The PBF-LB (laser-based powder bed fusion) process is subject to a large number of variables, including the characteristics of the processed powder. Since a powder with a given specification can be supplied by various powder manufacturers, the transferability of optimized parameter settings and statistical processing models is of major interest. This work therefore investigates the processing windows of two Ti-6Al-4V powders supplied by different manufacturers following the Design of Experiments (DoE) approach. The fitted regression models for porosity and roughness demonstrate a significant influence of the powder and its size distribution. Further, the powder type significantly interacts with laser power, scanning speed and hatch spacing. It is shown that an increase of the powder size distribution quantiles by less than 10 µm leads to a shift of optimum settings towards a higher volume energy density by 6.4 J/mm³ as well as to higher roughness on the top and side surfaces.