[ODE] More speed???

[Aras Pranckevicius]

>     The "system" is LCP A*x = b+w with certain conditions. Solve such
>     system require many math skills (which I don't have :( ), you can
>     read the famous Barraff's paper about it.

Ah, right... I got confused by those extra conditions :)

>         2) Change it-> use iterative methods : theoritically faster
>         but may break ODE stability. (But actually, I prefer this way)

If that's plain Ax=b, some sort of Conjugate Gradient might actually be
better. For CG, A must be symmetric and positive definite (I don't know what
that means :)), though. If A is also sparse, that can be exploited
efficiently.

Some nice stuff for all kinds of these things is in Iterative Template
Library - http://www.osl.iu.edu/research/itl/ and Matrix Template Library -
http://www.osl.iu.edu/research/mtl/



Aras Pranckevicius aka NeARAZ
http://www.gim.ktu.lt/nesnausk/nearaz/