Control Data Corporation

After purchaing ERA, Remington Rand kept the ERA team separate from the UNIVAC developers. Rand was most interested in the magnetic drum memory systems that ERA had pioneered. By the mid-1950s, Remington Rand's initial leadership with Univac was being whittled away by IBM machines that used fewer vacuum tubes.In 1955, the Sperry Corporation bought Remington Rand, forming the Sperry Rand Corporation, headed by Retired General Douglas MacArthur. The new corporation combined the Univac and ERA teams into one division, with Norris as its leader.

At first, the ERA engineers provides support for a variety of projects. Eventually, ERA was assigned to the UNIVAC II Sperry Rand wanted to use the ERA Atlas as the basis of the Univac 1101 computer. Sperry Rand then developed three Univac 1102 units for the US Air Force. They were followed by the 36-bit UNIVAC 1103. The UNIVAC projects were plagued by lengthy delays and rivalries between the UNIVAC and ERA engineers.

6600 design

Meanwhile, at the new Chippewa Falls lab, Seymour Cray, Jim Thornton, and Dean Roush put together a team of 34 engineers, which continued work on the new computer design. One of the ways they hoped to improve the CDC 1604 was to install better transistors. Cray selected the new silicon transistors that were developmed using the planar process. Manufactured by Fairchild Semiconductor,these silicon transistors were much faster than the germanium transistors in the 1604, without the drawbacks of the older mesa silicon transistors. The speed of light restriction forced Cray to employ a more compact design, with refrigeration designed by Dean Roush.

While designing the 6600, CDC set up Project SPIN to supply the system with a high speed hard disk memory system. At the time it was unclear if disks would replace magnetic memory drums, or whether fixed or removable disks would become the more prevalent. SPIN explored all of these approaches. The team eventually delivered a 28" diameter fixed disk and a smaller multi-platter 14" removable disk-pack system. Over time, the hard disk business pioneered in SPIN became a major CDC product line.

6600 release

In 1963, CDC introduced the CDC 6600 mainframe computer. It out-performed its competition by roughly ten times. When CDC sold over 100 6600's at $8 million ($83 million in 2025 dollars) each; it was considered a supercomputer.

The 6600 had a 100ns, transistor-based CPU (Central Processing Unit) with multiple asynchronous functional units, using 10 logical, external I/O processors to off-load many common tasks and core memory. That way, the CPU could devote all of its time and circuitry to processing actual data, while the other controllers dealt with the mundane tasks like punching cards and running disk drives. Using late-model compilers, the machine attained a standard mathematical operations rate of 500 kiloFLOPS, but handcrafted assembly managed to deliver approximately 1 megaFLOPS.

Lawrence A. Liddiard and E. James Mundstock at the University of Minnesota developed a FORTRAN compiler, known as MNF (Minnesota FORTRAN), in 1970 for the 6600.

7600

In the same month it won its lawsuit against IBM, CDC announced its new supercomputer, the CDC 7600 (previously referred to as the 6800 within CDC). This machine's hardware clock speed was almost four times that of the 6600 (36 MHz vs. 10 MHz), with a 27.5 ns clock cycle, and it offered considerably more than four times the total throughput, with much of the speed increase coming from extensive use of pipelining. The 7600 memory had a split primary- and secondary-memory which required user management but was more than fast enough to make it the fastest uniprocessor from 1969 to 1976.

The 7600 did not sell well for a number of reasons; it was introduced during a 1969 downturn in the U.S. economy, the complexity in its construction hindered its reliability and it was also not fully compatible with the 6000-series. It also required a completely different operating system from the 6000-series; which like most new OSs, was primitive. The 7600 project paid for itself, but damaged CDC's reputation. A few dozen 7600s were the computers of choice at supercomputer centers around the world.

8600

In 1968, Cray began designing the CDC 8600 supercomputer. This design included four processors similar to those in the CDC 7600, packaged in a single, smaller case. The smaller size and shorter signal paths allowed the 8600 to run at much higher clock speeds. Combined with its faster memory, the 8600 had good performance gains.

However, the 8600 had significant problems. In manufacturing its circuit boards, Cray chose to solder individual components to the board. The design of the 8600 made it impossbile to cool effectively. As a result, the hot-running solder joints degraded the computer's reliability. The compact design also complicated maintenance of the 8600.

Cray soon recognized that the 8600 needed a redesign. However, Norris believed that CDC did not have enough research and development funding for an 8600 redesign along with the development of the new STAR-100. In 1972, Cray left CDC to form Cray Research to build supercomputers. Norris remained a staunch supporter of Cray and even invested in Cray Research.

