Computer Organisation (Short Answers 2pm)
Computer Organisation (Short Answers 2pm)
Computer Organisation (Short Answers 2pm)
(SHORT ANSWERS)
Q3) In a single bus CPU what are the steps to fetch a word from the memory.
Ans) In a single bus CPU, fetching a word from memory involves a coordinated dance
between the processor and memory:
1. Instruction Register (IR) Prep: The processor starts by placing the address of the
desired word (obtained from the program counter) into a special register called the
Instruction Register (IR). Think of IR as a note holding the memory location you want
to visit.
2. Address Broadcast: The processor broadcasts the address from the IR onto the
address bus, which acts as a highway for memory locations. This address essentially
tells the memory where to look for the data.
3. Control Signal: Simultaneously, the processor sends a control signal (often called
"Read") on the control bus. This signal informs the memory that the processor intends
to read data.
4. Memory Access: The memory receives the address and control signal. It then locates
the requested word and prepares it for transfer.
5. Data Transfer: The memory places the data word onto the data bus, which acts as
another highway for data itself.
6. Data Register Load: Finally, the processor reads the data word from the data bus and
stores it in a dedicated register, often called the Memory Data Register (MDR). This
register holds the retrieved data ready for use.
These steps occur in rapid succession, with each bus carrying the necessary information for
the memory fetch operation. The single bus design simplifies the communication but might
become a bottleneck for high-performance processors.
Q4) What are the fundamental design issues in designing an instruction set.
Ans) Designing an instruction set (ISA) involves balancing power and efficiency. Here are
some key issues to consider:
1. Instruction complexity:
o CISC (Complex): Offers powerful instructions that can do more in one go,
but can be slower to decode and less efficient.
o RISC (Reduced): Uses simpler instructions requiring multiple steps, but can
be faster to decode and allow for better pipelining.
2. Number of operands: How many data elements (registers or memory locations) an
instruction can handle at once. More operands might be convenient but could require
complex instructions.
3. Addressing modes: How instructions specify the location of data operands. More
options offer flexibility but can increase instruction complexity.
4. Instruction size: Fixed-length instructions simplify decoding for RISC processors,
while variable-length instructions in CISC can be more efficient for specific tasks.
5. Register usage: Emphasis on registers for operands (RISC) can speed up access
compared to relying heavily on memory (CISC).
The choice of design philosophy (CISC vs. RISC) depends on the target application. It's a
trade-off between instruction complexity, execution speed, and overall efficiency.