System Architecture
Unit 1.1 covers the brain of the computer. Master the CPU, Von Neumann Architecture, Performance Factors, and Embedded Systems.
Concept Explorer: Performance Configurator
The performance of a CPU isn't just about speed. It's a combination of Clock Speed, Number of Cores, and Cache Size. Tweak the PC below to see how they combine to affect the processing bottleneck.
FDE Cycles per second.
Independent processing units (Parallel Processing).
Super-fast memory near the CPU.
Pedagogical Note: Increasing cores only helps if the software is written to support parallel processing. A 5GHz single-core might be faster than a 2GHz quad-core for linear tasks!
Modules
1. The CPU System
The Fetch-Execute cycle overview and the individual components (ALU, CU, Cache).
2. Von Neumann Architecture
The specific Registers (PC, MAR, MDR, ACC) and what they store.
3. Factors Affecting Performance
Deep dive into Clock Speed, Cores, and Cache levels.
4. Embedded Systems
Dedicated computers built into larger devices. Purpose and characteristics.
Interactive Practical Labs
Launch the web-based simulations to solidify your theory before examining.
Check Your Understanding
These questions are specifically engineered to test the biggest Examiner Traps where students lose marks on Section 1.1.
1. What is the explicit purpose of the Memory Address Register (MAR)?
2. The Program Counter (PC) increments by 1 during the Fetch stage. What does it store?
Written Exam Scenario (AO2/AO3)
"Company X is buying new computers. They are choosing between Computer A (3.8 GHz Single-Core) and Computer B (2.1 GHz Quad-Core). Discuss which computer is likely to be faster. Justify your answer." (4 marks)
It depends on the software being used. Computer B is a Quad-core, allowing for parallel processing (fetching four instructions simultaneously).
However, if the software is not designed to be split across multiple cores (i.e. it is linear logic), then three of the four cores in Computer B will sit idle.
In this scenario, Computer A would perform faster, because its single core completes 3.8 billion FDE cycles per second, compared to the active core on Computer B which only completes 2.1 billion cycles per second.
Examiner Tip: Never say "Quad Core is automatically 4x faster". This is a guaranteed dropped mark.