Having rapidly grasped the workings of a computer, Matyuhin began developing a detailed design for the arithmetical-logical unit and a magnetic drum unit for external control of memory. His first assistants arrived shortly.

Tamara Minovna Alexandridi was sent to Brook's laboratory for her diploma project in September 1950. She was also "hand-delivered" there by the Moscow Energy Institute's personnel department, knowing that Brook preferred to hire young specialists not thorough questionnaires, but after evaluating their capabilities. The young Alexandridi had no spots in her personnel file other than her unusual surname, which raised suspicion among some people.[3] But the Energy Institute's staff officers decided to avoid any risks by retaining her, even though they knew everything about her history since the Great Patriotic War. I will discuss Alexandridi and her career in more detail later in this chapter.

Brook immediately included Alexandridi in the computer design effort, offering her to work on either electronic or magnetic storage devices. Alexandridi chose the electronic option. Then, Brook proposed that she investigate the possibility of creating memory storage devices on cathode ray tubes that had been used in oscilloscopes. While a graduate student, she was supported by the laboratory engineer Vyacheslav Vasilievich Karibsky. Naturally, Brook did not expect the diploma project of a female student (he was mistrustful of women) to become a part of the scientific report on the M-1 computer.

In autumn 1950, Mikhail Alexandrovich Kartsev, a student of the Energy Institute's radio faculty, came to the lab to work part-time. Brook put him to design the M-1's control unit, which was the most complex part of the computer. At the same time, Kartsev prepared his diploma project by solving problems of Hamming code usage. This code, which improved the reliability of information transmission, was implemented in the development of the M-1's central control unit.

The young specialists were assisted by the technicians Lev Mikhailevich Zhurkin, who designed the external storage on magnetic drums, Yuri Vasilievich Rogachov, who worked on electric mounting and fitting, and Rene Shidlovsky, assigned to electrical mounting and fitting.

In 1951, they were joined by Alexander Zalkind, who had graduated from the Energy Institute in February 1950 and Igor Alexandrovich Kokolevsky, a construction engineer who designed the M-1's frame. Zalkind had participated in the adjustment of the arithmetic logical unit and the design of the input-output unit.

For this small group of young inexperienced specialists, building a computer was a difficult challenge; fortunately for them, they did not immediately realize it. These types of projects were only just beginning to unfold in the Soviet Union and other parts of the world. Moreover, because of Brook's personality, the group worked in complete isolation from other enterprises.

The laboratory's facilities were not appropriate for such large-scale work as building a computer with hundreds of electronic vacuum tubes. The project experienced frequent delays due to a constant shortage of parts and supplies, but Brook's energy and resourcefulness kept them going. He proposed using appropriated German electronic parts for the computer – copper-oxide rectifiers and reliable pentodes (the Soviet analogues were the 6Zh4 electronic vacuum tubes). Cheap and available oscilloscope cathode ray tubes were used as storage devices, and wide rolls of German army teletype paper were used for data input-output. In the end it all came together and the M-1 became the first Soviet small-size computer, using semiconductor elements and memory storage on ordinary oscilloscope cathode ray tubes.

The laboratory's young collective was full of enthusiasm. They worked from morning till late in the evening, inspired by the idea of building the first digital computer that would open the new age of scientific technical progress.

Matyuhin was living with his mother at the outskirts of Moscow, in a tiny 5 square meter room, which could barely accommodate a table and two beds. Matyuhin was so absorbed in his work, that he usually finished around midnight, by which time he neither had the energy nor strength to go home. He would often end up spending the night in the laboratory. This cycle went on for months on end. Kartsev's personal situation was similarly uncomfortable. To make matters worse, while studying at the Energy Institute, he came down with tuberculosis but quickly recovered.

