Eurocode 8 spectra are scaled by the peak ground acceleration values that are defined for the given site through a probabilistic seismic hazard analyses (PSHA). However, spectra that are created by combining the empirical shapes with the single PSHA-defined scaling factor will not represent the so-called Uniform Hazard Spectra (UHS). Moreover, the very shape (Type 1 or Type 2) of the Eurocode 8 elastic spectrum is selected with respect to the magnitude of the earthquakes “that will contribute most to the seismic hazard defined for the purpose of probabilistic hazard assessment…”. Such definition is somewhat obscure since these “most contributing” earthquakes are, even at the same site and for the same probability i.e. for the same “return period”, different (in a general case) for different vibration periods, while the whole Eurocode 8 spectrum is scaled by using only the PSHA estimate of the peak ground acceleration. In this Paper we present an illustrative example of the Uniform Hazard Spectra for the city of Belgrade and compare the obtained UHS spectra, as well as the scenario empirical spectra scaled for different earthquake parameters, to the corresponding Eurocode 8 spectra, further pointing out the intrinsic ambiguities in the current EC8 suggestions for creation of design spectra. [Projekat Ministarstva nauke Republike Srbije, br. 36043