Difference GUI and batch mode

- May 17, 2024 at 7:53 am
  
  Thies Blaakman
  Subscriber
  
  Dear all,
  I've been trying to run a parameter sweep using batch mode. I first tested some random datapoints I had in the GUI version, all of which gave a converging solution. Then, I tried to do the parameter sweep using batch mode. None of my analyses converged in batch mode. I tried running the same jobs I tested in the GUI in batch mode to see if there was a discrepancy between the solver in batch mode and GUI mode and it seems like this is the case (non-convergence and convergence). The errors I'm given are excessive distortion errors after which my simulation crashes. The errors themself don't really influence my results too much so having a few does not matter. I tried forcing different solvers in batch and GUI modes to see if this was the reason for premature termination of the analysis, but this was also with no succes. Is there an explanation for the difference in solution between batch mode and GUI? The only thing I can think of at this moment is that I run my batch using the -smp function whereas my GUI simulations are ran using distributed computing of type INTEL MPI. I hope someone can help me out.
  With kind regards,
  Thies
- May 17, 2024 at 10:23 am
  
  peteroznewman
  Subscriber
  
  What do you mean by batch mode?
  What was the benefit of batch mode you imagined over using Workbench and getting the results there?