dolfin team mailing list archive

[Anders Logg <logg@xxxxxxxxx>]

On Fri, Dec 12, 2008 at 08:55:15AM +0000, Garth N. Wells wrote:
> 
> 
> Anders Logg wrote:
> > I thought there was a bug in the Stokes Python demo (Taylor-Hood) but
> > it gives exactly the same results as the C++ demo (with heavy
> > oscillations in the pressure).
> > 
> > I think I remember that we've had this problem before and fixed it,
> > but don't remember why. Does anyone remember?
> >
> 
> The last time we had this problem it was due to the MeshFunction that 
> determines the subdomains on which the boundary conditions are applied. 
> It got messed up with the mesh re-ordering business. The result was 
> sensitive to the order in which the MeshFunction was read in and the 
> mesh was re-ordered.
> 
> Garth

ok, I'll take a look.

-- 
Anders

Attachment: signature.asc
Description: Digital signature