refinement towards a node in hermes3d

[moritz braun]

Dear Pavel
I can report, that I managed to pull my private branch
up to the new hermes and will be able to push some of my changes soon.
I have now  "translated"  the refinement bug  example from 3D down to 2D
What I did is to create the same refinement in two dimensions for a simple harmonic oscillator problem using the bulitin refinement_towards_vertex and so to speak by hand by refining the four elements containing the vertex with
this gave the same eigenvalue to 15 digits . Thus in two dimensions the effect is the same, while in 3D
i have seen, that this attempt fails, snce the energy comes out below the exact value which must no happen
for a variational calculation.
I will be committing and pushing  this stuff in the next days

regards

Moritz

 

--
Prof M Braun         Tel.:27-12-4298006/8027
Physics Department  Fax.: 27-12-4293643
University of South Africa (UNISA)     
moritz...@gmail.com
P.O. Box 392    
0003
UNISA
 South Africa

[moritz braun]

Dear Pavel

I have finally managed to  push my new   stuff.
I didn't realize I had to type
git push origin HEAD
to indicate a new branch to go to my private github repo

My previous message describes some of the details of this change.

regards



Moritz

 

--
Prof M Braun         Tel.:27-12-4298006/8027
Physics Department  Fax.: 27-12-4293643
University of South Africa (UNISA)     
moritz...@gmail.com
P.O. Box 392    
0003
UNISA
 South Africa