Manchester Baby Simulator

The Manchester Small Scale Experimental Machine (nicknamed the Baby) was built at the University of Manchester shortly after the Second World War. Heavily influenced by the ideas of John Von Neumann and company at the Moore School, Freddie Williams, Tom Kilburn and Geoff Tootill built the first operational stored program computer. This was a massive step in the development of the computer allowing programs to be executed from an electrical rather than an electro-mechanical store with the order of magnitude increase in speed that this brings. The machine was built primarily to test out the use of tubes as a storage device and is the first use of the Williams Tube as it later became known.


Background

The first program to be executed calculated the highest common factor of two integers and ran correctly on the 21st June 1948. After this date, the experimental machine was expanded upon and led to the development of the Manchester Mark 1 and later the Ferranti computers. In preparation for the 50th anniversary of the first program's execution, a replica of the original machine was built at Manchester by a team lead by Christopher Burton a member of the Manchester Computer Conservation Society. Christopher was of great help in advising on matters of historical accuracy for this simulator of the Baby.

The Baby's switch panel, taken in December 1948


Simulation

Unlike most other simulators that exist (including one I developed previously for RISC OS) this attempts to accurately simulate the switch panels that were used to laboriously program the machine along with many of the idiosyncracies that would have resulted from its incorrect use.

The program is written in java and requires a JVM supporting Java 1.2 or later to be installed. Full source code is included, as is a detailed user guide and an indepth discussion of the historical accuracy of the simulator. The simulator was written for the History of Computing course as part of my BSc in Computer Science at the University of Warwick.

The simulator's tube

The simulators's switch panel


future development

Alan Burlison has done some excellent work tidying up the source code and annotating it with javadoc comments with a view to perhaps pushing the simulator a little further. Alan has visited the replica machine and has some interesting comments and some very good photographs on his blog. One particular idea that we'd like to implement if we can find the time and resources is a photo-realistic user interface of the tube and control panels built up from high quality photographs of the replica machine.

Download

Discussion of the historical accuracy of the emulator 184 Kb PDF File
How to operate the Baby/Emulator 82 Kb PDF File
Emulator, Source code and documentation 317 Kb Zip Archive
Latest source code