was presented, currently I cannot > put the > paper itself, with all the details, on the web, but actually Marco is > working > quite ahead of it, and will surely produce something better at mpn > level, > with many new ideas, so I think that the paper is actually already > obsolete. > > Just to point out that Toom-Cook-k methods with k as high as 8 (may be > even > more ?) can be effective and worth implementing. > > Alberto > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-devel@googlegroups.com To unsubscribe from this group, send email to mpir-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en -~----------~----~----~----~------~----~------~--~---
