In addition to image processing and neural network, inverting the Geophysical Model Functions (GMFs) is one of the most widely used ways to retrieve oceanic parameters, i.e. surface wind speed, temperature, salinity, etc., from Synthetic Aperture Radar (SAR) data. More exact the description of the GMFs is, more accurately the results are obtained. For this problem, one can find two principal approaches: empirical (EP) GMFs based on in situ measurements and electromagnetic (EM) GMFs based on EM calculations of radar scattering from sea surface roughness. In order to explore the potentials of the GMFs for inverse problem in C-band, we compare in this paper radar scattering calculated by CMOD5.N (EP GMF), and TSM and SSA (EM GMFs). Likewise, we compare wind speed estimates by inverting the studied GMFs. Based on comparisons, several solutions may be proposed to improve the calculation of radar scattering, and then the retrieval of oceanic parameters.