How to code "if calculation returns a negative value, result = 0"

Using Arduino Programming Questions
#1

Hi forum,

I am having trouble thinking through this simple problem, I am subtracting 2 variables, both have a range of 0 to 15 and I was using the constrain function to keep the result of the calculation between 0 and 15 (intending to code - if the subtraction results in a negative value then return 0) ie:

byte result = x - y;
result = constrain(result, 0, 15);

But in this code, if the result of the subtraction is a negative value I get the upper end of the constrain function (15) as the output where I thought I would get 0.

How would I code this argument correctly?
Thanks in advance!

#2

problem with byte is that if you subtract 10 from 5, you don't get -5, because bytes can't represent -5, they can only do 0 - 255, and if you go over the limit, it wraps around, so 5 - 10 = 250 (give or take one)
I would suggest the following code:

if (y > x) {
  result = 0;
} else {
  result = x -y;
}

or, if you want to simplify it, you could use the ? : expression:

result = (y > x) ? 0 : x-y;
[code]

[/code]

#3

The problem is that the byte data type

See Arduino - Byte

It does not hold negative numbers

Use the int data type

#4

Do you actually know that x is supposed to be the larger number ?

If you just want to know the difference between them, and you don't make the assumption that a specific one is the bigger one, I'd write

result = abs ( x - y );
#5

Thanks to everyone for the replies, all nice tips.

TRex:
problem with byte is that if you subtract 10 from 5, you don't get -5, because bytes can't represent -5, they can only do 0 - 255, and if you go over the limit, it wraps around, so 5 - 10 = 250 (give or take one)
I would suggest the following code:

if (y > x) {

result = 0;
} else {
  result = x -y;
}



or, if you want to simplify it, you could use the ? : expression:

result = (y > x) ? 0 : x-y;

[code]


[/code]

Thank you very much T-Rex, this is the perfect solution for this application. I didn't realise that byte loops back around to 255 when you pass 0 and that was causing the confusion so this ensures that I get 0 if the result is negative.

Thanks again!

#6