Certainly, this group would not have been as productive if not for their involvement in team sports. Every Sunday was devoted to exercise and the team often went hiking at Istra water storage pond. They also constructed a volleyball court near the laboratory building and enthusiastically played during their infrequent breaks.[4]

The M-1 computer went into operation less than a year and a half after its inception. Out of the nine members of the team, only Brook had an advanced scientific degree. Considering the conditions they worked in, the M-1 was quite a remarkable accomplishment for the young scientists. Brook, Alexandridi, Zalkind, Kartsev, Matyuhin, and the other members of the team kept the final project report, "The Automatic M-1 Computer," approved by Director of the Power Engineering Institute Academician Krzhizhanovsky on December 15, 1951. Kartsev later commented on this period:

In 1950, the Electric Systems Laboratory of the Power Engineering Institute, managed by Corresponding Member of the Academy of Sciences Isaak Brook, assembled the original team of young scientists who would advance Soviet computer technology. Nikolai Matyuhin, currently a Corresponding Member of the Soviet Academy of Sciences, was the first one of us to receive his diploma. Back in the 1950, he was a young specialist, assisted by several graduate students from the Moscow Energy Engineering Institute. I, a fifth-year undergraduate engineer, was admitted on a part-time basis. Yuri Rogachov came to us after his release from the Army. Now he has the State Prize of the Soviet Union, a Candidate of Technical Sciences, and is a Senior Engineer at the Institute. Rene Shidlovsky was a young specialist, a recent graduate from a technical college that was sent to work in out lab. He is now Deputy Senior Constructor, manager of one of the leading departments of the Institute, and Laureate of the State Prize of the Soviet Union. Altogether, we were a group of about ten people. Before coming to the laboratory, not only were we not computer specialists, but we didn't even know about the existence of computers or their possibility. Nevertheless, we began making one of the first Soviet computers—the M-1. We may have been overconfident, but never careless and always professional in out work.

In the beginning of 1950, a strange part was discovered amongst the items delivered from a war reparations warehouse. I cannot say exactly who found it; maybe Brook, maybe Matyuhin or Rameev, who had worked for us earlier. For a long time none of us could guess its origin or purpose, until later we figured out that it was a miniature copper-oxide rectifier. Once the value of this part was fully understood, the M-1 became the world's first computer whose logic circuits were based on semiconductors.

In the summer of 1951, roughly at the same time, both the MESM and the M-1 computers began operating.[5] The first tasks to be solved on the M-1 were set by the academician Sobolev, who at the time was academician Igor Kurchatov's deputy for science exploration. The M-1 had an operating speed of only 15-20 (not thousands or millions) instructions per second with 23-digits numbers, and a storage capacity of 256 words. Many famous scientists and official state visitors came to the laboratory to see our technical marvel.

Such interest in Brook's new device was quite natural. There were no other functioning computers in Moscow. At this time the BESM was still being assembled at the Institute for Precision Mechanics, and SKB-245's Strela was in the same condition.

Yuri Rogachov remembered:

In May of 1950 I was discharged from the army, where I had been a radio operator, and began looking for a job. I had no specialized education and – as was the rule – I was expected to become a student before any job prospects would be open to me. I didn't like that plan. One day, as I was walking along the Lenin Prospect [in Moscow] – at that time it was still called the Great Kaluga Street – when I noticed a small sign on the wall of building No. 18. It read "Electro Systems Laboratory." I decided to inquire within. I was ushered into the office of the Laboratory Director, where a number of people were congregating. During our conversation, a short, heavy-set man walked into the room. He stopped near me, blurted out, "Are you looking for a job?" and immediately started asking about my experience in the army. He said that I would have to work on devices and working principles for a new direction in technology. He spoke as if I was already a member of the laboratory staff. That was my first encounter with Brook.

I began working at the lab in June of 1950. On my very first day on the job, Brook told me about the automatic digital computer project, and the new group that was being formed under Matyuhin's leadership. He pointed to a tall, slender young man in his office, and that was how I met Matyuhin. Nikolai Yakovlevich briefly told me about the lab and showed me around. The Electro Systems Laboratory was divided into two halves: one part was located within the main building belonging to ENIN – No. 19 Lenin Prospect. The other part was on the first floor and in the basement of the right wing of the building No. 18. Most of the time, engineers and energy specialists had the ENIN building at their disposal. A mechanical integrator was located there, which they used to solve various problems. Building No. 18 had a calculating stand with alternating current, intended for modeling complex electrical circuits.

