To floating-point binary
Assume 10 bits for the mantissa and 6 bits for the exponent. Give your answer as a normalised number.
Convert the denary number to floating-point binary: 0
Floating-point uses a mantissa and exponent to represent numbers in the form: mantissa × 2exponent
Normalised form: The mantissa should start with 0.1 for positive or 1.0 for negative numbers.
Steps:
- Convert the number to binary (including fractional part)
- Normalise by shifting the binary point
- Store the mantissa (10 bits) and exponent (6 bits in two's complement)
Floating-point binary represents numbers using scientific notation in binary. A number is stored as mantissa × 2exponent.
Normalisation
A normalised floating-point number has the binary point positioned so that:
- Positive numbers: Start with
0.1(the first significant bit after the point is 1) - Negative numbers: Start with
1.0(in two's complement form)
Method
- Convert the denary number to binary (whole and fractional parts)
- Normalise by moving the binary point to the correct position
- Count how many places you moved the point - this is your exponent
- Store the mantissa (10 bits) and exponent (6 bits in two's complement)
Example: Convert 6.5 to floating-point
6.5 in binary: 110.1
Normalise: 0.1101 × 2³
Mantissa: 0110100000 (pad to 10 bits)
Exponent: 3 = 000011 (in 6-bit two's complement) References:
- Wikibooks Computer Science A level (floating-point): AQA OCR
- Wikibooks Computer Science A level (normalisation): AQA
Further reading: