1.1.2 - Types of Processor Flashcards
(8 cards)
1
Q
What are the key facts about CISC?
A
Complex Instruction Set Computing
- Each Instruction may take multiple clock cycles
- Instructions are different sizes (bits)
- Complicated processor design - circuitry is expensive
- More instructions than RISC (larger instruction set)
- Some instructions in CISC will rarely get used
- Uses less RAM
2
Q
What are the key facts about RISC?
A
Reduced Instruction Set Computing
- Each Instruction will take one clock cycle
- Instructions are the same size (bits)
- Simple processor design (Less transistors), so cheaper to produce
- Less instructions than CISC
- Instructions are used regularly (due to limited range)
- Produces less heat and uses less power than CISC
- Uses more RAM
3
Q
What are the key facts about a GPU?
A
GPU - Graphics Processing Unit
- Is a co-processor specialised in rendering graphics
- Specialist instructions available
- 1000s of cores that can apply the same instruction to multiple pieces of data simultaneously (SIMD).
- Suited to performing simple operations in parallel on larger data sets
4
Q
What are the benefits of using a GPU to process and display the graphics in computer games?
A
- More responsive gameplay
- Shorter loading/rendering times
- More realistic graphics
- Higher frame rates/smoother gameplay
- Higher resolutions graphics
5
Q
What are some non-graphical uses of GPUs?
A
- Data Mining (e.g, Supermarket Sales Data)
- Crypto Currency
- AI / Machine Learning
- Big Data (e.g. DNA sequencing)
6
Q
What is parallel processing?
A
- Executing more than one instruction simultaneously
- Can be achieved through multiple processors or distributed across multiple cores in a single CPU or GPU.
7
Q
What is a multicore system?
What are the benefits?
A
More than one processing unit in a single processor
- Each core can independently process instructions at the same time.
- Improves multi-tasking
- Enables parallel processing
- Increases the performance of the computer
8
Q
What are some limitations of parallel processing?
A
- Writing algorithms for parallel processing is more challenging.
- Some tasks cannot be done in parallel due to relying on each other’s output
