From: Nicolas C. <war...@fr...> - 2005-04-25 09:57:46
|
> > If there's any concern about the speed of 'big_int_of_int', perhaps it > > could be optimized by having big_int representations of range > > -128..128 stored in an array and looking these up in case the input > > parameter falls into this range. If the input parameter is out of > > this range, then the generic 'big_int_of_int' would be used. > > Yes, you're right. To be honest I was getting very frustrated trying > to write a Miller-Rabin implementation for ExtBig_int, and was > thinking about moving to Numerix, which is probably what I should have > used in the first place. Only problem is that the GODI distribution > on my laptop is having upgrade problems ... > > I'll release what I have at some point and we can discuss whether it's > worth adding to extlib or not. > > Rich. I personaly never had the use for Bigints, so I might not the best to decide whatever extensions are needed or not. Does some other people here already felt frustrated with current Bigint API ? Nicolas |