The calculation of light diffraction is of great impor-tance for many essential optical applications, including optical lithography, optical tweezers, and super-resolution imaging, and is a research topic that has been extensively studied over the past few decades. Because both scalar and vector diffraction can be formulated as Fourier trans-forms, the standard method is to use the fast Fourier transform (FFT) algorithm. However, the use of the FFT requires a fixed sampling relation between the dis-cretization of the input field and that of the output field. For the vectorial case, this requirement often causes a significant waste of computational resources, which hin-ders real-time applications.