The first blocks in the scientific foundation of digital electronic technology were laid in Moscow, but things changed by the end of the 1940s. Because of Lebedev, the center of the new science shifted to Kiev. When academician Nikolai G. Bruevich was appointed as acting director of the Institute for Precision Mechanics and Computer Technology in Moscow on July 16, 1948, he did not know that secret work on the MESM was going on full speed in Kiev. In 1949, the first news about modern digital computers arrived at the Institute for Precision Mechanics from abroad. Foreign journals reported that in 1946, the first computer in the world, ENIAC [the Electronic Numeric Integrator and Calculator], had been created in America. It employed 18,000 vacuum tubes and performed about one thousand operations per second. Later, news of faster computers with fewer vacuum tubes appeared in the Soviet Union. Because the reports were very sketchy, the principles of building a machine remained unclear.

A year after the Institute for Precision Mechanics was established, a commission from the Soviet Academy of Sciences Presidium headed by Keldysh assessed the institute's work. Quite possibly, Lavrentiev's letter to Stalin prompted this investigation. Keldysh's commission came to a disturbing conclusion: digital electronic computer technology was developing rapidly in the West, whereas in the Soviet Union, it received very little attention. Spurred on by the commission's conclusion, Bruevich approved a decision to organize a department of high-speed computers in the institute through the bureau of Soviet Academy of Sciences Department of Technical Sciences. In September 1949, he charged a group of six scientists from his department to develop the necessary components for the creation of a digital computer.

A participant in the project, Petr Petrovich Golovistikov recalled:

Many visitors came to watch when we began to assemble the basic parts of the computer such as flip-flop triggers, a computer serial adder, rectifiers, and a decoder. At the time, I did not understand why Bruevich had invited them. It seemed to me that we had very little to show. Among the visitors were people such as the Minister of Machine Building and Instrument Engineering of the USSR Parshin, ministry board member Loskutov, and academician Blagonravov. This excited and compelled me to work every day from the wee hours in the morning until late at night. At last, I got used to these visits. I especially remembered one visit that took place in January of 1950. Bruevich brought in two men. One was tall, stately, and well-mannered; he listened attentively to the explanations. The other was short, wore glasses, and left a strong impression on me. He spoke directly to me and asked many questions. He wanted me to show him the computer's signals at various points and to demonstrate signal delay times in different circuits. He asked to change the generator frequency in order to define the range of scheme operation. He criticized many things and advised us to do some things in differently. To top it off, he asked me to make a prototype of a long circuit of controlled rectifiers. He wanted each tube to carry an additional load corresponding to similar tubes, so that the signal in the circuit would not dampen and so that the circuit would have minimal delay. Thus, I made the acquaintance of Lavrentiev and Lebedev. By this time, I knew that developments in the field of computers had begun at the ASUSSR Power Institute under I.S. Brook's supervision and at the Ministry of Machine Building and Instrument Engineering where SKB-245 [in Russian, Spetsial'noe konstruktorskoe byuro] was created. But completely unexpected for me was the fact that the first computer in the Soviet Union was developing rapidly under Sergei Alexeevich's supervision in Kiev.^[8]

Bruevich worked with the Soviet Ministry of Machine Building to establish by 1949 three organizations that composed a powerful scientific production collective: the Scientific Research Institute for Calculating Machine Construction [in Russian, Nauchno-Issledovat'elskii Institut Schetnovo Mashinostroyeniya – NII Shetomash], SKB-245, and the Calculating Machine Factory [in Russian, Zavod Schetno-Analiticheskhik Mashin, or SAM]. The director of the both the Schetmash and the SKB-245 was Mikhail Avksentievich Lesechko.

Despite the fact that these three organizations were ordered by the government to build differential analyzers similar to the ones the Americans had built in the 1930s and 1940s, Lesechko, relying on his engineering intuition, agreed with Bruevich's proposal to design and build vacuum tube based electronic computers.^[9] Yet a visit to the Institute for Precision Mechanics changed his mind. Lev Israelevich Gutenmakher, director of one of the Institute's laboratories, was attempting to build an electronic computer based not on vacuum tubes, but on electromagnetic non-contact relays [magnetic amplifiers]. His work appealed to of the Minister of Machine Building and Instrument Engineering, Petr Ivanovich Parshin, who took a keen interest in it.

