search

svalinn / DAGMC

Direct Accelerated Geometry Monte Carlo Toolkit
https://svalinn.github.io/DAGMC
Other
104 stars 65 forks source link

Update DAG-MCNP5 Test Case Reference Results #52

Open kldunn opened 11 years ago

kldunn commented 11 years ago

Some of the test cases for DAG-MCNP5 are producing results that are not consistent with the reference results. From the diff files I checked it seems to be primarily an ordering issue of the cell/tally numbers. The Meshtally/ test cases all seem to be working but the Dagmc/ test cases are not. This issue occurs for both the SVN repository and the GIT repository versions. Will need to update the DAG-MCNP5 test case reference results to be consistent for both 5.1.51 and 5.1.60 when the code has been shifted to its new repository.

kldunn commented 11 years ago

Per the discussion in #36, Meshtally/cases/stu_cyl and Meshtally/cases/stu_cyl2 currently use incorrect heating multipliers. These should be fixed and their corresponding reference results should also be updated when the new testing repository has been created.

kldunn commented 11 years ago

I have changed the tally multipliers for stu_cyl and stu_cyl2 from "fm -4 1" to "fm -1 0 -4 -1". The reference files have been updated and all of these changes currently exist on my forked version of svalinn/DAG-MCNP-TESTING under the meshtally_refactor branch.

kldunn commented 10 years ago

All VTK reference files for the Meshtally waf test cases have been updated on my forked DAG-MCNP-TESTING repo under the new 5.1.60 branch. These test cases were run using the latest version of DAGMC tallies, which includes Andy's new track length method.