Matyuhin taught me about the digital electronic computer, about how with the assistance of an electronic circuit one could execute arithmetic operations. He explained that the most suitable basis for those calculations was the binary system, which consists of only two digits – 0 and 1. He showed me how these digits could be represented in electronic trigger circuits. He also explained in great detail how to work with the arithmetic unit. In the end, I modeled an electronic trigger scheme based on Matyuhin's design.

Despite the fact that Matyuhin had just graduated from the Energy Institute, he handled the role of chief designer of the M-1 brilliantly. Moreover, together with Brook he invented the concept of a "small" computer, which they managed to build despite the scarcity of the materials available to the laboratory. The entire project was financed by the Soviet Academy of Sciences. To prepare the space for assembly and installation of the computer, they constructed a 1.5 by 1.5 meter base in a room of only 15 square meters. At the center of the base, they installed a square vertical ventilation column, with holes for cooling. There were three frames for mounting electronic circuit panels positioned on the sides of the column: a frame for the arithmetic unit, a frame for the main program unit, and a frame for a storage device. Under the base, a ventilator that supplied the frames with a cool airflow was installed. As the completed panels were received from the assemblers they were installed according to their positions in the frames. The off-line adjustment of the unit as a whole was a step-by-step process that didn't rely on having the complete set of panels. The mounting and separate circuits were checked at the same time.

Using this kind of assembly method considerably shortened the timetable for starting the complex fine-tuning of the computer. After the mounting of the panels was completed in December 1950, the off-line adjustments of the arithmetic unit only took a month and a half, and were completed in January 1951. At the same time the preparation and the off-line adjustment of the main program control unit continued. During the adjustment phase of the equipment on the frames, Matyuhin and Kartsev worked 16–18 hours per day. By spring of 1951, the magnetic drum with a cylinder coated with ferromagnetic material had been produced as well. The adjustments of the magnetic memory – alignment of the magnetic heads and electronic circuits for reading and recording – began after that. This was done by Zhurkin under Matyuhin's supervision. When Zalkind came to the laboratory, he also became involved in the fine-tuning of the arithmetic unit and later designed the input-output unit.

The first half of 1951 was devoted to the adjustment of units in the off-line mode, the fitting of their electrical and functional interfaces and then, the complex adjustment of the computer as a whole. By the summer holidays, the computer had reached the point where it could perform all arithmetic operations in manual (non-automatic) mode. The overall success of the project established a family-like atmosphere among the team members, which was nurtured by Brook's fatherly attitude toward his staff. We worked very hard; motivated in part by Brook's overwhelming ambition and in part because we were young and just beginning our careers. The project never seemed tedious either, mostly because we were delighted to be at the forefront of this emerging technology. In fact, it was actually quite enjoyable. The newness of the work, our enthusiasm and desire to obtain new results as soon as possible – each step forward yielded exciting results – made us lose track of time. We gladly worked overtime without extra compensation, starting in the early morning and finishing late in the evening.

At the end of August we began the final calibrations: running the arithmetic and logic operations in automatic mode. And with the operation of the input-output device, programming began. The first programs were designed for simple tasks. One of them was a calculation table for the function y = x2. This task had one feature: the values of the "y" function for positive and negative values of "x" were identical. Thus, we could check the computer's calculations by comparing the results. It was a lucky break because at the time we had no idea about special test programs for checking the accuracy of a computer. One could say that the parabolic equation y = x2 was the first test program for the M-1. The second one was a program for the equation y = 1/x. With the solution of these equations, the M-1's complex adjustment was complete.

At the beginning of 1952, the M-1 went into trial operation. Various tasks were solved on it, with the aim of checking the technical solutions and refining the programming technology. It became clear, for example, that we needed a control panel and a "Stop" operation, which the design engineers had not foreseen.

During that period, everybody took an active role in operating the computer and identifying the strong and the weak features in its circuitry.

Zalkind remembered an interesting episode concerning the M-1's operation and usage by the military:

Machine time on the M-1 became extremely important for 'The Beard' group, a clandestine governmental department.[6] The Beard's right-hand man responsible for mathematics – because the term "software" did not yet exist – was the well-known Sergei Lvovich Sobolev. He often visited the M-1's room and was very supportive of our work. His group needed to calculate the reverse matrix of large dimensions, which was done on the M-1 in the beginning of 1952.

