(By Virgilio
Pasquali- November 2003)
In 1962, the Canadian Ferranti subsidiary,
Ferranti Packard, had quickly designed and produced the FP6000 computer, after
a lengthy visit to the UK to study Ferranti UK ideas (including Orion). The
first FP6000 was installed and running at the US Federal Reserve Bank in
In 1963 the Ferranti Computer Department in
the
The team concluded that
"There
are certain facets of the system we do not like. However, were we to begin
designing now a machine in the same price/performance range as the FP6000, we
would have in some 18 months' time a system that would not be significantly
better - if indeed it were any better - than the FP6000."
It was becoming clear that IBM was planning a compatible range of machines covering a large spread of processing power, and the Ferranti need to be able to deliver proven hardware quickly was probably the key factor in the decision to adopt the FP6000 design as the basis of the new Ferranti range of systems. An FP6000 was delivered to Ferranti West Gorton Labs, another to Ferranti Bracknell, and development work started on the new Ferranti medium system range, the 1900 (that later became the 1904/5).
From January 1963 ICT was exploring with Ferranti a merger of the Ferranti Computer Dept. with ICT. Very soon after the publication of the Ferranti FP6000 visit report a joint ICT/Ferranti party including Arthur Humphreys, “Echo” Organ and Tom Sheppard from ICT and Peter Hall and Hugh Devonald from Ferranti visited New York and Toronto to assess the FP6000. After September 1963, when the ICT shareholders approved the acquisition of the Ferranti Computer Dept. by ICT, development work started in ICT Stevenage on the1902/3, while ICT (Ex Ferranti) West Gorton Labs initiated development of the 1906/7 to complement the already ongoing development of 1904/5 (FP6000), both projects under the overall leadership of Charlie Portman.
Some key architectural changes were introduced to make the FP6000 design better suited to be the basis of a wide compatible range., the key ones being order code changes to allow expansion of memory capacity beyond 128K 6 bit characters (32K 24 bit words), initially in the 1906/7, and the introduction across the range of the ICT Standard Interface that had been specified by the ICT Stevenage Labs and RCA.
Confronted with challenging timescales, the teams of engineers from the development labs of the newly merged company worked together in an extraordinary spirit of cooperation and mutual respect to achieve immediate outstanding results. The ready acceptance of the Ferranti 1900 design by the ICT development engineers (especially the excellent ex-EMI people who formed the nucleus of those assigned to the work at Stevenage) and the ready adoption by the ex-Ferranti engineers of the ICT Standard Interface was, to say the least, unusual for those times.
The ICT engineers in Manchester West Gorton
and Stevenage were able to develop 2 working systems (1903 and 1904) to be
exhibited at the Business Efficiency Exhibition at Olympia (London) immediately
after the ICT 1900 Announcement of 28 September 1964 (some six month after the
IBM announcement of the 360). While the timescales achieved by both development
teams were short by the standards of the time, particularly noteworthy was the
achievement of the
ICT
1900 Announcement Summary (Simplified extract from ICT document Sept 1964)
ICT announced a range of seven 1900 system models (1902, 1903, 1904, 1905, 1906, 1907, and 1909), Programming Aids (Software) and peripherals.
In the announcement, great emphasis was placed on programming compatibility and portability between all models in the 1900 range (present and future) and modularity via the ICT Standard Interface (“Each 1900 system will grow with the customer’s needs by replacing individual modules as needed, without needing to replace the whole system”). Another key message was “value for money”. Storage discs (both fixed and exchangeable) were announced, but initially emphasis was placed on Magnetic Tape systems.
The first 1900 system (1905) was delivered to
Northampton CAT (
A year later the range was extended with a new system, the 1901, of approximately half the power of the 1902. The 1901 was a key addition to the 1900 range. It extended downwards the 1900 range into a price area that IBM normally could not reach with the 360 range. ICT was able to provide entry into its fully compatible range at a price well below the 360 level.
Subsequently, a system below the 1901 and one at the top of the range (the 1908) were investigated and defined, but they never went into full development for different reasons.
With the addition of the 1901, the viable span of the 1900 range was established, and this span continued in the subsequent manifestations of the 1900 range till the ICL top down announcement of the 2900 in 1974.
The 1900 was in continuous development during this period:
The first phase of development after the original 1900s, while keeping the same processor hardware technology, concentrated on improving key aspects of the architecture (see 1.4), and produced the 1904/5 E/F and 1906/7 E/F, replacing the 1904/5/6/7/9. Extended addressing, dual systems and the introduction of Segmentation and Paging as an alternative to Datum and Limit were the major improvements.
The subsequent phases of development, taking advantage of a stable and well understood architecture, applied state of the art technologies to the designs.