I don't know what you would consider a "slam-dunk", but it sounds like the old allegations I had encounter years ago that you can't trust the adaptivity. I heard this a lot from NASTRAN/ANSYS users, so to prove them wrong, I made a small presentation to show that they all (p vs h) basically produce the same results. The attached file is certainly not a public reference, but makes a valid point on a simple model.
↧
