4 Replies Latest reply on Oct 6, 2010 11:05 AM by Anthony Botting

    Viewing intermediate results?

    Tony Gatta
    Tony Gatta Oct 2, 2010 3:51 PM

      Maybe this is a "wouldn't it be nice" comment -- but thought I'd ask anyway..

      any way to "preview" results from timesteps that have been computed already?

       

      Often find myself running long analysis (2+ hrs typically) .. sometimes only to find

      I've mis-defined (or omitted!) a critical restraint.  I've taken to stopping the simulation

      after a few steps and checking the results to verify the study is configured

      correctly.

       

      And stopping these simulations is no easy task!

       

      Thoughts?

       

      (we also run CFDesign for .. well, CFD.. and it does a nice job of updating the screen

      and making the results available upon completion of each iteration).

      • 793 Views
      • Categories: Simulation
      • Tags:
        • Re: Viewing intermediate results?
          Ryan Werner
          Ryan Werner Oct 4, 2010 9:38 AM (in response to Tony Gatta)

          Hi Tony,

           

          The only way I know of checking the progress of a study is to do exactly what it sounds like you are doing.  If I am running an Iterative Solver analysis I will just stop it and save results to hopefully get a little better idea of how things are progressing.

           

          Ryan W.

          • Re: Viewing intermediate results?
            Anthony Botting
            Anthony Botting Oct 5, 2010 10:46 PM (in response to Tony Gatta)

            Hi Tony: The nonlinear module is good for that - it has a "re-start" switch and "save data for restarting' switch you could use. I believe the re-start switch should really be labeled "continue from last time step" (to me, the term "re-start" seems to imply the analysis would start over again from the beginning). Sematics aside, if you are running a linear analysis, you can use the nonlinear module as it provides inherent saving of data at each step, anyway. I hope that helps. It's a nice function (turn off the option to use "large displacement" and and "large strain" you get a linear analysis solved by the nonlinear module). Cool beans.