yade-dev team mailing list archive

[Václav Šmilauer <eudoxos@xxxxxxxx>]

PS I think the way both Dem3DofGeom and ScGeom currently work is not
prospective. ScGeom is dumb and only stores a few numbers inside, making
Law doing all the computation; Dem3DofGeom is better (right? ;-) ) but
uses many virtual functions. In addition, both classes have different
interfaces.

As we already discussed, it would be much more efficient to precompute
normal and shear displacements in the Ig2 functor and store it in the
*Geom class. Only writing back to *Geom from Law2 (such as upon a
plastic slip) would need a few virtual methods, which could be
interface-wise identical for incremental and total formulations BTW.

If properly designed, it could be easily extended to 6-dof contacts as
well, which we currently are not able to handle (I mean, gracefully).

I don't see anyone doing this ATM (not me anymore, definitely), though.

Cheers, v.