byte?

[rikotech8]

what is the difference between declaring:
int8 var_name
and
byte var_name

[Ttelmah]

Nothing.
Just other names.
In fact if you look in the include file for the processor, you will find the line:

#define BYTE int8

int = int8 = char = byte
on the PIC18/16.

Danger is though that if you switch to a PIC24, int is no longer in this list.

This is why in general it is 'better' practice to use:

intn sizes (like int8, int16, int32) for numeric values.

Then use 'char' for alphabetic text, and (possibly) use 'byte' for things that are being sent over 'byte wide' transport media (RS232 etc.).

This way you are confident about the sizes, and 'know' they are suitable for what is being done with them.....


Best Wishes

[rikotech8]

Thx a lot but what happens with int at pic24? As I understand it is no longer int8.

[ckielstra]

Sigh.... what is the part you didn't understand?
Yes, that is what Ttelmah wrote:

[Battery David]

I was poked in the ribs by a "real" programmer once to get me to be explicit on every variable type. Since an 'int' could be different on different parts, I now always do full declarations:

[asmboy]

Because of my ugly roots in ASM land - I second batterydavids reply.
I have ALWAYS explicitly declared my integer Vars as signed or unsigned
( except for int 1 ;-))

and it is a good way to keep your thoughts and your program clear.

[jeremiah]

It's also worth noting that for PIC24 at least, that set of equalities isn't always true. Some versions of the compiler in the late 120s or early 130's default char to signed and int8s to unsigned, which can snag you sometimes. I haven't checked to see if that is still the case