Howdy. Has anyone implemented SSOR (symmetric successive overrelaxation) or ILU (incomplete LU) or IC (incomple Choleski) factorizations for preconditioning the conjugate gradient method (or some other iterative method for solving large sparse linear systems)? I have found very little information on the implementation from the web :/ - Mikko Kauppila

Sadly, no.

But im wondering what you are working on .

To NewDeal:

I''m programming implicit integration (backward Euler) for solving the differential equations of a spring-mass system (cloth simulation). It involves a large sparse linear system which I''d want to solve as fast as possible. Elaborate preconditioners (SSOR and IC) would do this, but there is very little information on their implementation on the Internet.

- Mikko Kauppila