With julia, there is scarcely any difference between "built-in" and having a package---package code is not at a disadvantage compared to code that's in base (other than for loading time). Over time, it's expected more stuff will be moved from base into packages. So you shouldn't think of https://github.com/JuliaLang/IterativeSolvers.jl as being second-class simply because it's not in base.
--Tim On Sunday, June 21, 2015 05:19:27 AM [email protected] wrote: > Thank you Christoph. > > In the "Linear Algebra" part of the users manual I saw a number of built-in > matrix decomposition functions. In my understanding they are not that > different than iterative solvers, and I thought that if there is built-in > support for them, in the future there might be support for iterative > solvers and many other similar things too. > > The AMG solution that you mentioned. It may work fine, but it just does not > sound natural for a language that is designed for scientific computing. > > Cuneyt > > On Sunday, June 21, 2015 at 12:14:36 AM UTC+3, Christoph Ortner wrote: > > The tendency seems to be to provide packages (easy to install) instead of > > built-in. > > > > For iterative solvers look at IterativeSolvers.jl. For AMG, I have just > > started to use PyAMG (via PyCall) which was very easy to setup and seems > > to > > work fine. > > > > Christoph
