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.

CCSTUDIO: CCSv5.3 has a problem with the svd function

Part Number: CCSTUDIO

Tool/software:

Hi Experts,

Can you assist this query from customer?

The software I use is CCS5.3 and the hardware is 6678, I used matlab coder to convert a c source file of svd function and ported it to CCS project, but during the running process I found that the running result has a big error with that of vscode, specifically, the first few rows of the feature matrix obtained by running the singular value decomposition of complex matrix in CCS Specifically, the first few rows of the eigenmatrix obtained by running the singular value decomposition of a complex matrix in CCS have almost no error, and the later the error increases. I would like to know why?

See attached file shared for detailed report:

error report.docx

Regards,
Archie A.

  • Hello,

    The software I use is CCS5.3

    Please note that CCS 5.3 is very old and unsupported. 

    I used matlab coder to convert a c source file of svd function and ported it to CCS project, but during the running process I found that the running result has a big error with that of vscode, specifically, the first few rows of the feature matrix obtained by running the singular value decomposition of complex matrix in CCS Specifically, the first few rows of the eigenmatrix obtained by running the singular value decomposition of a complex matrix in CCS have almost no error, and the later the error increases. I would like to know why?

    This is not really a tools specific question. Some one from the device team will need to take a look at this first.

    I will bring this thread to their attention.

    Thanks

    ki

  • Hello Ki,

    Thank you for your redirection.

    Regards,
    Archie A.

  • Hi, are they running the code on 6678? what board is this?

    thanks.