Which pair typically composes a floating-point number in binary systems?

• A. Sign and exponent only
• B. Sign, significand, and exponent
• C. Only a sign bit
• D. Mantissa and offset

Answer: (B)

Binary floating-point numbers are stored using three fields: a sign bit, an exponent, and a significand (often called the mantissa). The sign bit decides if the value is negative or positive. The exponent scales the magnitude by a power of two, and the significand provides the precision of the value. In normalized form, the leading 1 of the significand is typically implicit, which means the stored bits efficiently represent the significant digits. The overall value is computed as (-1) raised to the sign, times the significand, times 2 to the power of (exponent minus a bias). This combination of sign, significand, and exponent is what encodes a binary floating-point number. The other options miss essential pieces: a only sign bit can’t convey magnitude, only a sign and exponent omits the actual precision, and mantissa with offset omits an explicit exponent in the stored fields.