1.1.2 // Types of Processors: CISC & RISC, GPUs, Multi-Core/Parallel Systems

Question 1

What does CISC stand for?

Answer 1

Complex Instruction Set Computers

Question 2

What does CISC stand for?

Answer 2

Reduced Instruction Set Computers

Question 3

What are CISC, and describe their instructions.

Can I give an example?

Answer 3

• CISC contain a large set of instructions used to accomplish tasks in as few lines of assembly as possible.
• A CISC instruction can combine a “load/store” instruction with the instruction that carries out the actual calculation.

For example, the line ‘MULT A, B’ could be used to multiply A by B, then store the results back in A.

Question 4

What are RISC, and describe their instructions.

How long does each line take? What about a jump/branch command?

Answer 4

• RISC instructions are reduced to very simple operations
• Because of this, each line takes one clock cycle; This means pipelining can be facilitated to up processing. This happens by fetching one instruction, and while decoding it, fetching the next one etc etc.
• However, during a jump/branch command, the pipeline is flushed and remade again.

The minimal number of instructions are used to accomplish the required operations from multiple GPRs.

Question 5

List some Advantages of CISC.

Answer 5

1. Programs can be coded quicker
2. Meaning, the Compiler has relatively less work to do to translate to Machine Code.
3. Because code is relatively short, little RAM is needed to store instructions.

Question 6

List some Advantages of RISC.

Answer 6

1. The Hardware is simpler to build, requiring fewer circuits for carrying out complex instructions.
2. Pipelining is possible because each instruction takes the same amount of time to complete (1Hz.)
3. Because RAM is relatively cheap, you can afford larger caches. Larger instruction caches mean higher hit rates, increasing overall processing.

Question 7

What is Multi-Core Processing, and how must a computer take advantage of it?

Multi-Core Processors are also known as Parallel Systems; But why?

Answer 7

• M-CPs are able to distribute a workload across multiple processor cores; Thus achieving significantly higher processing performance by performing several tasks in Parallel.
• A single OS coordinates these tasks (as long as the software is written to support multiple cores.)

Many computers have 2-8 cores; Supercomputers have thousands of general-purpose cores.

Question 8

What is a Co-Processor?

Answer 8

• An extra processor used to supplement the functions in the Primary Processor (CPU.)
• It may be used to perform Floating-Point Arithmetic, Graphical, and Digital Signal Processing.

Question 9

Describe the Structure and Function of the GPU.

Note that the GPU is a Co-Processor, as it supplements the CPU with graphical processing.

Answer 9

Graphical Processing Unit: It’s very efficient at manipulating computer graphics and image processing.

• It consists of thousands of small, efficient cores designed for Parallel Processing.
• It can process large blocks of visual data simultaneously.
• Because of this, it can cooperate with the CPU to accelerate scientific calculations.

GPUs can also be used in Machine Learning/AI because it excels in performing a large number of simple calculations simultaneously.
In a PC, a GPU may be present inside a Graphics Card.