Return a normal number `y` and exponent `exp` satisfying `x = y * 2^exp`.

Return a normal number `y` and exponent `exp` satisfying `x = y * 2^exp`.

Split a double-precision floating-point number into a normalized fraction and an integer power of two.

The minimum biased base 2 exponent for a subnormal double-precision floating-point number.

The maximum base 10 exponent for a subnormal double-precision floating-point number.

The minimum base 10 exponent for a subnormal double-precision floating-point number.

The maximum biased base 2 exponent for a subnormal double-precision floating-point number.

The minimum biased base 2 exponent for a subnormal half-precision floating-point number.

The maximum biased base 2 exponent for a subnormal single-precision floating-point number.

The minimum base 10 exponent for a subnormal single-precision floating-point number.

The minimum base 10 exponent for a subnormal half-precision floating-point number.

The maximum base 10 exponent for a subnormal single-precision floating-point number.

Explore direct neighbors and limits of IEEE floating-point values.

Assessing the impact of subnormals on your CPU's performance

The maximum base 10 exponent for a subnormal half-precision floating-point number.

The minimum biased base 2 exponent for a subnormal single-precision floating-point number.

Split a single-precision floating-point number into a normalized fraction and an integer power of two.

The maximum biased base 2 exponent for a subnormal half-precision floating-point number.

⚙️ Calculate the maximum base-2 exponent for float16 constants, enabling precise numerical computation in web and Node.js environments.