• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 41 subscribers
  • Views 1090 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 points in the CC3200

Umar Khan24
Prodigy 220 points
Other Parts Discussed in Thread: CC3200


Hi,

       I have a question about the floating point syntax for the CC3200. It seems that the CC3200 give the correct value for a particular syntax but a false value for another syntax which is equivalent.

I read the adc and print it out using the following command,

UART_PRINT("\n\r ADC Voltage = %4.2fV \n\r",adcVoltageValue);

However,

adcVoltageValue = adcReading*(1.4/4096);             always gives     65536.00

adcVoltageValue = adcReading*(14/10)*(1/4096);   always gives     0.00

and

adcVoltageValue = (((adcReading*14)/10)/4096);   gives the correct value.

Can any one please provide any pointers to what I may be missing. My floating point support is fpalib.

Thanks.

  • 0
    TI__Genius 16085 points
    Hi Umar,

    in adcVoltageValue = adcReading*(14/10)*(1/4096); -> (1/4096) is treated as integer division and that will result in 0. To force floating division operation you need to type cast either as double ie (((double)1)/4096).

    in adcVoltageValue = adcReading*(1.4/4096); type cast either as double as double hold more precise value than float.



    Regards,
    Aashish
  • 0 in reply to Aashish Bhanawat
    Prodigy 210 points
    Hi,
    What is the type of adcVoltageValue and adcReading? can you provide more details about your compiler?
    if adcVoltageValue is float , then:
    adcVoltageValue = adcReading*(1.4/4096); //should work as expected since most compilers treat float*int and float/int as float

    adcVoltageValue = adcReading*(14/10)*(1/4096); //will not work since (1/4096) is zero because both operands are integers ,also 14/10 = 1 for the same reason

    adcVoltageValue = (((adcReading*14)/10)/4096); //will work but not as accurate as it should be when adcReading is integer, because there will be integer approximation when the result of (adcReading*14) is divided by 10 and then when the result is divided by 4096

    the best approach is to use type casting as Aashish pointed.
    adcVoltageValue = (float)adcReading*1.4/4096.0;

    Good luck,
    Jaafar
  • 0 in reply to Jaafar Kh
    TI__Genius 16085 points
    Hi Umar,


    We are closing this thread, for follow up queries please open a new thread and add a link to this one for reference.


    Regards,
    Aashish