• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 99 subscribers
  • Views 492 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Floating point to fixed point conversion

gil albo
Intellectual 725 points

Hi All,

regarding floating to fixed point conversion,

i dont manage to find something suitable that will explain all the considerations,such as accumulator size,etc'...

what i find usually is how to simulate this on matlab.

 

i want good explaination and examples.

 

thanks a lot for the help

Gil

 

  • 0
    TI__Mastermind 30825 points

    Gil,

    Is this question director towards C66x based devices?  The C66x devices fully support floating point instructions and there would not be a need for float to fixed point conversions.

    If you're referring to another fixed point device, then this would be more of a software question as the HW itself doesn't perform a conversion, it's done via a Software Routine.

    Best Regards,

    Chad

  • 0 in reply to Chad Courtney
    Intellectual 370 points

    Hi Gil,

     

    If you have not tried it you can try with IQ math libraries offered by TI. I used it for floating point calculation on fixed point device C6472. It helped. Further, calling the functions as inline calls reduces the function calling overheads also and speeds up the execution.The setup for the libraries and the user guide(e.g. SPRUGG9 as applicable for your device) are available in the TI website.

     

    Regards,

    AC.