At this time we began receiving the first domestic 6Zh4 electronic vacuum tubes, but our attempts to change out the German variants for domestic ones were a complete fiasco because of the wide deviation of the cut-off voltage in domestic pentodes. All work on the M-1, even the testing, completely stopped. It was very frustrating for Sobolev, but for our team it was a complete disaster.

I was sent to the Svetlana Vacuum Tube Manufacturing Plant in Leningrad [today St. Petersburg] to obtain several hundred 6Zh4 vacuum tubes that had passed the quality control inspection. To fill this order, we had built a simple stand with a wall outlet plug and a single tube socket, a power supply for the pentode and a current tester. We had also prepared a basic business letter: "As part of your technical assistance to us, we kindly ask you to allow our representative, Mr. Zalkind, to select from your 6Zh4 tubes. We guarantee payment."

Just before leaving for Leningrad, Sobolev visited us. He told me: "Should any problems come up, you must make a telephone call... and at the start of the conversation don't forget the password..." Sobolev then mentioned the name of a well-known flower.

After such an unusual briefing, I remember feeling fearful and insecure as I walked up to the office of Svetlana's Senior Engineer, Mr. Gavrilov. I nervously waited near the door, while Gavrilov, without getting up from his armchair, suddenly barked: "Looking for tubes?" "Yes," I answered. He retorted: "Get out of here!"

Feeling miserable, I returned to the hotel, but then remembered Sobolev's parting words. I made the call. After someone answered, I named the flower. The voice on the other end of the receiver gave me the address of an apartment building on Nevsky Prospekt [the main boulevard in St. Petersburg] opposite a knitwear shop. I went there; it looked like a typical apartment. They let me in, listened attentively and said: "We act only on the level of the Third Secretary of the Regional Communist Party Committee. You need to wait two days and then call us back in the same manner."

Two days later they answered my call: "Everything is in order with Gavrilov. You may visit him again."

At the Svetlana plant, Gavrilov smiled, shook my hand and issued an order to provide me with everything I needed. I brought 300 6Zh4 vacuum tubes to Moscow.

Those were the type of strategies used by the "Gordorstroi" [the common Soviet abbreviation at the time for the MGB, the Security Ministry responsible for supplying the Soviet Union's nuclear weapons project]. With the M-1 working around-the-clock, Sobolev was pleased.

Returning to Rogachov's reminiscences:

Encouraged by the M-1's success, in April of 1952 Brook asked a group of engineers and technicians under Kartsev's management to create a new and improved computer with more advanced features that ever before. Again, the young team pulled off this seemingly impossible task, and by the end of 1952, just six months later, a new, more powerful computer was being installed and prepared for adjustment.

Kartsev talked about how he started out in science and the M-2 computer project at the 15th Anniversary meeting in 1967, before a group that he established himself at the Institute of Computer Complexes of the Ministry of Radio Production of the Soviet Union:

In the spring of 1952, immediately after I had received my diploma, Brook assigned me to a group of seven people to design and build the M-2 computer. Today, it's difficult for me to understand how we managed to do it. We developed the technical documentation first, then watched as parts were manufactured at a number of sites – an experimental factory of the Institute of Combustible Mineral Resources of the Soviet Academy of Sciences, an experimental construction branch of the MEI, the Sokol medical equipment plant, and about ten other places. Then we gathered all of the parts and began assembling the computer. We started in the spring of 1952 and by October 10th the first two frames – the control and arithmetic units – were operating, just in time for the opening of the 19th Congress of the Communist Party of the Soviet Union. By November 7, 1952, we had finished the power supply frame and the magnetic drum. And on December 5th – the Constitution Day – the electronic memory was installed and tested. By January 1953, the computer began operating using a magnetic drum, and in summer 1953 it was fully operational.

Generally speaking, the M-2 machine remained the only one of its kind. The Chinese tried to copy it, but we have no definite confirmation that it actually worked.[7] It was a serious computer, used for large-scale and very important calculations were. Strictly speaking, over the course of several years, there were only two operating computers in the Soviet Union: our M-2, and the BESM of the Institute of Precision Mechanics.

