Consider a 32 bit number cons.ts of two parts: a 24 bit Nebo. part and an 8 bit has-Ronal part
For instance the number:
0000 0000 0000 0000 0000 0111 0110 0000
Represents:
7.375
Ind. lab you will design a convergent rounder thuds different in functionality from the regular rounder that you are familiar with.
The convergent rounder intl. lab rounds the number based on fractional part ofthe number. Tbe following . tbe criteria [baryon need to consider when design Maddsrounder:
1. When the fractional part is larger than 0.5, the number is rounded up.
2. When the fractional part is less than 0.5 the number is rounded up.
3. Menthe fractional parcis 0.5, the number is rounded up if the whole part is odd and rounded down is the whole part is even.
Design a synthesirabie convergent rounder with the above specification.
Your design should work with both signed and unsigned numbers.
1. Draw a block diagram of the comp.te design that implemen6 your algoritbm.
2. Write a OHM code to model this divisicm by constant circuit
3. write a MR testbencb to verify your design. Presents a test vector that verifies the functionality of design for all critical case, Your tertbench should cover both signed and unsigned numbers.