On 4/7/22 11:27, Teo Collin wrote:
Thanks for your response! I've taken a look at those and I agree it changes
the computation of the integrals to use the correct mapping, but I don't see
how it changes the mesh representation. Does it? Or do I need to do something
else?
You need to describe
Thanks for your response! I've taken a look at those and I agree it changes
the computation of the integrals to use the correct mapping, but I don't
see how it changes the mesh representation. Does it? Or do I need to do
something else?
I feel like what I want is closer to step-49 - ideally I'd
On 4/7/22 10:46, Teo Collin wrote:
**
I am asking about the mapping. I know it won't make a difference in terms of
the accuracy of my solution, but it will make a difference in terms of
time/flops/memory/etc, which is what I want to see.
Then all you have to do is define a mapping class, and
I am asking about the mapping. I know it won't make a difference in terms
of the accuracy of my solution, but it will make a difference in terms of
time/flops/memory/etc, which is what I want to see.
On Thursday, April 7, 2022 at 12:22:04 PM UTC-4 Wolfgang Bangerth wrote:
> On 4/7/22 10:14, Teo
On 4/7/22 10:14, Teo Collin wrote:
Suppose I have read in a linear mesh of quads or hexes (via gmsh). How could I
create a higher order mesh that represents exactly the same mesh (i.e the
elements will be linear but represented with polynomials)? Is there an easy
way or step-x someone can poi
Hello all,
Suppose I have read in a linear mesh of quads or hexes (via gmsh). How
could I create a higher order mesh that represents exactly the same mesh
(i.e the elements will be linear but represented with polynomials)? Is
there an easy way or step-x someone can point me to?
I am not tryin
Hi Timo,
The output I get is:
#DEAL_II_WITH_HDF5 set up with external dependencies
#HDF5_DIR = /Users/fr714/dealii-candi-basic/hdf5-1.10.7
#HDF5_INCLUDE_DIRS =
/Users/fr714/dealii-candi-basic/hdf5-1.10.7/include
#HDF5_USER_INCLUDE_DIRS =
/Users/fr7
