Floating Point numbers not supported???

[runtime]

Hello,

I want to do a simple math division and get a result in decimal format, I'm reading a voltage with read_adc() which returns an integer, now I want to divide that number by 2 and get the result with 1 digit after the decimal point, if I try the following I get 'Floating point numbers not supported' from the compiler...

int TRead;
float CValue;

in main:

TRead = read_adc();
CValue = TRead/2.0;

fprintf(COM,"CValue: %3.1f \r\n",CValue);

any ideas?

Thank you in advance

[Ttelmah]

[Ttelmah]

[KerryW]

or you could do it like this:

int TRead;

in main:

TRead = read_adc();

fprintf(COM,"CValue: %u.%c \r\n",TRead>>2, (TRead&1)?'5':'0');

Which breaks down like this:

fprintf(COM,"CValue:
%u.%c \r\n" // print an unsigned float, a decimal point, a character, and a newe line
,TRead>>2 // print TRead divided by 2 (right shifted once)
, (TRead&1)?'5':'0'); // if TRead is odd, print '5', if it's even, print '0'

The last digit can only be 0 or 5 anyway. This method doesn't use any multiplication or divide routines. It will be very short and very quick.

[Alan Murray]

The problem is that when you divide 2 decimal numbers the compiler stores the result as a decimal thereby throwing away the remainder.
For dividing by powers of 2 shifting left works as suggested above.

However a much simpler option is to cast both divisor and dividend as floats before doing the division - this will yield a floating point result. e.g.

int adcReading;
float tempADC, result;

tempADC = (float)adcReading;
result = tempADC/2.0;

This will handle all integers - not just powers of 2.