When a change in leadership occurred at the Institute for Precision Mechanics in mid-March of 1950, the newly appointed director, Mikhail A. Lavrentiev, found himself under pressure on many fronts. First, there were very few specialists in the field of digital computer technology at the Institute. Second, the few relevant scientific departments that did exist were scattered throughout Moscow. Third, it appeared that the government was about to adopt a resolution mandating that the Institute develop a new digital electronic computer – a gigantic construction project consisting of thousands of vacuum tubes and significantly more complex than what Lavrentiev had seen in Kiev at Lebedev's laboratory. Thus, in an order dated March 20, 1950, Lavrentiev appointed Lebedev as Director of Laboratory No.1 of the Institute: Lebedev was now employed in both Kiev and Moscow. Also by this time, the Ministry of Machine Building and Instrument Engineering was commissioned to design and develop the other digital electronic computer based on Gutenmakher's ideas.

Before approving the governmental decree to develop and build two computers, Stalin required that several individuals be assigned to oversee each machine. Lavrentiev and chief designer of the BESM, Lebedev, were appointed from the Soviet Academy of Sciences. From the Ministry of Machine Building and Instrument Engineering, Lesechko and Yuri Yakovlevich Bazilevsky, chief designer of the Strela (Arrow) computer were appointed. Thus, Lebedev's powerful rival collective of the SKB-245, the SAM, and the SRI Schetmash was born. Clearly, the center of activity of the Soviet Union's digital electronic computing had shifted from Kiev to Moscow. For anyone else but Lebedev, the situation at the Academic Institute for Precision Mechanics would have seemed hopelessly complicated. Instead, he brought from Kiev his already sketched out plans for the BESM.

According to Golovistikov:

There is a legend that the BESM's full schemata was drafted on separate sheets of paper and kept by Lebedev on Kazbek cigarette boxes. This is simply not true. He had many thick notebooks that contained scrupulously drawn diagrams and designs, with all of the structural schemata of the machine, provisional diagrams of the units' operations, and detailed descriptions of all methods of individual operations. Having arrived from Kiev, he immediately began transferring this enormous volume of information to us...

Initially, I wasn't confident about the work I was performing. Lebedev assigned me to design the arithmetic unit, but also wanted me to understand the principles of other operational units. Kiril S. Neslukhovsky was commissioned to develop the control unit and needed to know how the machine as a whole would operate. Neslukhovsky became Sergei Alexeevich's technical operations manager. Later, this post was filled by Bardizh.

To find appropriate personnel for the Institute, Lavrentiev and Lebedev went to colleges to hand-pick students with practical experience. They were placed on staff and immediately received specific design assignments, while the development of all main computer units for the preliminary prototype was overseen by specialists. Students also took part in preparing project outlines relating to their own area of specialization. With only small modifications, according to the requirements of their colleges, these materials satisfied their graduation requirements.

By the end of 1950 the prep work on the prototypes of BESM's individual components was in full swing. In spring 1951, the staff of Laboratory No. 1 had grown to about fifty people – many were the most highly skilled specialists lured away from the Moscow Power Institute. Candidate of Technical Sciences O.K. Guschin (an assembly worker at that time) recalled the formation of the young team at the Institute under Lebedev's supervision:

It seems to me that everyone was proud of their participation in this great and important affair – the creation of Soviet computer technology, a gigantic device in its infancy, the so-called "electronic miracle" with hundreds of thousands of components. It should not be forgotten that the most complex radio-electronic device of that day was the KVN-49, the first Soviet television set.

Our work went on day and night, and nobody expected to have any time off. We built the prototype elements and components of the BESM. We made chasses and stands; we drilled and riveted, assembled and adjusted different versions of flip-flop triggers, counters and computing units, and then checked their reliability in operation. At every stage of the work, Sergei Alexeevich demonstrated an exemplary selflessness. After an intense workday, he often sat up until 3:00 or 4:00 a.m., by the control desk or oscillograph, actively taking part in the fine adjustments of the machine.

