01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011 01101111 01101101

1.1.2 Types of Processor

01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001 01100101 01101110
01100011 01100101 01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011
01101111 01101101 01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001

Specification & Learning Objectives

A Level Specification point description
1.1.2a The differences between, and uses of, CISC and RISC processors
1.1.2b GPUs and their uses (including those not related to graphics)
1.1.2c Multicore and parallel systems
 Key question: What are the differences
01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011 01101111 01101101
01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001 01100101 01101110

between the RISC and CISC architectures?

01100011 01100101 01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011
01101111 01101101 01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001

• In CISC they have more complex hardware as there is a larger number of complex instructions of variable length
whereas in RISC it is made with more simpler hardware because it has a smaller number of simple standardised
instructions.
• Also in RISC they use a single clock cycle for instruction while CISC can take multiple clock cycles per instruction.
• In CISC they have multiple address modes available while in RISC there are fewer addressing modes available
• In CISC the compiler has to do less work because emphasis is on the hardware but in RISC the emphasis is on
the software therefore the compiler needs to do more work.
• CISC also cannot make use of pipelining but RISC can
• RISC requires heavy usage of RAM whereas CISC makes more efficient use of RAM than RISC
 Key question: How does having multiple
01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011 01101111 01101101
01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001 01100101 01101110

cores affect the speed of processing?

01100011 01100101 01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011
01101111 01101101 01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001

• Having multiple cores can have the benefit of having each core fetch decode and execute its own instructions. A
multicore system should process multiple instructions faster than one with a single core.

• However double the number of cores doesn’t necessarily double the performance as some programs cant make
maximum use of all cores.

• The performance is also improved by on-chip shared cache,intercore communication. It is important the core
can communicate with each other so they can work on different tasks working towards the same problem/goal.

• A multicore processor is a single chip containing 2 or more independent processing units -> each core can fetch-
decode-execute its own instructions

• Parallel processing is the processing of program instructions bu dividing them up between multiple processors or
processor cores
 Key question: What are the different characteristics of
01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011 01101111 01101101
01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001 01100101 01101110
CPUs vs GPUs, and what else besides graphics can GPUs be
01100011 01100101 01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011
used for?
01101111 01101101 01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001

CPU
CPUs are generic processing units and good all rounders whereas GPUs are highly specialised in what they do
CPUs excel at performing complex operations of small data sets
CPUs tend to be better for simple operations on larger data sets

GPU
GPUs are superior in speed and efficiency BUT only for certain tasks
They are initially used for rendering graphics but as technology advanced the large number of cores in GPUs have
been exploited to process many parallel streams of data at the same time ,no matter what that data might be
While GPUs have thousands of stream processors they typically run slower than a standard CPU core
GPUs can be used for; oil exploration, weather modelling ,machine learning ,linear algebra, statistical analysis ,stock
option predictions ,gravitational field theory

CPUs and GPUs are designed with two different goals in mind so they have different performance characteristics.
Although a typical CPU might perform one floating point operation quicker than a GPU ,a GPU will be able to
perform a thousand floating point operations much quicker due to the benefits of its bespoke design for parallel
processing
01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011 01101111 01101101

Typical exam questions

01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001 01100101 01101110
01100011 01100101 01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011
01101111 01101101 01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001

1. Modern games consoles come with GPUs. Describe the role of the GPU and explain how it works with the CPU to order to help speed up
processing. [4]
The role of the GPU is that it performs thousands or millions of simple calculations over and over again on multiple points of data (SIMD) for e.g. loading up a scene on
the game it renders the graphics while the CPU tends to work on the more complex operations of small data sets. As the GPU has the advantage of parallel processing

2. Explain two advantages of and two disadvantages of a CISC processor architecture over a RISC processor architecture. [4]
Advantages:
an advantage of a CISC is that they have multiple address modes available while in RISC there are fewer addressing modes available this makes it more efficient.
Also In CISC the compiler has to do less work because emphasis is on the hardware but in RISC the emphasis is on the software therefore the compiler needs to do
more work therefore it is easier to translate into assembly language
Disadvantages :
A disadvantage of CISC over RISC would be CISC also cannot make use of pipelining but RISC can therefore the CPU can work on maximum efficiency as it allows as one
instruction is being fetched the second is being decoded while the third is being executed. Another disadvantage of CISC would be that it costs more as it is physically
larger ,has more complex circuitry ,requires more silicon to make whereas RISC is smaller, less complex circuitry is required ,less silicon is needed to produce so it is
cheaper to make.
3. Some computer systems take advantages of multiple cores and parallel processing. Explain these concepts making reference to how they
help to improve a computers performance. [5]
4. Multiple cores mean that it is more efficient as each core can work on its own FDE cycle a multicore processor is a single chip containing two or more
independent processing units .A multicore system should process multiple instructions faster than one with a single cor. However if you double the number of
cores this doesn’t necessarily double the performance as some programs cant make maximum use of all cores as some are not designed with multiple core usage
in mind. Parallel processing is the processing of program instructions by dividing them between multiple processors or processor cores. The extent to which
parallel processing can speed up computation will be highly dependent on the task being carried out and if the program has been designed to make use of
parallel processing.

5.
Assessment Target: Overall grade:

Minimum expectations & learning outcomes

Terms 22-26 from your A Level Key Terminology should be included and underlined.
You must include a section which clearly compares the differences between, and uses of CISC and RISC processors
You must include a diagram which illustrates the different types of processor systems such as multicore systems, parallel systems and co-processor
systems (e.g. GPUs)
Answer the exam questions.

Feedback
Breadth Depth Presentation Understanding

 All  Link/Create new material  Excellent  Excellent

 Most  Explain/Compare/Contrast  Good  Good

 Some  Describe/Identify  Fair  Fair

 Few  Little shown  Poor  Poor

Comment & action Student response

01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011 01101111 01101101
01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001 01100101 01101110
01100011 01100101 01000001 00101101 01001100 01100101 01110110 01100101 01101100 00100000 01000011
01101111 01101101 01110000 01110101 01110100 01100101 01110010 00100000 01010011 01100011 01101001

Reflection & Revision checklist

Confidence Clarification
 Candidates need to understand the differences between the CISC and RISC processors and the key features and benefits of each.
 Candidates should be aware of the relative benefits of each architecture.
 Candidates need to understand what is meant by a parallel system and the benefits and limitations of parallel processing.
Candidates need to understand that parallel processing can be achieved through different (i.e. multiple processors in the same computer or distributed
 or multiple cores in a CPU or GPU).
 Candidates need to understand the benefits of a multicore system in terms of parallel processing and running multiple programs at the same time.
Candidates need to understand the purpose of GPUs and what applications they are used for (candidates need to understand how GPUs are used to
 aid graphics, but also other applications for example their use in modelling, data mining, etc.).
Candidates should understand the benefits and using GPUs and why they are suited to certain tasks (specialist instructions, multiple cores and SIMD
 processing).