A proper evaluation of the building energy consumption implies an accurate assessment of the solar gains and consequently an exact estimation of the direct and diffuse solar radiation on tilted and oriented surfaces. The new standard EN ISO 52016-1: 2017 introduces a new hourly methodology for the calculation of energy performance of buildings and is complemented by EN ISO 52010-1:2017, which specifies the procedure for the conversion of climate data for energy calculations. These standards represent a good compromise between the old EN ISO 13790:2008 and Trnsys, since they allow for a more accurate evaluation of energy consumption than the former, while being less onerous than the latter. In the present work, EN ISO 52010-1:2017 is analysed for the first time and an alternative calculation procedure is proposed. These two algorithms are then compared with two other irradiance splitting methods implemented in Meteonorm and Trnsys code. In order to test performance and limits of the different calculation algorithms, we make a comparison among such methods on five European cities characterized by different climatic characteristics. Results show that the model implemented in EN ISO 52010-1:2017 overestimates the diffuse solar irradiance on horizontal plane for all the considered locations, while the proposed procedure and those implemented in the Trnsys and Meteonorm softwares show similar results. Using Trnsys algorithm as a baseline, the proposed calculation algorithm gives the best approximation among the considered models for what concerns the direct, diffuse and global irradiance on vertical surfaces exposed to south, east and west.