Math problem in ESP32?

Using Arduino Programming Questions
1

Hello. Why when performing the following mathematical operation with float, the result is -0.0?
I don't make sense of the "-" sign.
In the following code you can see the result:

float val1 = -0.1;
float val2;

void setup(){
Serial.begin(115200);
}

void loop(){
val2 = val1 + 0.1;
Serial.println(val2);
}

fallo float
fallo float1252×652 26.2 KB

2

What does it print when you increase the number of decimal places
Serial.println(val2, 8);
Leo..

3

In C++, a floating-point literal defaults to type double. On the smaller Arduino platforms like Uno, double is the same as float. But on ESP32, it's the normal size, 64 bits.

void setup() {
  Serial.begin(115200);
  auto f0 = 0.1;
  float f1 = -0.1;
  Serial.println(sizeof(f0));
  Serial.println(sizeof(f1));
  Serial.println(f1 + 0.1);
  Serial.println(f1 + (float) 0.1);
  Serial.println(f1 + 0.1, 15);
  Serial.println(f1 + (float) 0.1, 15);
}

void loop() {
  delay(10);
}

produces

8
4
-0.00
0.00
-0.000000001490116
0.000000000000000

on ESP32. If you want to work in float to save memory at the cost of precision, you need to use the f suffix, e.g. 0.1f

0.1 is about 0.100000000000000005551
0.1f is only 0.100000001490116119385

1 Like
4

Thanks for the answers.
Part of the code I use is the following:

float can2volt = 0.0F;

void setup() {
Serial.begin(115200);
}

void loop() {

 if (can2volt > 4.9) {
        can2volt =  -5.1;
      }
      can2volt = can2volt + (float) 0.1;
      //tft.drawString(String(can2volt, 1), 270, 71, GFXFF);
      Serial.println(String(can2volt,1));

delay(50);
}

Still showing -0.0
I have tried various combinations with (float) , F and the result is the same.
What is it that I am missing?

5

It is a rounding error. The result of your calculation is very close to zero but slightly negative. Hence the - sign.
Remember that 0.1 cannot be represented in float representation as precise as in decimal presentation. Like 1/3 cannot be represented in decimal format (at least not without an infinite amount of threes after the .
You could add an if to test for very small negative value and replace with 0.0...

if (x<0) {
  if (x>-0.0001) {
       x=0.0;
   }
}
1 Like
6

That this is not a math problem. The ESP32 and code are functioning as designed.

Why do you care that the result is a very tiny negative number?

7

See Signed zero - Wikipedia

The IEEE 754 standard for floating-point arithmetic (presently used by most computers and programming languages that support floating-point numbers) requires both +0 and −0. Real arithmetic with signed zeros can be considered a variant of the extended real number line such that 1/−0 = −∞ and 1/+0 = +∞; division is undefined only for ±0/±0 and ±∞/±∞.

Negatively signed zero echoes the mathematical analysis concept of approaching 0 from below as a one-sided limit, which may be denoted by x → 0−, x → 0−, or x → ↑0. The notation "−0" may be used informally to denote a negative number that has been rounded to zero. The concept of negative zero also has some theoretical applications in statistical mechanics and other disciplines.

8

Thank you very much everyone for the responses. And I stay clear. I have fixed it as indicated by build_1971.