On Sun, 22 Oct 2006 19:23:00 -0500, Bill Hart <[EMAIL PROTECTED]> wrote: >> The only thing I'm aware of is the bit-operations I mentioned in an >> earlier email about his SSMul function. > > The thing that really sux is that earlier version of MAGMA computes the > whole product in less time than NTL computes just the pointwise > multiplications, and that is just the multiplications too, not > including the non-predictable branches that follow. > > It simply isn't possible to get down to the times of that earlier > version of MAGMA with this algorithm. It's FFT to be sure, but must be > some kind of FFT over the reals or something.
I wonder why they switched from that faster algorithm to a slower one? Maybe it produced incorrect results (possibly due to rounding errors)? William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
