version 5.0 update 9.1

For fft plot, frequencies and amplitude are calculated and displayed.

The frequency increament should equal to FS/N, but it equals to FS/(N+2) in the plot, N = 2^k

Is this a bug of plot window?

version 5.0 update 9.1

For fft plot, frequencies and amplitude are calculated and displayed.

The frequency increament should equal to FS/N, but it equals to FS/(N+2) in the plot, N = 2^k

Is this a bug of plot window?

Hi CraigG,

I use a simple demo to show the frequency fault.

1. Delcare a global float array in a program and run simulation of a board (my board is adsp 21369 eval-kit).

**float data[4] = {1, 2, 3, 4};**2. Setup a new plot for "data"

3. In "Data processing" tab, use "FFT Magnitude" to get a fft plot of the data.

Sampling frequency (FS) is 10 Hz for instance.

4. In the plot, the frequencies given have mistakes.

The 3 fft bin (half + 1 of the data length) should be 0 Hz, FS/data_length=10/4=2.5 Hz and 5 Hz.

But in the figure, the frequencies are 0 Hz, 1.67 Hz ( FS/(data_length+2)=10/6 ) and 3.33 Hz.

I encounter this problem in vdsp 3.5 several years ago.

Now it still remains incrrect, so I think it should work better. Otherwise, I have to export data to matlab and plot by myself.

Hi Minidick,

Thank you very much for reporting this issue. I've reproduced this problem in my workstation: whatever the length N of plot data is (tried 4, 8, 16), the frequency step is always calculated by FS/(N+2) instead of FS/N.

I've filed this issue as a bug in our bugging tracking system and it should be fixed by our R&D team in next release version of VDSP.

Sorry for any conveniences caused by this problem.

Hi Minidick,

Thank you very much for reporting this issue. I've reproduced this problem in my workstation: whatever the length N of plot data is (tried 4, 8, 16), the frequency step is always calculated by FS/(N+2) instead of FS/N.

I've filed this issue as a bug in our bugging tracking system and it should be fixed by our R&D team in next release version of VDSP.

Sorry for any conveniences caused by this problem.