Photos: Secrets of early computing revealed by rediscovered EDSAC documents

Image 1 of 5

When volunteers began rebuilding one of the world's first fully operational general-purpose computers they had to painstakingly piece together the inner workings of the 65-year-old machine.
The original EDSAC - the Electronic Delay Storage Automatic Calculator - was built immediately after World War II at the University of Cambridge, where it aided research into areas including genetics, meteorology and X-ray crystallography. The machine's design was later used to create LEO, the world's first business computer.
With the historic EDSAC scrapped decades ago, the 20 volunteers recreating the computer had to scrape together information about the machine. For guidance the team consulted documents held by individuals and in libraries at the University of Cambridge, and examined the one surviving chassis used in the EDSAC.
Now the team working out of Bletchley Park in England, home to the famous World War II codebreakers that included father of computing Alan Turing, have had confirmation the rebuild is on the right track after stumbling across diagrams of the original machine.
The 19 detailed circuit diagrams, which came to light more than 60 years after they were drawn up, were delivered to the team by chance.
For decades the drawings sat in the home of John Loker, who worked as an engineer in the maths lab at the University of Cambridge and came across the diagrams in 1959, just after EDSAC had been decommissioned.
"In a corridor there was a lot of stuff piled up ready to be thrown away, but amongst it I spotted a roll of circuit diagrams for EDSAC. I'm a collector, so I couldn't resist the urge to rescue them," he said.
"It wasn't until I visited TNMOC [The National Museum of Computing at Bletchley Park] recently and learned about the EDSAC Project that I remembered I had the diagrams at home, so I retrieved them and gave them to the project."
Many of the diagrams, which date from between 1949 and 1953, were drawn after EDSAC had been constructed and are thought to have been an aid in refining the original machine and in designing its successor. They are believed to have been part of a much larger set of at least 150 drawings and are in remarkably good condition.
Andrew Herbert, leader of the EDSAC Project rebuild, said: "Thankfully, the documents confirm that the reconstruction we are building is basically correct, but they are giving us some fascinating insights about how EDSAC was built and show that we are very much in tune with the original engineers: both teams have been exercised by the same concerns.
"Importantly, the drawings clearly show that the aim of EDSAC's designer, Sir Maurice Wilkes, was to produce a working machine quickly rather than to create a more refined machine that would take longer to build. The refinements could come later -- and many did as the sequence of diagrams over the five-year period shows."
Above, from left, is EDSAC technical leader Chris Burton, with John Loker, the engineer who saved the diagrams, and leader of the EDSAC Project rebuild Andrew Herbert.

The National Museum of Computing
The schematics have also shed light on how the machine was refined after its completion. Changes included improvements to the instruction set and error correction so programming errors could be distinguished from machine malfunction, as well as redesigning the circuitry to obtain stronger signals.
While the volunteer team were mostly on the right track with the rebuild the diagrams did throw up some differences between the original and the reconstruction, namely in how the machine set up its initial instructions in memory, known as "initial orders".
The team will now modify the rebuild to reflect the instructions in the original diagrams.
In some respects it is remarkable the rebuild managed to so closely mirror the design laid out in the plans, as the few remaining parts of EDSAC that have helped guide the rebuild are thought to be left over from early prototypes of the machine.
"Very few artifacts of EDSAC remain. However these papers give a clue as to why a few such as a chassis do exist. We think that the existing artifacts were discards from a very early version of the machine," Herbert said.
The goal of the project is to recreate EDSAC as it existed on 6th May 1949, when it ran its first program.The reconstruction is due for completion by late 2015 and the ongoing reconstruction story featuring videos of progress at different stages can be seen at www.edsac.org.

Copyright Computer Laboratory, University of Cambridge. Reproduced by permission.
The computer was built from 12 racks spread over 20 square metres. The racks held the more than 3,000 vacuum tubes it used to carry out instructions and the mercury-filled tubes that acted as its memory.
The machine's principal designer and creator sir Maurice Wilkes, seen here on the left, led the team that built the original Edsac for the Mathematical Laboratory at the University of Cambridge. Wilkes died in November 2010 at the age of 97.
A key element of Edsac's design survives in computers to this day, the practice of storing both instructions and data in memory. The stored program concept was based upon the ideas devised by the brilliant mathematician John von Neumann in 1945.

Computer Laboratory, University of Cambridge
Edsac could carry out 650 instructions per second and had just 2KB of memory, which was provided by mercury-filled tubes. While obviously orders of magnitude less powerful than a modern PC it effectively computed some 1,500 times faster than the mechanical calculators it replaced.
Edsac read programs from a punched paper tape and printed its results. Here you can see the machine's five hole tape reader and Creed teleprinter.

Computer Laboratory, University of Cambridge
Edsac rebuild project manager Andrew Herbert credited the late professor Wilkes with having the vision and the drive to realise the potential of computers to carry out mathematical calculations that underpin scientific research.

Computer Laboratory, University of Cambridge

When volunteers began rebuilding one of the world's first fully operational general-purpose computers they had to painstakingly piece together the inner workings of the 65-year-old machine.

The original EDSAC - the Electronic Delay Storage Automatic Calculator - was built immediately after World War II at the University of Cambridge, where it aided research into areas including genetics, meteorology and X-ray crystallography. The machine's design was later used to create LEO, the world's first business computer.

With the historic EDSAC scrapped decades ago, the 20 volunteers recreating the computer had to scrape together information about the machine. For guidance the team consulted documents held by individuals and in libraries at the University of Cambridge, and examined the one surviving chassis used in the EDSAC.

Now the team working out of Bletchley Park in England, home to the famous World War II codebreakers that included father of computing Alan Turing, have had confirmation the rebuild is on the right track after stumbling across diagrams of the original machine.

The 19 detailed circuit diagrams, which came to light more than 60 years after they were drawn up, were delivered to the team by chance.

For decades the drawings sat in the home of John Loker, who worked as an engineer in the maths lab at the University of Cambridge and came across the diagrams in 1959, just after EDSAC had been decommissioned.

"In a corridor there was a lot of stuff piled up ready to be thrown away, but amongst it I spotted a roll of circuit diagrams for EDSAC. I'm a collector, so I couldn't resist the urge to rescue them," he said.

"It wasn't until I visited TNMOC [The National Museum of Computing at Bletchley Park] recently and learned about the EDSAC Project that I remembered I had the diagrams at home, so I retrieved them and gave them to the project."

Many of the diagrams, which date from between 1949 and 1953, were drawn after EDSAC had been constructed and are thought to have been an aid in refining the original machine and in designing its successor. They are believed to have been part of a much larger set of at least 150 drawings and are in remarkably good condition.

Andrew Herbert, leader of the EDSAC Project rebuild, said: "Thankfully, the documents confirm that the reconstruction we are building is basically correct, but they are giving us some fascinating insights about how EDSAC was built and show that we are very much in tune with the original engineers: both teams have been exercised by the same concerns.

"Importantly, the drawings clearly show that the aim of EDSAC's designer, Sir Maurice Wilkes, was to produce a working machine quickly rather than to create a more refined machine that would take longer to build. The refinements could come later -- and many did as the sequence of diagrams over the five-year period shows."

Above, from left, is EDSAC technical leader Chris Burton, with John Loker, the engineer who saved the diagrams, and leader of the EDSAC Project rebuild Andrew Herbert.

The National Museum of Computing

By Nick Heath

Nick Heath is a computer science student and was formerly a journalist at TechRepublic and ZDNet.

| Full Bio
| See all of Nick's content

Related Topics:

By Nick Heath

Join Discussion