This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Extenden C28x floating point unit library soon?

Other Parts Discussed in Thread: CONTROLSUITE

The c28x floating point unit library has only got real FFT algorithms. I have to do inverese FFT and I need a inverse FFT or a complex FFT to do the inverse. Can anyone tell me if this library will be updated with new functionality like IFFT or CFFT in the near future?

 

  • This is on our to do list for the 2H of the year. 

  • Hi,

    I also need a complex FFT for the C28346. Can you give us an update on when TI might release the FPU library with complex FFT?

    Thanks.

  • Charles,

    Can you use the fixed point version?

    http://focus.ti.com/docs/toolsw/folders/print/sprc081.html

    Our software team has a re-release of the DSP libraries as a priority for the remainder of the year.  Working on it daily.

     

  • Hi Chris,

    Fixed point should work for my application, so I will give it a try.

    Thanks.

     

    - Charles

  • Hi Chris,

    The documentation for sprc081 doesn't explicitly state what the maximum FFT size is, but seems to only mention up to 1024 point FFT. What is the maximum size it supports?

    I need to do a 4096 point complex FFT to implement a correlation.

     

    regards,

    Charles

  • The FFT is for N points (this is in the readme one level up, the documentation should be more clear on this I agree), no issue doing 4096.  Make sure you allocate your SRAM appropriately to handle all the data.

     

  • CharlesOram said:

    Hi,

    I also need a complex FFT for the C28346. Can you give us an update on when TI might release the FPU library with complex FFT?

    Thanks.

    Charles,

    The next controlSUITE update will include the complex FFT for floating-point.  The update will be within the next couple of weeks.     If it is needed before then, I can see about putting it on an FTP drop for you.

    Regards

    Lori

     

  • Hi Lori,

    I am hoping to finish this project by the end of the month, so if you could arrange an FTP drop that would be great.

    Thanks.

     

    - Charles

  • Charles,

    I have uploaded a example project for CFFT test. You can download from the temp ftp as following instructions. If any problem, please feel free to let me know.

    To retrieve the files from the pickup directory:

      1. Open an ftp session to ftppickup.ti.com.
             ftp ftppickup.ti.com

      2. Login with the userid 'pickup'.
             Name: pickup

      3. Enter the password.
             Password: piurh!

      4. Change directories to the pickup directory.
             ftp> cd /pub/share/eemiixe

         Note: If you are using a graphical ftp client, you will not
               see the hidden pickup directory name eemiixe appear
               in the file list.  You will need to use a manual
               'Change Directory' or 'CD' command to change into the
               dropoff directory.

      5. Set the file transfer mode to ascii or binary as necessary.
             ftp> bin

      6. Transfer the file.
             ftp> get <file>

      7. End the ftp session.
             ftp> quit

    The files can also be retrieved with the url:

      ftp://pickup@ftppickup.ti.com/pub/share/eemiixe

    List of files in the pickup directory 'eemiixe':
    Example_CFFT_test.zip

    -Yu Cai

  • Thanks Yu Cai.

    Tthat's great. I have downloaded it and got it compiling and running on my C28346. Now I need to check that it is working correctly and then increase the FFT size to 4096 points.

    Is there any documentation for the library? The handling of the buffers (e.g. CurrentInPtr and CurrentOutPtr) is a bit confusing and the code comments and header file don't really tell me enough.

     

    - Charles

  • -Charles,

    Sorry, I haven't attached the document to this standalone project.  But you can get document from here,

    (by the way, the matlab code is inside the matlab folder. The structure of the matlab code is exactly the same as the assembly code. The CurrentInPtr and CurrentOutPtr explaination can be found inside the pdf file under newFPU4. You can also check the header of the assembly file of CFFT_f32.asm.

    Any question, please feel free to let me know. 

     

    The ftp pickup directory 'fojeexoo' was created on 10/19/2010 18:15:00.

    The files from the dropoff area have been moved to the pickup directory and are available for 7 days.

     

    To retrieve the files from the pickup directory:

     

      1. Open an ftp session to ftppickup.ti.com.

             ftp ftppickup.ti.com

     

      2. Login with the userid 'pickup'.

             Name: pickup

     

      3. Enter the password.

             Password: piurh!

     

      4. Change directories to the pickup directory.

             ftp> cd /pub/share/fojeexoo

     

         Note: If you are using a graphical ftp client, you will not

               see the hidden pickup directory name fojeexoo appear

               in the file list.  You will need to use a manual

               'Change Directory' or 'CD' command to change into the

               dropoff directory.

     

      5. Set the file transfer mode to ascii or binary as necessary.

             ftp> bin

     

      6. Transfer the file.

             ftp> get <file>

     

      7. End the ftp session.

             ftp> quit

     

    The files can also be retrieved with the url:

     

      ftp://pickup@ftppickup.ti.com/pub/share/fojeexoo

     

    List of files in the pickup directory 'fojeexoo':

    newFPU4.zip

     

  • it's for you convenience. The final version of CFFT code under controlSUITE is like this. But i am afraid you havn't installed the full package of controlSUITE so the version i gave you is a standalone version which is not heavily commented but it is easier to use.

  • Hi Yu Cai,

    Thanks. I have now got it working with a 4096 point FFT and was able to check the results using Octave and the matlab script provided.

    I really like the graphing feature in Code Composer Studio - very useful for checking the output.

    regards,

    Charles