Float rounding and truncate functions?

Projects Programming
1

Are there functions available for the Arduino to round or truncate floating point (real) values?

2

Just add 0.5 and cast to an int:

float f = 2.345f;
float f2 = 2.845f;

int rounded = (int)(f+0.5f); // roundedĀ  == 2
rounded = (int)(f2+0.5f); // roundedĀ  == 3
3

I am familiar with type casting to int and the inherent rounding to the whole number. I need more flexibility and functionality than round to a whole numbet. I need to round or truncate floating point values., ie round 2.56 to 2.6.

4

(At least in IDE version 1.6.5) there is a built in round() function. Use that with a pre and post multiplication by a power of 10.

  Serial.println(round(1.550 * 10) / 10.0);
  Serial.println(round(1.499 * 10) / 10.0);
  Serial.println(round(-1.499 * 10) / 10.0);
  Serial.println(round(-1.550 * 10) / 10.0);

... prints ...

1.60
1.50
-1.50
-1.60

5

Was about to suggest something along those lines, only issue is, the multiplication and division gives you the decimal place accuracy, so might be worth sticking it in a function that works out the multipliers based on the desired d.p

6

There are functions round(), floor() and ceil(), depending what you want to do.

7

guix:
There are functions round(), floor() and ceil(), depending what you want to do.

Is there a help link for them?

8

They are standard C++ functions. This page is about round and has links to the others.

http://www.cplusplus.com/reference/cmath/round/

9

There are functions round(), floor() and ceil(), depending what you want to do.

These won't help on their own, he stated he wants to round to a set decimal place, not just the integer part.

You'll need something along these lines:

 float round_to_dp( float in_value, int decimal_place )
{
	float multiplier = powf( 10.0f, decimal_place );
	in_value = roundf( in_value * multiplier ) / multiplier;
	return in_value;
}
10

Sorry, I missed this part :wink: