The Raman effect has been widely used for years for carbon layers analysis for qualitative analysis of samples as well as quantitative. Two main lines of attitude to the quantitative analysis of the Raman spectra have been de¬ve¬lo¬ped. The first one is applied mainly in forensic investigations when one has pre¬pa¬red comparative samples which consist of known substances with known concentration. This is the ba¬se for comparison of spectra from unknown sample. A tool used in these in¬ves¬ti¬ga¬tions is the Voigt profile based on Faddeeva function, allowing a very precise com¬pa¬rison of spectra with elimination of individuality of me¬a¬s¬u¬re¬ment equipment. This at¬ti¬tu¬de seems not to be the best solution when one has not comparative samples and the sam¬p¬le un¬der test is unique – the situations often meet in scientific investigation. Using Voigt profiles may cause a loss of in¬for¬ma¬tion car¬¬ried with the spec¬trum, especially related with small quantities of nanotubes or ful¬le¬renes generated in the process of imposing diamond layers in plas¬ma arc. Another function for approximation shape of the Raman peak based on Breit-Figner-Fano curve is frequently used but difficulties arise when iden¬ti¬fy¬ing physical meaning of parameters and subtracting idle pedestal from the ex¬pe¬ri¬men¬tal spec¬t¬rum. A different attitude based on simple Gaussian peak approximation is pre¬sen¬ted as better solution. Use of this procedure as a metrological tool has been dis¬cus¬sed leading to the conclusion that engaging more advanced and so¬p¬hi¬s¬ti¬ca¬ted me¬thods does not provide automatically better results.