I've talked about it with a colleague reverse phone lookup of mine, and she told me that convergence has to be verified, but that in fact using too small timeteps can cause errors, if the solver is not efficient enough.
My question is nba reddit (and sorry for the long introduction): can I rely on the "" command in Scilab, pcpartpicker or is it better to know what's inside, to know if my answer makes sense? And is it better implemented in programs like Octave, Matlab or Scipy? Or are there better methods to solve an Ax=b system, more precise and robust?
Thank you for your answers!