ffc team mailing list archive

[Kristian Oelgaard <k.b.oelgaard@xxxxxxxxxx>]

Quoting Marie Rognes <meg@xxxxxxxxx>:

> Kristian Oelgaard wrote:
> > Quoting Marie Rognes <meg@xxxxxxxxx>:
> >
> >
> >> Did we agree on upping the order of the RT/Nedelec 1st kind elements? So
> >> that they start at 1?
> >>
> >
> > I think so, then UFL can determine the correct order of the polynomials
> such
> > that FFC will use the correct number of quadrature points for exact
> > integration.
>
> Ok!
>
> > It shouldn't change much in FFC right?
> >
>
> Nope, I'll just call FIAT with "degree - 1" in the initialization. I'll
> push the change.

That's what I thought.

> I was a bit surprised that the generated code did not change except for
> the signature.
> Is this expected?

I think I noticed this before, it's probably because the polynomial order of the
demo can still be integrated by the same number of points. You can check this in
compiler.py where the order is determined.

Kristian

>
> --
> Marie
> _______________________________________________
> FFC-dev mailing list
> FFC-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/ffc-dev
>