cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Anonymous
Not applicable
4,316 Views

FFT 7.1 C Model gives blk_exp error when using C_HAS_BFP = 1

Hello Folks,

 

I'm running the C Modelexample provided with the package.

 

I'm using my data. I used Scheduled Scaling, but when I'm going to test Block Fixed-Point I get an error:

 

ERROR: blk_exp is incorrect: expected 14, actual 7

Some outputs from simulation are incorrect

 

Output of FFT is unusable, and very far from results I obtain with Scheduled Scaling.

 

Why is this happening?

How can I solve it?

0 Kudos
5 Replies
Highlighted
Xilinx Employee
Xilinx Employee
4,313 Views
Registered: ‎11-28-2007

Re: FFT 7.1 C Model gives blk_exp error when using C_HAS_BFP = 1

The blk_exp checking in the C model example is only valid for the input data used in the original example. If you use your own data, the blk_exp will be different, so you will need to remove the blk_exp checking in your C model.

 

The core automaically adjusts scaling in block floating mode, so blk_exp output can be different for each frame depending on the input data. You will need to left shift the FFT output by blk_exp to reconstruct the output data.


@Anonymous wrote:

Hello Folks,

 

I'm running the C Modelexample provided with the package.

 

I'm using my data. I used Scheduled Scaling, but when I'm going to test Block Fixed-Point I get an error:

 

ERROR: blk_exp is incorrect: expected 14, actual 7

Some outputs from simulation are incorrect

 

Output of FFT is unusable, and very far from results I obtain with Scheduled Scaling.

 

Why is this happening?

How can I solve it?




Cheers,
Jim
0 Kudos
Highlighted
Anonymous
Not applicable
4,304 Views

Re: FFT 7.1 C Model gives blk_exp error when using C_HAS_BFP = 1

Ok, I'll remove that check.

 

But anyway, I'd be glad if you can teach me why blk_exp should be equal to nfft, if it is possible.

 

Thanks in advance

0 Kudos
Highlighted
Xilinx Employee
Xilinx Employee
4,294 Views
Registered: ‎11-28-2007

Re: FFT 7.1 C Model gives blk_exp error when using C_HAS_BFP = 1

The blk_exp doesn't always equal to nfft. It happens to be nfft in the provided example when the input is a constant 0.5.

 


@Anonymous wrote:

Ok, I'll remove that check.

 

But anyway, I'd be glad if you can teach me why blk_exp should be equal to nfft, if it is possible.

 

Thanks in advance




Cheers,
Jim
0 Kudos
Highlighted
Anonymous
Not applicable
4,285 Views

Re: FFT 7.1 C Model gives blk_exp error when using C_HAS_BFP = 1

Ok, I managed to use BFP.

 

One problem was I was using bad data.

 

Block Floating-Point seems to work now.

 

 

 

What is faster for the LogiCORETM IP Fast Fourier Transform v7.1,

to use a Scaling Schedule or BPF ?


0 Kudos
Highlighted
Xilinx Employee
Xilinx Employee
4,282 Views
Registered: ‎11-28-2007

Re: FFT 7.1 C Model gives blk_exp error when using C_HAS_BFP = 1

If by "faster" you mean Fmax, the difference between scaled and block floating point is negligible. You can easily generate two netlists and compare the Fmax yourself.

 


@Anonymous wrote:

Ok, I managed to use BFP.

 

One problem was I was using bad data.

 

Block Floating-Point seems to work now.

 

 

 

What is faster for the LogiCORETM IP Fast Fourier Transform v7.1,

to use a Scaling Schedule or BPF ?





Cheers,
Jim
0 Kudos