Lebedev's competitor, Gutenmakher, encouraged by the support of the Ministry of Machine Building, was also working very hard. In 1950, he came up with an additional schematic for an electronic computing device for the SKB-245 based on ferrite-diode elements. At this time the situation in the Ministry changed drastically: Bashir Rameev appeared at SKB-245 project, having been collaborating with Isaak Brook until 1948 on programming controls for electronic computers.

Rameev immediately got to work and prepared an advanced electronic computer project based on vacuum tubes. Subsequent events unfolded very strangely. In Rameev's absence, the technical council of SKB-245 considered Gutenmakher's project. Then, in Gutenmakher's absence, they listened to Rameev. In the end, the decision was made to support the machine based on vacuum tubes, instead of Gutenmakher's design. Thus the BESM had a serious competitor—the Strela. The SKB-245's managers appointed Rameev as the deputy director for Yuli Bazilevsky, the chief designer of the project. Rameev was only 32 years old at the time, but he already carried heavy personal burdens. Being the son of the "enemy of the people," he was expelled from the institute during his second year and ended up serving in the army; he was also a highly driven person and had an insatiable desire to work.

Thus Lebedev now faced a competitive-triumvirate: Lesechko, Bazilevsky and Rameev; while the Institute for Precision Mechanics faced a powerful competitor – SKB-245 together with the SAM and the Schetmash. The center of gravity of digital computing technology work had been relocated from Kiev to Moscow.

Having lost the support of SKB-245, Gutenmakher continued to work on his project by himself. In his laboratory at the Institute for Precision Mechanics, a machine based on ferrite-diode elements had already been designed and was being built at the same time as the BESM. I became acquainted with it sometime in 1954, when it was already operating. Its productivity was very poor. Due to the low quality of the elements it was also unreliable; the impulse power source was cumbersome and uneconomical. Under the pretext of secrecy, entrance to the laboratory was prohibited, and early in the 1960s it was completely closed. The strict secrecy in which Gutenmakher conducted his research ensured that very little was known about his machine. Nevertheless, it was a defining milestone in the history of computer technology.

On April 21, 1951, a State Commission assembled to review the two projects, BESM and Strela. The commission included such people as Keldysh, the Chairman, Minister of Machine Building and Instrument Engineering Parshin, academician Anatoly Arkadyevich Blagonravov, and others. Earlier, members of the commission visited Kiev where Lebedev demonstrated the MESM. A detailed analysis of the BESM and Strela projects had already been made in Moscow by the time the commission began evaluating these projects. Commission member Anatoly Alexeevich Dorodnitsyn remembered a curious argument that occurred during one of the sessions:

Strela's chief designer, Bazilevsky, declared that his machine, performing at 2,000 operations per second, could solve all the mathematical problems in the Soviet Union within four months. Therefore, the BESM with its higher performance of 8,000 -10,000 operations per second was unnecessary! Sergei Alexeevich parried the argument saying that due to its low capacity, Strela would not be able to work out a problem during the time between two errors and would display incorrect results, whereas BESM would succeed.

Both Bazilevsky and Lebedev successfully defended their proposals. The management of the Institute for Precision Mechanics decided to support the creation of an experimental model of the BESM. Sergei Alexeevich, in light of his experience creating and operating the MESM, recommend a modular design for the BESM – a bold move because many machines at that time were made with non-modular units. Luckily, the number of different modular blocks needed turned out to be small.

Lebedev and his team began the design and fabrication of racks, circuit boards, and machine units, but could not finish the project as planned. If this project had been completed successfully – all that it needed was a supply of industrial cathode ray tubes for the memory – then BESM would have far superseded its competition not only in the Soviet Union, but all over the world. Its projected performance of 10,000 operations per second would have been five times faster than Strela's. At that time, no machine had ever achieved such speed.

