Floating point conversion

Instructions

This is a tool to experiment with a simplified floating point format described in CSCI 23000 at IUPUI. It is an eight-bit format with a mantissa and exponent in signed magnitude notation.

The format is ABBBBCDD where:

• A is the sign of the mantissa (0 for positive, 1 for negative)
• B is the mantissa, four digits of binary
• C is the sign of the exponent (0 for positive, 1 for negative)
• D is the exponent, two digits of binary

Enter values into the text fields, noting the program requires binary input of the given length. Incorrect inputs will be ignored. Hit the 'calculate' button to calculate the results of the current inputs.

Use the "generate a problem" button to generate a random set of legal inputs.

Use the "generate a decimal challenge" to get a decimal value. This value will have at least one solution, and perhaps more than one. You might want to write the number down and experiment until you can find a way to generate it.

Note that each binary input has exactly one correct result, but a decimal number could potentially have more than one solution in this floating-point notation.

This notation is far too limited for use in real life, but it is a good simplified example of how floating point notations work.