An archived instance of discourse for discussion in undergraduate PDE.

Last Part of Instructions on Lab 2 Before the Exercises

clea

I don't know if anybody else noticed this but this is the initial state of the triangle that we formed with the hole punched inside of it. I was just curious as to why those random lines that look like cracks on the surface of the triangle appear. If anyone could explain this to me then I think you deserve an "Awesome" badge, regardless of what Mark thinks.


gigaliciousness

You mind posting your code? I didn't get those scratches/cracks (whatever we want to call them).

clea




Here ya go Giggey, I copied straight from the lab that Mark gave us in class. Sorry about the weird crack in the second graph, that is the product of me trying to screenshot my code and line it up perfectly...oops.


gigaliciousness

Thanks man! I'm writing a paper right now, but I'll get on it later tonight and try to get back to you!

gigaliciousness

So, essentially what is happening is simply that of a time-dependent numerical error that is happening within mathematica (numerical because we are doing a NDSolveValue function and Mathematica will occasionally round numbers such as 0.0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001 to 0 for obvious reasons).

If you change your time from 0 to 0.0000001 or the like, you'll see that these "cracks" - which are simply domains where Mathematica has incorrectly rounded to zero - will change significantly.

Hope this helps!