# <float.h>

**[<float.h>](https://en.cppreference.com/w/c/header/float)** defines platform-specific floating-point limits: `DBL_EPSILON` (smallest difference from 1.0), `DBL_MAX`, `DBL_MIN`, and precision constants. Essential for writing numerically robust code across platforms.

Use `DBL_EPSILON` as the basis for floating-point comparisons instead of exact equality.

## Example

This example demonstrates machine epsilon and shows the limits of floating-point representation.

```c
// compile: gcc -o floatexample floatexample.c
// run: ./floatexample
// description: query floating-point limits and test epsilon

#include <float.h>
#include <math.h>
#include <stdio.h>

int main() {
    printf("DBL_EPSILON = %.2e\n", DBL_EPSILON);
    printf("1.0 + DBL_EPSILON != 1.0:   %s\n",
           (1.0 + DBL_EPSILON != 1.0) ? "true" : "false");
    printf("1.0 + DBL_EPSILON/2 != 1.0: %s\n",
           (1.0 + DBL_EPSILON/2 != 1.0) ? "true" : "false");
    
    printf("DBL_MAX = %.6e\n", DBL_MAX);
    printf("DBL_MIN = %.6e\n", DBL_MIN);
    printf("DBL_DIG = %d decimal digits\n", DBL_DIG);
    
    return 0;
}
```

## Common macros

**Double precision:**
- `DBL_EPSILON`: smallest x where 1.0 + x != 1.0
- `DBL_MAX`, `DBL_MIN`: largest/smallest representable values
- `DBL_DIG`: decimal digits of precision
- `DBL_MANT_DIG`: mantissa digits
- `DBL_MAX_EXP`, `DBL_MIN_EXP`: exponent bounds

**Float versions** (FLT_* prefix):
- `FLT_EPSILON`, `FLT_MAX`, `FLT_MIN`, etc.

**Long double versions** (LDBL_* prefix):
- `LDBL_EPSILON`, `LDBL_MAX`, `LDBL_MIN`, etc.