syfi team mailing list archive

[Kent Andre <kent-and@xxxxxxxxx>]

As you refine the mesh the solution goes to the true solution. 
The condition number of the matrix seem to be correct. 
Furthermore, if f is interpolated into the same space as the
solution you get an exact match. 

-> I think this is a P2-P1 interpolation issue. Not a bug. 

Kent


On ti., 2009-03-10 at 08:42 +0100, SyFi wrote:
> One or more new changesets pushed to the primary SyFi repository.
> A short summary of the last three changesets is included below.
> 
> changeset:   350:c6ea76d1118e208c841a2e2b31418a26767d524f
> tag:         tip
> user:        "Martin Sandve Alnæs <martinal@xxxxxxxxx>"
> date:        Tue Mar 10 08:43:12 2009 +0100
> files:       examples/newsolvers/Hyperelasticity1D/cpp/Makefile examples/newsolvers/Hyperelasticity1D/cpp/main.cpp examples/newsolvers/Projection1D/cpp/Makefile examples/newsolvers/Projection1D/cpp/Projection.ufl examples/newsolvers/Projection1D/cpp/SConstruct examples/newsolvers/Projection1D/cpp/main.cpp examples/newsolvers/projection/cpp/Projection.ufl examples/newsolvers/projection/cpp/main.cpp site-packages/newsfc/codegeneration/cellintegralcg.py
> description:
> Added 1D projection demo, seems to be some 1D issues.
> 
> 
> changeset:   349:b9b2793e6974ca95c7a8f07b2eb204dfbda9f31b
> user:        "Martin Sandve Alnæs <martinal@xxxxxxxxx>"
> date:        Mon Mar 09 23:35:30 2009 +0100
> files:       examples/newsolvers/Hyperelasticity1D/cpp/Hyperelasticity1D.ufl examples/newsolvers/Hyperelasticity1D/cpp/Makefile examples/newsolvers/Hyperelasticity1D/cpp/main.cpp examples/newsolvers/Hyperelasticity1D/cpp/manufactured_solution.py site-packages/newsfc/codegeneration/integralcg.py
> description:
> Updated Hyperelasticity1D demo.
> Manufactured solution computed with a swiginac script.
> Some parameters give a solution close to the analytical, some don't.
> Need a simpler equation to validate both the compiler and the main.cpp file
> (I have too many bad experiences to trust dolfin in its current version).
> 
> Some formatting fixes, using braces around some blocks
> which will enable convenient folding in many editors,
> as well as separate scopes (which makes this a good
> precursor to extracting blocks as functions).
> 
> 
> changeset:   348:2031cc4d0c72333bdb0c6617e023eb2bee4115cb
> user:        "Martin Sandve Alnæs <martinal@xxxxxxxxx>"
> date:        Mon Mar 09 18:14:05 2009 +0100
> files:       site-packages/newsfc/codegeneration/integralcg.py
> description:
> Formatting fixes in generated tabulate_tensor code.
> 
> 
> ----------------------------------------------------------------------
> For more details, visit http://www.fenics.org/hg/syfi
> _______________________________________________
> SyFi-dev mailing list
> SyFi-dev@xxxxxxxxxx
> http://fenics.org/mailman/listinfo/syfi-dev