Sobolev conducted large-scale calculations for Kurchatov. By a special government order, we were commissioned to make calculations on the stability of the dams for the Kuibyshev and Volzhskaya hydroelectric power stations. These calculations were managed by the Institute of Precision Mechanics. We also conducted calculations on our computer for M.A. Mikheyev from the Alikhanov Institute of Theoretical and Experimental Physics (at that time called the Academy of Sciences Heat Engineering Laboratory) and many, many others.

All calculations done on the M-2 computer were approved and coordinated exclusively by Brook. However, the first actual shakedown calculation on the M-2 violated this "iron" rule, but Brook didn't find out about it until 15 years later. Here is what happened. At the end of 1953, when the adjustment of the M-2 was completed, Brook went on vacation to Kislovodsk, a famous resort town in the Soviet Union. At the same time, a group of scientists at the Combustion Physics Laboratory in the Power Engineering Institute, directed by Tatiana Bazhenova, were completing tables of thermodynamic and gas-dynamic air parameters necessary to build fire-proof protective shells for rocket hulls. The group had begun calculations in the summer 1953 and promised to finish them by December.

Bazhenova recalled:

Despite the fact that the calculations accounted for only two components of the air – oxygen and nitrogen – the task was extremely time consuming. It required solving dissociation equations for oxygen and nitrogen, then adding ionization equations for their atoms to form nitric oxide, then adding equations of collision processes, and finally, applying the law of conservation of energy and wave interference laws of gas dynamics. As a result, a system of thirteen differential equations had to be solved by a method of sequential approximation.

At first, the problem was given to two laboratory technicians, but no matter how hard they tried to complete the work on time, the calculations proved too cumbersome for them. So it was turned over to the First Moscow Factory of Mechanical Calculations, where an entire staff of young women with hand computers attempted to solve the problem. The work went faster, but the completion deadline loomed even closer. At that time, BESM was the only electronic computer that worked on serious, urgent orders and it was booked ahead for a long time. And then some unexpected help materialized.

We knew that in the neighboring laboratory directed by Brook, there was work was going on with some kind of a new secret machine. On my birthday, friends from that laboratory with whom I used to go on ski trips, came over and presented me with some ski grease, which back then was scarce. The cans of grease were stacked one on top of another and wrapped with paper tape covered with straight lines of numbers. Despite the fact that I knew almost nothing about computers, I noticed that the paper tape resembled computation printouts. "Is this your tape?" I asked the guys. "Yes, it's ours," they replied. After this it was easy to guess what kind of secret machine was being created in the laboratory next door. Through Prziemsky, our laboratory Communist Party organizer, we turned to our friends, Misha Kartsev, Yuri Lavrienyuk and Tamara Alexandridi. They understood our difficulties, plus the computer had not yet come on line and had no orders. Consequently, the "Brookians" decided to test the M-2 using our problem. At the same time, our missile specialists were camped out around the clock near the building No. 18 on Lenin Prospect. As the portions of the tables were completed, they took turns taking them back to our lab to perform additional calculations for the coating of the hulls of our first intercontinental ballistic missiles. In retrospect, the rush was justified: possession of such missiles gave our country weapons parity with the United States.[8]

The M-2 was never mass-produced, despite its excellent construction and time proven superb performance. During the fifteen years it was operational at the Power Engineering Institute in Moscow; it was responsible for solving a wide range of problems in many different branches of science and technology.

During the M-2's construction, Kartsev talent really shined through. Unlike the small M-1 computer, the M-2 was appropriately classified as a large machine. During its first period of operation it had the same speed of 2000 instructions per second as the Strela and the BESM. The M-2 represented Kartsev's first step towards founding his own scientific school, where the main focus would become the creation of specialized supercomputers.

In Brook's laboratory, almost simultaneously with the M-2, design work began on a smaller electronic computing machine – the M-3; Brook appointed Matyuhin to manage this project.

The decision to begin design work on two computers at the same time by such a small design team can be attributed to the fact that Brook's two leading scientists, Matyuhin and Kartsev, had been striving for independent work and begun to display leadership skills early on. Such characteristics were quickly noticed by the shrewd scientific director Brook.

