URL
https://opencores.org/ocsvn/dblclockfft/dblclockfft/trunk
Details |
Compare with Previous |
View Log
Line No. |
Rev |
Author |
Line |
1 |
36 |
dgisselq |
Here are the bench tests for the pipelined FFT. In general, there's a
|
2 |
|
|
`*_tb.cpp` file corresponding to every unit within the FFT. Feel free to
|
3 |
|
|
try them.
|
4 |
|
|
|
5 |
|
|
Be aware, however, the [fft_tb](fft_tb.cpp) doesn't truly
|
6 |
|
|
check for success--I just haven't gotten to the point of verifying that
|
7 |
|
|
the FFT result is *close enough* to the right answer in spite of actually
|
8 |
|
|
calculating the right answer. Instead, it creates a data file that can be
|
9 |
|
|
read in Octave via [fft_tb.m](fft_tb.m). That will show the first test output.
|
10 |
|
|
The second and subsequent outputs can be read via `k=k+1;` followed by calling
|
11 |
|
|
[plottst](plottst.m).
|
12 |
|
|
|
13 |
|
|
As another note (before I clean things up more), you'll need the `*.hex` files
|
14 |
|
|
in the same directory as the one you call [fft_tb](fft_tb.cpp) or
|
15 |
|
|
[fftstage_tb](fftstage_tb.cpp) from.
|
16 |
|
|
|
17 |
|
|
I expect the IFFT will work: it's just an FFT with conjugate twiddle factors,
|
18 |
|
|
although I haven't fully tested it yet.
|
© copyright 1999-2024
OpenCores.org, equivalent to Oliscience, all rights reserved. OpenCores®, registered trademark.