WJEC A-Level Computer Science 601-5345-3
Hardware: Architecture, Memory and Storage
Networks and Data Transmission
Logic Gates and Boolean Algebra
Data Representation and Data Types
Data Structures
Organisation of Data: Files and Records
Database Systems
The Operating System
Algorithms and Programs
Principles of Programming
Systems Analysis
Software Engineering and Program Construction
Types of Software System
Practical Programming
Data Security and Integrity
Computing, Society and the Law
Advanced Data Structures and Logic
Recursion, Sorting and Searching
Programming Paradigms and Translators
System Design and Human-Computer Interaction
Software Engineering Tools and Professional Practice
Advanced Architecture and Assembly Language
Communication Networks and File Organisation
Number Representation and Floating Point
Databases and Distributed Systems
Operating System Management
Software Systems and Applications
Data Security, Cryptography and Attack
Number Representation and Floating Point
Truncation, rounding and accuracy
- 3 things to know
- A2 Unit 4
Overview
Absolute error, relative error and accumulated error.
What you need to know
Answer each one before you reveal it.
What is truncation?
Which introduces more error on average, truncation or rounding, and why?
Questions from this subtopic
Each one with the model answer underneath.
- 1
What is truncation?
Model answer
Discarding the digits that will not fit, without adjusting the digits that remain.
- 2
What is rounding?
Model answer
Adjusting the last retained digit according to the value of the digits discarded.
- 3
Which of the two introduces more error on average, and why?
Model answer
Truncation, because it always moves the value towards zero, so the errors accumulate in one direction.
- 4
Why does rounding produce a smaller average error?
Model answer
The adjustment is upward about as often as downward, so errors tend to cancel over many operations.
- 5
What is absolute error?
Model answer
The difference between the stored value and the true value.
- 6
What is relative error?
Model answer
The absolute error expressed as a proportion of the true value.
Other ways to work this
The same content, four other ways.
What is inside Levo
Six ways to work the same material.
Every question in Number Representation and Floating Point, on a schedule.
Attempt a question from memory, write the correct version, then attempt it again. Levo records the jump between your first attempt and your last, and brings each question back on the day you are due to forget it.
Or £46 a year, about £3.83 a month. Cancel anytime.