After the work on the vacuum tube-based computers BESM-2 and M-20 was finished, the design of the second-generation supercomputer BESM-6 – the semiconductor based masterpiece – commenced at the Institute for Precision Mechanics. Two of his former students – Vladimir Andreevich Melnikov and Lev Korolev – assisted Lebedev with this project. Melnikov and Korolev both became Lebedev's operations managers and famous young scientists in their own right. They studied and analyzed everything they could get their hands on that was published about designing high-speed computers. Lebedev led the mathematical modeling of the machine. As a result, they developed a machine with an original system of commands that made programming user-friendly, had a simple internal structure, reliable system of elements, and a design that simplified maintenance.

The BESM-6 became the first Soviet computer that was approved by the State Committee with a complete software package, and many leading Soviet technical specialists were involved in its development. However, Lebedev was the first one to realize the effect of the joint efforts by the mathematicians and engineers on the creation of computer systems. The development of computer technology evolved from a purely engineering to a mathematical problem, which could only be solved by the pooling of resources.

Finally, and this is also important, Lebedev formulated all diagrams of BESM-6 with Boolean algebra, which opened vast horizons for the automation of design, and preparation of assembly and operational documentation. Later, the design was further streamlined by Gennady Grigorievich Ryabov, creator of the Pulse System for which he was awarded the Soviet State Prize.^[14]

BESM-6 featured: 1) a pipeline control system, or as Lebedev called it in 1964, "plumbing," according to which the flow of commands and operands were simultaneously processed (up to 8 machine commands at each stage); 2) the use of associative memory on super-speed registers that reduced the number of retrieval calls to the ferrite memory, thus optimizing calculations; 3) a stratification of operating memory into autonomous modules, which enabled simultaneous, multi-directional calls to memory units; 4) a multi-program operational mode for real-time work on several problems with specified priorities; 5) a hardware mechanism for transformation of mathematical addresses to physical ones, which made it possible for dynamic distribution of the operating memory in the computational process; 6) a page system of memory that in turn developed protection mechanisms for numbers and commands; and 7) an up-to-date interrupt system that facilitated the automatic transfer from one computational task to another, accessing external units and controlling their operation.

BESM-6 contained 60,000 transistors and 180,000 semiconductor-diodes. Its element base was brand new at the time and became the foundation of circuit engineering for third and fourth-generation computers. The principle of dividing complex machine logic built on diode blocks using single transistor amplification guaranteed a simplified manufacturing process and reliable operation. BESM-6 achieved average speeds of up to one million operations per second.

The BESM-6 prototype was tested in 1965; in 1967, the first manufactured model had its trial run. Three additional models were made at the same time; because of joint collaboration with the manufacturer, there was virtually no lag time to prepare the computer for mass production.

The State Committee headed by Keldysh, who was still the President of the Academy of Sciences in the 1960s, understood BESM-6's importance. With this technology, the Soviet Union established computing centers that facilitated real-time control systems and coordination of data tele-processing systems. BESM-6's were used for simulating complex physical processes and control processes, and also for development of new computer software in computer aided design systems. The basic technical design that Lebedev and his colleagues had employed during BESM-6's development gave the computer an enviable service life: BESM-6 was manufactured for over seventeen years. Their users loved these machines, and by the 1970s BESM-6 set the standard for high-speed computers in the Soviet Union.

The 1975 Apollo-Soyuz space mission was controlled from a new computer complex that included a BESM-6 and other domestic high-speed computers developed by Lebedev's students. Prior to this, the space mission telemetry data processing would have taken approximately thirty minutes. Using the new computer complex, the work was performed in one minute. Soviet scientists completed all of the Apollo-Soyuz mission's data processing one half hour earlier than their American colleagues. This marked Lebedev's real triumph: his school and his students developed a first-class computer that was capable of competing with the best machines in the world. For their work on BESM-6, Lebedev and his team won the State Prize.

While writing this manuscript, I came across the work of the German philosopher Frederick Nietzsche. One of his statements caught my attention: "The ability to show the way is a sign of genius." Immediately, I thought of Lebedev.

^[14] Editor's note: Ryabov became Director of the Institute for Precision Mechanics in 1984.