CSIRAC computer as a Musical Instrument?

How Australia played the world’s first music on a computer

WE have just republished a most interesting article from The Conversation on this topic by Paul Doornbush.

It’s available from the Audio Legends menu tab (top of page) or directly at

Graeme Huon has also contributed this summary on CSIRAC:

Building a Computer

The CSIR department of radiophysics was established in the grounds of the University of Sydney in 1939.

The group already had considerable experience in general and switching electronics from their work on radar pulse systems during the war. 

In 1947 Trevor Pearcey convinced the head of the Division, Ed Bowen that this expertise should be used for (then) high speed computing.

Trevor Pearcey teamed up with Maston Beard to design and construct an electronic computing system. A third member of the team, Reginald Ryan, built the necessary dynamic data storage using recirculating mercury acoustic delay lines. Pearcey, in collaboration with Geoffrey Hill, then worked on developing a more detailed logic design, to fix an instruction set and to devise a practical programming scheme.

By November 1949 the basic units were completed and the first test program run. The result was the first automatic electronic computer in Australia and one of the earliest in the world. This was the CSIR Mk1, later renamed CSIRAC. 

Audio output was used to alert the operator to program operation without having to continuously watch the machine. Successful operation and even particular routines had a certain sound.

First test programs were run in late 1949, and CSIR Mk1 was developed to a stage where it was in restricted operation in late 1950 and was demonstrated during the first Conference on Automatic Computing Machines held on 7-9 August, 1951. One of the demonstrations included a performance of computer generated music.

From 1951 to mid-1955 CSIR Mk1 was employed in the Division of Radiophysics, in part to support the cloud-physics and radioastronomy projects including for study of raincloud seeding, as well as a tool for developing programming techniques. It also provided a computing service for other divisions of CSIRO, to other universities, and a variety of other research, design and engineering organisations.

In 1955 the computer was dismantled and transported by road to the University of Melbourne where it was reassembled and on 14 June 1956 formally recommissioned and named CSIRAC.

Neville Thiele told the story of his mate Prof. Doug Lampard (Monash Electrical engineering) coming over City Road to Engineering, and drawing his attention to this machine when they (and it) were “installed” as undergraduates at the University of Sydney.  Neville was particularly interested in the mercury delay lines used as recirculating data storage.  When told that the machine was later residing in Melbourne, he said “I would not be game to turn it on now”. 

Incidentally, Neville and Doug were jointly interested in university stage show production particularly during the second year of their course. Neville was being badgered by Doug to rebuild the stage amplifier they were using. Neville was not really interested so Doug forced the issue by disassembling it, triggering a rather hasty rebuild to meet a production deadline.


Graeme Huon
July 2016