User Portlet

I am nuclear physicist an I am using Mathematica for ages. However, i am trained on classical Fortran codes so my acceptance and use of symbolic programming is not on the level. I am still using old numeric logic, and I am using Mathematica as an auxiliary tool for solving and checking standard numeric calculations.

Right now I am heavily involved in minimizations, so I am using minimization codes like Minimize, NMinimize, FindMinimum, NonlinearModelFit, etc.... I am currently facing the problem of numeric precision of Mathematica using these codes.