The M-3 could have also remained a one-of-a-kind machine (its development was not officially sanctioned by the government), had it not been for the academician Viktor Amazaspovich Ambartsumian. Upon his arrival in Moscow in 1954, he asked his friend Andronick Gevondovich Iosifian, Director of the All-Union Soviet Scientific Research Institute of Electro-mechanics [today the Vserossiskii nauchno-isledovatel'skii institut elektromekhaniki, or VNIIEM], to help him obtain a computer for the Armenian Academy of Sciences. Iosifian turned to Brook, and the two of them agreed to manufacture three M-3s at the Institute of Electro-mechanics, which had its own manufacturing facility. One computer was to be built for the Institute of Electro Mechanics, one for the Yerevan Institute of Mathematics of the Armenian Academy of Sciences, and another for Korolev's space program. They formed a joint group of specialists: Matyuhin (from Brook's laboratory), Boris Moiseevich Kagan, George Petrovich Lopato (from Iosifian's Institute), and others. In 1956, the first of the M-3's was fine-tuned and presented to the State Commission, together with technical documentation required for industrial mass- production.

Boris Kagan, the informal leader of the joint group, spoke about the M-3 computer at a ceremonial meeting celebrating Isaak Brook's 90th birthday:

...Because the M-3 was actually a self-initiated project, independent of any state plans, the State Commission headed by Bruevich and assisted by Shura-Bura showed its true character by not wanting to acknowledge this computer, claiming that it was "born illegitimately." Although they did eventually accept it, two years went by without it's going into mass production. During this time, the Yerevan Institute of Mathematical Machines was founded and began producing its own computers using our documentation for the M-3.[9] At the same time, the first industrial computer factory in Minsk, Belarus, was built, but had nothing to produce. The factory managers learned that Iosifian had a model of a computer but nobody had given him permission to make it. Only then were the M-3's documents transferred from the Institute of Electro Mechanics to the factory in Minsk and the M-3 computer became the basis for computer manufacturing in Yerevan and Minsk.

I would also like to note that the first computers in Hungary and China were built using our documentation. As for the Institute of Electro-mechanics, this work became the springboard for the development of a foundation for large-scale research and production of control computers and systems.

So in the end, the "Brook Brigade" succeeded in joining the ranks of the designers of industrial mass-produced computers.

^[3] Translator's Note: Alexandridi is a common name among thousands of Greeks living in the Black Sea region of the Soviet Union. However, for inhabitants of Central Russia the name sounded strange, and thus aroused suspicion at the domestic security service in Moscow.

^[4] Translator's Note:Sunday was the only official day off in the Soviet seven-day calendar week from 1936–1965. Before 1935, the official Soviet week was equal to six days that had no names such as Monday, Tuesday etc. Every sixth day was a day off. After Stalin's Constitution of 1936, a seven-day week was declared, with ordinary day names. Until 1965, Sunday was the only day off and Saturday was an ordinary working day.

^[5] Author's Note:Kartsev meant that the M-1 had started to do arithmetic operations in semi-automatic mode. The total outfitting of the M-1 was completed at the end of the year. As the M-1 designers have noted, the real operation of the M-1 started in January, 1952. In Brook's book The High Performance M-2 Computer (1957), he claimed that M-1 began to operate in spring of 1952.

^[6] Translator's note: In Russian boroda, or beard, was the affectionate nickname given by his employees to Igor Kurchatov, scientific head of the Soviet nuclear weapons program. Kurchatov fashioned wearing a long beard.

^[7] Author's Note: Tsai Chuan Yuan wrote for Druzhba magazine (no.11, 1958) "2000 Instructions Per Second," claiming that the Chinese M-2 began operating in October 1958.; Translator's Note: Druzhba [Friendship] was a popular, mass-circulation illustrated magazine, dedicated to Soviet-Chinese relations.

^[8] Author's note: In 1968, 15 years later, Bazhenova wrote about this incident in the article "Space in Tubes" in Science and Life magazine.

^[9] Author's Note: These Armenian machines included the Aragats, and Razdan-1 and Razdan-2 computers.