Testing the speed of FFT/IFFT using the NVIDIA graphics cards

Abstract

The article shows the limits in the processing speed which are achieved by the modern NVIDIA graphics cards (GPUs). Also the level of the computational errors that registered with FFT and inverse FFT at different plans of their implementation is considered and the absence of distortion type jitter is fixed. A comparison of the speed of the spectral analysis of large data sets using different plans and GPUs NVIDIA GeForce 660M and 680 was made.