mlopez14
commented
9 years ago mrvortex, you bring good points and we are acutely aware about the absence of thorough documentation and detailed case tutorials.
The core developers of HiFiLES are Ph.D. candidates and post-docs whose main drive is to publish developments in numerics and simulation results. Consequently, in these early stages of the code, we will see that the main test-cases will be derived from published results in journals and conferences (the International Workshop on High-Order CFD Methods is coming up!) and the documentation will be spread throughout the code itself in the form of comments.
Current active users of HiFiLES have gone through the comments to understand enough to run their own simulations. As they encounter problems, they ping us and we help with the details.
Current development is focusing on increasing the robustness of the simulations, implementing moving grids, and expanding the theory of shock-capturing from triangles to all other elements in 2D and 3D.
We hope that, by enabling the running of Large-Eddy Simulations with unstructured grids in small GPU clusters, the first users who make the effort of going through the -now fairly steep- learning curve will find satisfaction in obtaining accurate results in complex geometries with vortex-dominated flows.
Hi, Thank you for sharing the code. But the documentation is not very sufficient. Can you guys provide some documentations about the Code structure and important classes and processes of the Code? Besides, Can you provide more 3D testcases? (such as sphere, SD7003 etc) Than you very much!