PIC & floating point numbers

weg22 · Joined: 08 Jul 2005 Posts: 91

Hello all,

I am having an issue with floating point numbers. If I define 4 floats,
float q1, q2, q3, q4;

and after some calculations they have values of:
q1 = 0.9224;
q2 = 0.345;
q3 = 0.0024;
q4 = 0.05;

the PIC I am using (16F84) will view q1, q2, q3, q4 as zero. This I figured because I didn't think it was able to work with floating point numbers. However, I thought that if I did something like q1*10, q1 would take an integer value of 9 and it would not be like multiplying 0*10 which would give me zero. But, that's what I seem to be getting. Does anyone know why?

Also, are there any PICs that can work with floating point numbers directly?

Thanks in advance,
weg

Neutone · Joined: 08 Sep 2003 Posts: 839 Location: Houston

Try this

q1*10.0

It should fix you type casting problem.

Guest · Posted: Tue Aug 02, 2005 8:32 am

Hi,
just post the code where you have your calculations, a bad casting of variables in floating point maths could resuls in a variable value of 0.

i.e. if you multiply q1 with an integer variable without explicitely casting this integer to float, you will have a result being an integer and not a float... and 0.xxx in integer format IS 0

int myInt=2;
float myFloat=0.213;

myFloat=myFloat*myInt will be 0

myFloat=myFloat*(float)(myInt) will be 0.426