- Jul 04, 2024
-
-
-
Dmitri Naumov authored
[Tests] rotation test for elastic tensor See merge request ogs/ogs!5034
-
-
-
-
Dmitri Naumov authored
[T/CI] Allow vtk-osmesa >= 9.2.6 See merge request ogs/ogs!5038
-
- Jul 03, 2024
-
-
Dmitri Naumov authored
-
Dmitri Naumov authored
-
Dmitri Naumov authored
NumLib: Update setting of DirichletBCs in Newton See merge request ogs/ogs!4912
-
Dmitri Naumov authored
[Minor fixing] a not rendered equation in the web document CreepBGRa See merge request ogs/ogs!5036
-
- Jul 02, 2024
-
-
wenqing authored
-
Dmitri Naumov authored
-
Dmitri Naumov authored
Results are visually identical but not equal. Reason is the lesser number of iterations with the new method and the missing sigma/damage update in postTimeStep. Essentially the sigma/damage are from the previous iteration.
-
Dmitri Naumov authored
-
Dmitri Naumov authored
The results are visually undistinguishable. Stresses and loading curves are identical.
-
Dmitri Naumov authored
Otherwise "divergence is detected" in the first iteration of the first time step.
-
Dmitri Naumov authored
Otherwise the non-linear solver is stuck.
-
Dmitri Naumov authored
-
Dmitri Naumov authored
-
Dmitri Naumov authored
Instead of setting the solution in the Dirichlet nodes, the solution is enforced through the r.h.s. This usually saves one global iteration and more importantly avoids big jumps in the first iteration for mechanics processes.
-
Dmitri Naumov authored
-
Dmitri Naumov authored
The system solved by PETSc-SNES is different one and so far the methods could be reused. With upcoming changes to the Newton's version of applying Dirichlet-BCs this is not going to work anymore.
-
- Jul 01, 2024
-
-
Dmitri Naumov authored
Additional check for coordinate system and small refactorings. See merge request ogs/ogs!5033
-
Dmitri Naumov authored
Current error: /ParameterLib/CoordinateSystem.cpp(280): error C2672: 'checkTransformationIsOrthonormal': no matching overloaded function found /ParameterLib/CoordinateSystem.cpp(280): error C2784: 'void ParameterLib::checkTransformationIsOrthonormal(const Eigen::Matrix<double,Dimension,Dimension,0|Rows_==1&&Cols_!=1?Eigen::RowMajor:Cols_==1&&Rows_!=1?Eigen::ColMajor:Eigen::ColMajor,Rows_,Cols_> &)': could not deduce template argument for 'const Eigen::Matrix<double,Dimension,Dimension,0|Rows_==1&&Cols_!=1?Eigen::RowMajor:Cols_==1&&Rows_!=1?Eigen::ColMajor:Eigen::ColMajor,Rows_,Cols_> &' from 'Eigen::Matrix<double,3,3,0,3,3>' /ParameterLib/CoordinateSystem.cpp(26): note: see declaration of 'ParameterLib::checkTransformationIsOrthonormal' [609/644] Building CXX object Tests\CMakeFiles\testrunner.dir\Unity\unity_14_cxx.cxx.obj
-
Dmitri Naumov authored
-
Dmitri Naumov authored
avoiding STL iterators.
-
Dmitri Naumov authored
No measurable run-time difference.
-
Dmitri Naumov authored
-
Dmitri Naumov authored
-
Dmitri Naumov authored
-
- Jun 28, 2024
-
-
Dmitri Naumov authored
-
Lars Bilke authored
Default vtk 9.3.0 download misses ThirdParty/eigen/vtkeigen subdirectory. Use explicit download from GitHub-tag.
-
Lars Bilke authored
[ci] Update guix channels. See merge request ogs/ogs!5031
-
Lars Bilke authored
vtk 9.3 is now in upstream: https://issues.guix.gnu.org/70323
-
- Jun 27, 2024
-
-
Lars Bilke authored
-
Dmitri Naumov authored
[NL] Remove matrix/vector types from NR interface See merge request ogs/ogs!5027
-
Lars Bilke authored
Some CMake logic for building with conda See merge request ogs/ogs!5029
-
Lars Bilke authored
Install functionality is skipped on conda builds.
-
Lars Bilke authored
-
Lars Bilke authored
-