The Soviet Ministry of Machine Building and Instrument Engineering held a monopoly on storage tubes, and ignored the interests of not only Lebedev's institute but of science in the Soviet Union in general: the Ministry supplied only the Strela workers with storage tubes. Even Lavrentiev could not help Lebedev, and the BESM team found itself in a difficult and humiliating position. One could imagine Lebedev's shock in coming so close to meeting an advanced technical goal and then receiving such a blow. He had always been cooperative and offered his help, even to SKB-245. A few years earlier, when Lebedev and Bazilevsky visited Kiev, Sergei Alexeevich had explained MESM's operation in great detail to the scientists on the project, and also helped them reach an agreement with the Institute of Physics for the development of a magnetic-tape storage system. Moreover, the BESM was not a secret project as much as it was a rival. Unfortunately, the competitors did not respond in kind. Former SKB-245 project co-worker, F.N. Zikov, remembered that when Lebedev visited SKB-245 to learn about the Strela, the staff showed him the machine "...packed in boxes, ready to be shipped out."

Without access to cathode ray tubes for BESM's memory, Lebedev was forced to use acoustic mercury-delay line storage tubes, reducing BESM's performance to Strela's level and causing many problems. The amount of mercury needed for rapid access storage at full volume needed to be several hundred kilograms. The memory included 70 mercury tubes, each nearly one meter long: 64 were for storage, one tube was made for clock rate, and 5 served as spares. The mercury tubes had been developed in 1949 on Lebedev's order at the Institute of Automation. All of the tubes were placed inside a giant thermostat, which in turn was installed in a special room equipped with an exhaust hood, where work with mercury could be carried out.

The electronic parts of each channel were assembled in one large standard unit. The control panels and power supply units were substantial, while the external dimensions of the memory panels occupied an entire room. It was located at the end of a corridor on the first floor, a sizeable distance from the arithmetic unit, which was connected by cables carefully soldered onto foil. A large console for each of the mercury memory units included a dot-matrix indicator, a very elaborate console that simplified the life of the shift engineer by allowing him to examine each of the 64 channels. The layout of the memory unit was further complicated by the analog and electronic schema that worked in a closed ring circuit. Yet Lebedev worked tirelessly to adjust the rapid access memory. For two months, he practically lived in the room where the rapid access memory was located. According to Elena Petrovna Landera, "When Sergei Alexeevich made construction decisions, he rarely stopped halfway, and often went to great lengths to make any additional mechanical and assembly adjustments himself."

By the summer of 1952, the machine was basically completed and the adjustment phase had begun. All designers of the machine took part in the work, which continued around the clock. The main defects were in the acoustic-mercury tubes. Many of them burned out during the first hours of operation. But if a tube had already worked several hundred hours, then it was much less likely to fail.

By April 1953 the BESM was commissioned by the State Committee to go into operation and the design engineers stepped aside for the mathematicians. Even though in the beginning the machine performed at below capacity, it was able to solve several national economic problems. Strela had been completed at the same time and had been approved for serial production. Its creators had been awarded I, II, and III degree State prizes. Chief designer Bazilevsky was awarded the title of Hero of Socialist Labor.^[10]

On the recommendation of academician Lavrentiev, who had become a vice-president of the Soviet Academy of Sciences, Sergei Alexeevich was appointed as the director of the Institute for Precision Mechanics and he was also elected to be a member of the Academy. At the election banquet, the son of Otto Yulievich Schmidt, Sigurd Ottovich, proposed a toast: "Today we have elected as academicians two distinguished scientists—S.A. Lebedev and A.D. Sakharov!"^[11] In 1956, when the BESM had been commissioned by the State Committee again (this time it was made with vacuum storage tubes), S.A. Lebedev was awarded the title of Hero of Socialist Labor. The other main designers also received State prizes.

^[8] Author's Note: SKB was the general Russian abbreviation for Special Construction Bureau, whose task was the design and preparation of any product for mass production.

^[9] Editor's Note: In the United States, Vannevar Bush led the effort to build the first differential analyzer at MIT beginning in the 1930s.

^[10] Author's Note: This was the highest state award presented for the most valuable work completed in a specific field.

^[11] Author's Note: Otto Schmidt was a renowned polar explorer. Andrei Dmitrievich Sakharov was the chief designer of the Soviet H-Bomb and later in his life a human rights advocate.