IEEE 754 Analyzer

Visualize how the computer stores real numbers. This analyzer breaks down 32-bit single precision floats into sign, exponent, and mantissa components.

S
Exponent
Mantissa / Fraction
Sign Bit
0
Exponent (Biased)
0
Hexadecimal
0x00000000

The Paradox of Digital Decimals

Unlike integers, real numbers (floats) cannot be represented exactly in binary because they can have infinite precision. The IEEE 754 standard solves this by using a form of scientific notation in binary: Value = (-1)^S * 2^(E-127) * 1.M.

This method provides a massive range of values but introduces subtle rounding errors that are the bane of every software developer.

The 0.1 + 0.2 Trap

In base-10, 0.1 is a simple fraction. But in base-2, it's a repeating decimal (like 1/3 in decimal). Because memory is finite, the computer must truncate this sequence.

> 0.1 + 0.2
0.30000000000000004

Special IEEE 754 States

NaN (Not a Number)

Result of undefined operations like 0/0. Represented by an exponent of all 1s and a non-zero mantissa.

Infinity (∞)

Represented by an exponent of all 1s and a zero mantissa. Can be positive or negative.

Denormal Numbers

Used to represent values extremely close to zero where the leading '1' of the mantissa is no longer assumed.

Memory Storage Breakdown

1 bit

Determines the positive or negative polarity of the value.

8 bits

The power to which 2 is raised, shifted by a bias of 127.

23 bits

The fractional part of the number, providing approximately 7 decimal digits of precision.