CS1011 introduces the basic concepts and terminology of Computer Science & Computer Systems. It covers a great breadth of material, but not very deeply, providing a framework that reveals the relevance of, and relationships between the other CS course units, which go into more detail.
We do not know of any book that covers the course, so you are not expected to buy any for CS1011. However, you might find some of the following helpful. All of them are in the Departmental Library and the Student Resource Centre, so you can read them there, or at least look at them before buying.
These are a readable introduction at about the level of an "Equinox" or
"Horizon" programme if you are not familiar with computers (one of them should
be recommended for CS1311):
"Computers" Long & Long
"Computers: Tools for an Information Age" Capron
"Essentials of Computing" Capron
"Computers & Information Systems with hands-on software tutorials"
Szymanski et al.
These go into more detail, but often about different examples than we use.
They may be helpful if you want a different viewpoint:
"Computer Science: an overview" Brookshear
"Introduction to Computing" Greenfield
This is recommended for CS1211 and CS1222, and has lots of information about
hardware, which is mainly missing from the other books:
"Principles of Computer Hardware" Clements
1623 Wilhelm Schickard (1592-1635), of T*:ubingen, W*:urttemberg (a friend of the astronomer Kepler), makes his "Calculating Clock" - a 6-digit machine that can add and subtract.
1786 J. H. M*:uller, of the Hessian army, conceives the idea of a
"difference engine": a special-purpose calculator for tabulating values of a
polynomial, which can approximate almost all useful functions.
e.g. for f(x)=$x sup 4$ we expect:
x= 0 1 2 3 4 5 . . . f(x)= 0 1 16 81 256 625 . . .If we repeatedly take the differences between adjacent pairs of numbers we get:
0 1 16 81 256 625 . . . 1 15 65 175 369 . . . 14 50 110 194 . . . 36 60 84 . . . 24 24 . . .Because the highest power of x in f(x) is 4, the 4th set of differences is constant.
A difference engine works by repeatedly performing additions to generate
values of a polynomial, for as many values of x as we want - we initialise
the values to the first diagonal of differences above (0,1,14,36,24):
(-> means the value is unchanged)
x: 0 1 2 3 4 5 6 7 8 f(x): 0 1 16 81 256 625 1296 2401 4096 D1: 1 15 65 175 369 671 1105 1695 2465 D2: 14 50 110 194 302 434 590 770 974 D3: 36 60 84 108 132 156 180 204 228 D4: 24 -> -> -> -> -> -> -> ->
initialise:
fx=0 D1=1 D2=14 D3=36 D4=24
repeatedly:
new fx = fx + D1
new D1= D1 + D2
new D2= D2 + D3
new D3= D3 + D4
D4 is unchanged
1822 Charles Babbage (1792-1871), of London, reinvents the difference engine & obtains government funding.
1832 produces a prototype segment, which operates on 6-digit numbers and 2 levels of differences. The complete engine was to use 20-digit numbers and 6 levels of differences.
Starting from x=0, moving in steps of 1, to calculate f(X) for an N-th order
polynomial requires X steps, each of N additions. A simple serial
method using a single adder would take XN time units.
A faster parallel method would do the N additions of each step
simultaneously, taking just X time units, but at each step we would need
both the set of "old" values and a set of "new" values to stop the additions
interfering with each other.
Babbage used a pipelined method that, like the serial method, needs only one set of values, but is nearly as fast as the parallel method, as it takes 2X time units. He split each step into two, first calculating every other value, and then calculating the remaining interleaved values:
initialise, and then repeatedly:
new D1= D1 + D2, new D3= D3 + D4, new D5= etc.
new fx = fx + D1, new D2= D2 + D3, new D4= etc.
(as before, keep the final difference constant)
x: 0 1 2 3 4 f(x): 0 -> 1 -> 16 -> 81 -> 256 D1: -1 1 -> 15 -> 65 -> 175 -> D2: 2 -> 14 -> 50 -> 110 -> 194 D3: -12 12 -> 36 -> 60 -> 84 -> D4: 24 -> -> -> 24 -> -> -> ->[the bold numbers are the previous initial values] 1836 Babbage produces the first design for his "Analytical Engine".
1842 Babbage's difference engine project is officially cancelled.
1853 Pehr George Scheutz of Stockholm, completes the first really useful difference engine, operating on 15-digit numbers and 4th-order differences, with a printer.
1858 used to calculate astronomical tables for an observatory in Albany, New York. A second similar machine had a long useful life in the British government.
1871 Babbage produces a prototype section of the Analytical Engine's "mill" (CPU) and printer. No more is ever assembled.
Babbage spent the last 40-odd years of his life designing and redesigning his difference and analytical engines. The implementation of the difference engine was never completed, and that of the analytical engine barely started.
Applications
Correctness
Automation
Algorithms
Notations
Analysis
Speed, Size, Cost
Technology
Techniques & Tools
Reliability
Infrastructure
Standardisation
We typically use a series of layers and transformations between them,
each dealing with a different level of detail/complexity.
e.g. customer requirements
-> problem specification
-> solution design
-> program implementation
-> machine instructions
-> execution on a computer
-> digital logic
-> circuits
-> transistors
-> quantum electromechanics
Layers are as independent as possible but some interaction is inevitable
Each layer/transformation usually consists of sub-layers & sub-transformations.
Some transformations can be automated,
others can't (yet)
Scientists try to show they understand something, by predicting what
will happen.
Engineers try things out to see what will happen.