Recently, I figured out that there is a problem in separating the simulation and post-processing phases in my analysis. Since my simulation is quite big so I decided to separate the simulation phase and the post-processing one. But after visualizing the data in the post-processing phase I understood that after a certain point the responses are not reliable anymore and are actually ridiculous. There was nothing wrong with the simulation because after the end of the simulation I checked the response using SMTShowMesh and it was correct. However as I was also saving the responses during the simulation using the following command,


I understood that the responses of the simulation and post-processing are different after exactly 193 saves.

Has anyone had the same experience or it is just my problem? Is there any limitations for SMTPut[]?

Thanks in advance

  • $\begingroup$ I tried the example from documentation ("Perform Visualization Phase Separate from Analysis Phase") with many steps and everything works fine. Try saving your intermediate results with SMTPut[SMTData["Step"], {yourOtherVariables}], because the first argument (step number) represents the identification key and it is easier to distinguish in postprocessing phase than some time stamps close together. $\endgroup$ – Pinti Jun 15 '18 at 6:55
  • $\begingroup$ @Pinti Thanks for your response. Actually, yesterday after posting the question I reperformed my simulation as you are recommending now. But the problem still exists. This time I also saved the SMTShowMesh values during the simulation. In order to be able to compare the gifs of my responses that are created during the simulation with those of post-processing phase. If you want to see I can send the gifs and you can observe how the responses of the post-processing phase show ridiculous values after a certain step, exactly from step 194 again. $\endgroup$ – KratosMath Jun 15 '18 at 7:12
  • $\begingroup$ I recommend that you first install the latest AceFEM version (6.901). Then try to make your problem as small as possible (short code, short running time) and see if problem persists. If you manage to make a suitable MWE, then you can also edit your question to include it. $\endgroup$ – Pinti Jun 15 '18 at 8:23
  • $\begingroup$ @Pinti thanks a lot for your suggestions. I will try it out. $\endgroup$ – KratosMath Jun 15 '18 at 8:31
  • $\begingroup$ @Pinti is there any way to visualize the response of SMTShowMesh (of any field) for the previous steps without saving them? Because SMTShowMesh just shows the data of the current step. $\endgroup$ – KratosMath Jun 15 '18 at 9:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.