Skip to content
Commits on Sep 1, 2010
  1. @mhanus

    Updates to tutorials with time dependence.

    They mainly ensure that the initial exact solution is used in the weak forms as long as possible (during the whole first time step) and is not replaced by a non-EXACT projection on the initial mesh.
    mhanus committed Sep 1, 2010
  2. Merge remote branch 'hanus/new-api' into change

    Conflicts:
    	tutorial/22-newton-timedep-heat-adapt/main.cpp
    	tutorial/24-newton-timedep-gp-adapt/main.cpp
    committed Aug 31, 2010
  3. @mhanus
Commits on Aug 31, 2010
  1. @mhanus

    Applies changes made to tutorial 24 also to tutorial 22.

    Still does not work, however.
    mhanus committed Sep 1, 2010
  2. @mhanus
  3. @mhanus

    Disable Newton's method on coarse mesh after global derefinement (24-…

    …newton-timedep-gp-adapt).
    
    Also moves visualization out of the adaptation loop.
    mhanus committed Sep 1, 2010
  4. @mhanus

    Updated comments in traverse.cpp.

    mhanus committed Aug 31, 2010
  5. @mhanus
  6. @mhanus
  7. @mhanus
  8. Updating examples 22 and 24.

    committed Aug 31, 2010
Commits on Aug 30, 2010
  1. @mhanus
  2. @mhanus
  3. @mhanus

    Corrections to neutronics-heat-conduction-adapt.

    Still fails after some adaptation paths, current version fails in Time step 3, adaptivity step 1, displaying
    E Invalid element ID 213, current range: [0; 85]
    mhanus committed Aug 30, 2010
  4. @Zhonghua
  5. Updating examples 22 and 24.

    committed Aug 29, 2010
Commits on Aug 29, 2010
  1. Updating example 24.

    committed Aug 29, 2010
  2. Fixing final time in test for example 23, long version.

    (The test was doing 300,000 time steps.)
    committed Aug 29, 2010
  3. Updating example 43.

    committed Aug 29, 2010
  4. Example 42 updated.

    committed Aug 29, 2010
  5. @mhanus
  6. @mhanus
  7. @mhanus

    Bugfix in neutronics-2-group-adapt benchmark and its test.

    Wrongly defined Ord version of forms.
    mhanus committed Aug 28, 2010
  8. Examples 40 and 41 work.

    committed Aug 29, 2010
  9. Problems with freeing spaces fixed, Valmor's test updated.

    DiscreteProblem cannot free spaces on destruction, since
    it is used for intermediary operations such as global
    projections. There is still an issue with NULL mesh
    pointer in example 40, so do not worry about the temporary
    control messages.
    committed Aug 29, 2010
  10. @Zhonghua
  11. @val-github
  12. Minor extension of Valmor's test.

    Valmor added to AUTHORS.
    committed Aug 28, 2010
  13. @val-github
  14. Rest of examples converted. Everything builds fine now.

    I believe that the Trilinos issues have been resolved,
    but an error was introduced into project_global() which
    now damages Space that comes there as an argument. Thus
    several examples and test fail now, but this should be
    easy to fix.
    committed Aug 28, 2010
Commits on Aug 28, 2010
  1. @certik
  2. @certik

    Don't run Cython for hermes_common

    It just uses the autogenerated files
    certik committed Aug 28, 2010
  3. @certik

    Add _hermes_common.cpp file

    certik committed Aug 28, 2010
Something went wrong with that request. Please try again.