Simulation domain volume and reduction weight sum differ

Create issue
Issue #640 closed
Ian Hinder created an issue

When I run qc0-mclachlan.par with the current trunk, I get this warning

WARNING level 1 in thorn CarpetReduce processor 0 host kop70.datura.admin
  (line 120 of /home/ianhin/Cactus/EinsteinToolkit/arrangements/Carpet/CarpetReduce/src/mask_test.c): 
  -> Simulation domain volume and reduction weight sum differ
INFO (CarpetReduce): Simulation domain volume: 432000
INFO (CarpetReduce): Reduction weight sum:     431999.996160507

This happens very frequently - more frequently than the CarpetIOBasic output. I think people who try the released ET with the example BBH parameter file will be concerned by this warning.

Keyword:

Comments (6)

  1. Erik Schnetter
    • removed comment

    This will then need to be mentioned and explained in the release notes.

    I assume that you are using CarpetMask, aren't you?

  2. Ian Hinder reporter
    • removed comment

    I am using Cactus/par/qc0-mclachlan.par, which uses CarpetMask. Could we remove the use of CarpetMask from this example parameter file, as well as mention the problem in the release notes? I'm uncomfortable with having such a prominent warning appear on every iteration. Do you know the cause of the problem?

  3. Erik Schnetter
    • removed comment

    The problem is that CarpetMask reduces the volume of the simulation domain by excision the interior of the black hole from the reduction operations.

    Without CarpetMask, many reductions will be useless and/or nan.

    We can add a check whether CarpetMask is active, and if so, reduce the severity or omit the warning.

  4. Ian Hinder reporter
    • removed milestone
    • removed comment

    I have added the information to the release notes. Removing release milestone.

  5. Log in to comment