todo-projeto-1.txt

15. getting wrong results for logg (star 1), and temperature is a bit off for star 2. why? ew minimization problem? not detecting ew well enough? it seemed to work okay for sisma spectra. then again, these had higher resolution. Is the minimization alg good? Is there any reason why particular lines would have consistently high residuals? Such that it might affect the parameters obtained? is it the determination of EW? using actual area gives really bad results. using formula seems to work better.

20. i think maybe the final uncertainty isn't a perfect estimator of error in parameters. i changed N, window size for the grid and suddenly i started getting different results - even though the uncertainty didn't change -- although it only changed one step, right? so it's not a real problem.