I have a PXI-5412 arbitrary waveform generator generating an arbitrary waveform that is read by a PXIe-5122 Oscilloscope. In the picture attached the waveform labeled "1" (in red) is the graph of the waveform as read by the oscilloscope. The waveform labeled "2" is the waveform after being passed through an FFT node on the Host (how it should look). Then I have a PXIe-7846R FPGA module that I am transferring the waveform (pre-FFT) to through Peer-to-Peer streaming and back to the Host and graphing (graph labeled "3") to make sure the initial waveform looks as expected. The graph labeled "4" is the graph of the returned waveform from the FPGA after it is passed through an FFT Node in the Host VI. The graph labeled "5" is the waveform after being returned to the Host where an FFT was performed ON the FPGA instead of on the Host. My question is why the post FFT on FPGA waveform does not look the same as the post FFT on Host waveform (how it should look)?
Attached are the waveforms, my Host VI, my FPGA VI, my FFT settings, and the settings for the Peer-to-Peer FIFO and the DMA FIFO for sending and receiving the waveform from the FPGA.