Timeline of systems releases

CDC 1604 et al. – 1604, 1604-A, 1604-B, 1604-C, 924 (a "cut down" 1604 sibling) * CDC 160 series – 160, 160A (160-A), 160G (160-G) * CDC 3000 series – 3100, 3200, 3300, 3400, 3500, 3600, 3800 * CDC 6000 series – 6200, 6400, 6500, 6700 * CDC 6600 * CDC 7600 * CDC CYBER – 17, 18, 71, 72, 73, 74, 76, 170, 171, 172, 173, 174, 175, 176, 203, 205, Omega/480, 700 * CDC STAR-100

1957 – Founding

1959 – 1604

1960 – 1604-B

1961 – 160

1962 – 924 (a 24-bit 1604)

1963 – 160A (160-A), 1604-A, 3400, 6600

1964 – 160G (160-G), 3100, 3200, 3600, 6400

1965 – 1604-C, 1700, 3300, 3500, 8050, 8090

1966 – 3800, 6200, 6500, Station 6000

1968 – 7600

1969 – 6700

1970 – STAR-100

1971 – Cyber 71, Cyber 72, Cyber 73, Cyber 74, Cyber 76

1972 – 5600, 8600

1973 – Cyber 170, Cyber 172, Cyber 173, Cyber 174, Cyber 175, Cyber 17

1976 – Cyber 18

1977 – Cyber 171, Cyber 176, Omega/480

1979 – Cyber 203, Cyber 720, Cyber 730, Cyber 740, Cyber 750, Cyber 760

1980 – Cyber 205

1982 – Cyber 815, Cyber 825, Cyber 835, Cyber 845, Cyber 855, Cyber 865, Cyber 875

1983 – ETA10

1984 – Cyber 810, Cyber 830, Cyber 840, Cyber 850, Cyber 860, Cyber 990, CyberPlus

1987 – Cyber 910, Cyber 930, Cyber 995

1988 – Cyber 960

1989 – Cyber 920, Cyber 2000

Note: The 8xx & 9xx Cyber models, introduced beginning in 1982, formed the 64-bit Cyber 180 series, and their Peripheral Processors (PPs) were 16-bit. The 180 series had virtual memory capability, using CDC's NOS/VE operating system.The more complete nomenclature for these was 180/xxx, although at times the shorter form (e.g. Cyber 990) was used.

CDC then purchased Bridge Engineering, which offered a line of lower-cost as well as higher-speed card punches. All card-handling products were moved to what became the Valley Forge Division after Bridge moved to a new factory, with the tape transports to follow. Later, the Valley Forge and Rochester divisions were spun off to form a new joint company with National Cash Register (later NCR Corporation), Computer Peripherals Inc (CPI), to share development and production costs across the two companies. ICL later joined the effort.

Printers

The CDC printer business was initially supported by Holley Carburetor in Rochester, Michigan. The two companies later formalized this partnership by creating a jointly held company, Holley Computer Products. Holley later sold its stake back to CDC, the remainder becoming the Rochester Division.

The Rochester Division developed train printers and band printers in a joint venture with National Cash Register (NCL) and International Computers Limited (ICL), with CDC holding a controlling interest. This company was named Computer Peripherals, Inc. (CPI). CDC in 1982 bought a controlling interest in Centronics, a manufacturer of dot matrix printers and merged CPI into it.

OCR systems

Norris was particularly interested in breaking out of the punched card–based workflow, dominated by IBM. He bought Rabinow Engineering, a pioner in the manufacture of optical character recognition (OCR) systems. The idea was to bypass the entire punched card stage by having the operators simply type content onto normal paper pages, using an OCR-friendly typewriter font. The operator could then use a reader to upload the content into the computer. Since a typed page contained much more information than a punched card (which has essentially one line of text from a page), this would offer savings all around.

This seemingly simple task turned out to be much harder than anyone expected, and while CDC became a major player in the early days of OCR systems, OCR has remained a niche product to this day. Rabinow's plant in Rockville, MD was closed in 1976, and CDC left the business.

Service bureaus

Another side effect of Norris's attempts to diversify was the creation of a number of service bureaus that ran jobs on behalf of smaller companies that could not afford to buy computers. This was never very profitable, and in 1965, several managers suggested closing the bureaus to save money. Norris chose to keep the bureaus and instead ordered an across-the-board "belt tightening".

Control Data Institute

CDC created an international technical/computer vocational school from the mid-1960s to the late 1980s. By the late 1970s there were sixty-nine learning centers worldwide, serving 18,000 students.

Cyber product lines

A variety of systems based on the basic 6600/7600 architecture were repackaged in different price/performance categories of the CDC Cyber, which became CDC's main product line in the 1970s. An updated version of the STAR architecture, the Cyber 205, had considerably better performance than the original.

Sales of the STAR were weak, but CDC produced a successor system, the Cyber 200/205, that gave Cray Research some competition. CDC also embarked on a number of special projects for its clients, who produced an even smaller number of black project computers. The CDC Advanced Flexible Processor (AFP), also known as CYBER PLUS, was one such machine.

Another design direction was the "Cyber 80" project, which was aimed at release in 1980. This machine could run old 6600-style programs, and also had a completely new 64-bit architecture. The concept behind Cyber 80 was that current 6000-series users would migrate to these machines with relative ease. The design and debugging of these machines went on past 1980, and the machines were eventually released under other names.

In 1986, CDC introduced the Cyber 910 family. These were rebadged Silicon Graphics International (SGI) workstations; CDC had taken a 20% stake in SGI. The first Cyber 910 workstation, the 910-300, was based on SGI's Iris 3100 workstation, employing the Motorola 68020 processor. Later Cyber workstations would reflect SGI's migration to the MIPS processor architecture. CDC introduced workstation and server models employing the R2000 and R3000 RISC microprocessors and running SGI's IRIX operating system.

CDC was also attempting to diversify its revenue from hardware into services and this included its promotion of the PLATO computer-aided learning system, which ran on Cyber hardware and incorporated many early computer interface innovations including bit-mapped touchscreen terminals.