Forum Discussion

RGroo3's avatar
RGroo3
Icon for New Contributor rankNew Contributor
6 years ago

FFT IP MATLAB C-model seems not bit true

We included the Intel FFT IP (altera_fft_ii) in our Arria 10 and ran into the following 2 issues. The first issue (bug?) is related to the data representation. The core is configured with “Single F...
EMijn's avatar
EMijn
Icon for New Contributor rankNew Contributor
6 years ago

Hi Chee Pin,

Thanks for your reply.

Regarding your questions:

  1. We are using Quartus Prime version 17.1.1. Internal Build 593 and the target device is Arria 10 (10AS066H2F).
  2. In the screenshots below you see for 6 sweeps the FFT outcome (represented as re^2 + im^2) and the associated absolute difference between the HDL simulation results and the Matlab model. Each sweep consists in this case of 128 samples. The max. absolute difference observed in this case is ~5*10^-6 at the peak value of the FFT, which is in this case ~80.

3. We were indeed expecting a bit-accurate Matlab model. We have this experience with the fixed-point version of the FFT IP and were also expecting this behavior when using floating-point.

The FFT IP user guide states that the Matlab model is bit-accurate: https://www.intel.com/content/www/us/en/programmable/documentation/hco1419012539637.html

4. The peak-frequency seems indeed identical, its magnitude almost...

Kind regards,

Erwin Mijnen