The First Semiconductor Multipurpose Control Computer “Dnepr” | History of Computing in Ukraine

The First Semiconductor Multipurpose Control Computer “Dnepr”

By the 1950s, the idea of creating a universal control machine had been around for some time. The first practical implementation to emerge was “Dnepr”, the first semiconductor-based multipurpose control computer, created by the Computing Center of the Academy of Sciences of the Ukrainian SSR. The development took only 3 years, and within 10 years, about 500 “Dnepr” machines were produced. They were actively used by many industrial, military and scientific organizations all over the USSR.
EVM_Dnepr

Bio:

Glushkov
Viktor M. Glushkov

Malinovskiy
Boris N. Malinovsky

When V.M. Glushkov proposed the idea of creation of a universal control machine at the All-Union Conference in Kyiv in 1958, it was met with general hostility. At that time, universal machines were always vacuum tube-based; necessitating large halls with air-conditioning, which wasn’t compatible with the typical conditions in industrial facilities for which the computer was intended.

Nevertheless, despite the strong opposition, active development of the universal control machine began in Kyiv in 1958. The person entrusted with the project was B.N. Malinovsky, who was appointed chief designer; V.M. Glushkov became the scientific director.

The machine was intended to be semiconductor-based with calculations only up to 26 digits, which was at the time sufficient for most typical process control tasks. Average productivity was supposed to be 10,000 operations per second, and the memory block-based in order to enable future upgrades.

One of the unique features was the presence of an integrated computer-process interface, a modifiable set of devices for connecting to sensors and actuators of the controlled object.

The use of semiconductor elements and a new-generation memory storage device ensured the machine was highly reliable and relatively small sized.

Devices common at the time were selected for program I/O: punch tape input device and an electrical-driven digital printing device.

The bulk of the work on “Dnepr” and the innovative systems based on it was carried out by the technical departments of the Computing Center of the Academy of Sciences of the Ukrainian SSR (60 workers), and also by many engineers, designers, technicians from other departments of the Center, including some from the newly-created Research and Design department. Other organizations also participated in the project at various stages.

The development of the machine (from proposing the idea at the conference in 1958 to its launch in 1961) took only three years, a record speed in this kind of development. During this process, the name of the machine was modified from “universal” to “multipurpose” control computer, as per chief designer Malinovsky’s suggestion.

When the developers saw the first model produced at a factory, they were shocked. It looked like a pile of scrap, because its 1,000 welded connections were poorly constructed and constantly malfunctioned; socket contacts would break constantly.

Such a machine was impossible to fine-tune. It was later discovered, when the director of the factory responsible for production had heard that the requested machine was 6 times bigger than an oscillograph, he hired a bunch of recently graduated high school students to do a large amount of the work. When yesterday’s school kids with soldering irons attacked the machine with soldering irons, they inevitably broke a lot of the connections with their careless, inexperienced handling. Still, the deadline for the machine’s launch was approaching, so the scientists themselves had to re-weld most of the machine and replace most of the contacts. This allowed the fine-tuning to finally be done, but cost the developers a lot of effort and countless 24-hour work days.

By the time the State Commission headed by A.A. Dorodnitsyn arrived to test the new machine, it had already undergone two weeks of serious trialing. The scientists evaluated the machine’s average productive work times, its ability to complete standard control tasks, its compliance with technical specifications for heating and cooling and for replacement of standard parts, etc. After further testing, the State Commission recommended the experimental-industrial model of the semiconductor multipurpose control computer “Dnepr” be mass production as the control machine for automation of manufacturing processes in industry, for complex physics experiments, and for military purposes.

The semiconductor multipurpose control computer “Dnepr” is significant as the first Soviet semiconductor machine. It coped well with any weather and climate conditions, was fairly shock-resistant and very reliable overall.

The creation of “Dnepr” in Kyiv also led to the building of the Computing and Controlling Machines Plant, later renamed “Elektronmash”. This idea was initiated by the Institute of Cybernetics of the Academy of Sciences and supported by the government. Thus, the birth of “Dnepr” also laid the foundation for building the first major computer-producing plant in the USSR.

In 1963, the epic story of building “Dnepr” was nominated for the Lenin Award of 1964, under the name of “Design and implementation into the Soviet national economy of the multipurpose control computer “Dnepr”. Among the nominated team members were B.N. Malinovsky, G.A. Mikhailov, N.N. Pavlov, B.B. Timofeev, A.G. Kukharchuk, E.S. Oreshkin, V.S. Kalenchuk, L.A. Korytnaya, V.M. Yegipko, F.N. Zykov, Y.T. Mitulunskiy, A.I. Tolstun, V.I. Skurikhin, Y.L. Sokolovskiy, M.Z. Kotlyarevskiy, and M.S. Galuzinskiy. While they did not ultimately receive the award, their work would have made them worthy winners.

Still, the machine earned high praise from its users: out of the 500 machines produced by the Kyiv “Elektronmash”, all were being used successfully; scores remained in Ukraine, several hundred were used in Russia, and a handful went to the other Soviet republics and even abroad.

It is worth noting that the United States also had their first semiconductor universal control machine RW300 in development, and completed it at approximately the same time as “Dnepr”, in 1961. According to Glushkov, this was one of the significant moments when Soviet science was able to close the technology gap between the two countries completely, in a very significant field.

#photodescription4-9

#photodescription4-9

 

#photodescription4-2

#photodescription4-3

#photodescription4-4

#photodescription4-5

 

#photodescription4-7

#photodescription4-8