Second generation: transistors
taken from Wikipedia

Initially, it was believed that very few computers would ever be produced or used. This was due in part to their size, cost, and the lack of foresight into the types of uses to which computers could be applied.
In 1951 starts the first calculation machine which is made in series and there is a big development of these machines, because of the introduction of new techniques, of new unities and programming methods. In 1953 the calculation machines' number all over the world raises till about 100 unities.
In 1958 only the United States have about 2500 models in all.
In Italy the first calculation machine was settled in 1954 at the Politecnico di Milano University and only in 1957 it is used by a firm. In 1958 it is settled in Italy a tenth of calculation machines, which support about 700 meccanographic employee. In conclusion of the First Generation, at the end of the fifties, the electronic calculation machines have won their users' trust. At the beginning they were considered more as calculus instruments and useful for the university researches than machines useful to the corporations or firms' operative needs for their capacities to execute information.
The calculation machines overcome more and more the restrictions due to some building and not refined programming techniques .
Their use doesn't represent any more an "adventure" for the firms and the corporations which settle them, but it answers to the necessity to solve the various operative problems.
About at the end of the fifties the empty tubes were replaced by the Transistors. So it rises what is known as the "second generation" of the calculation machines.
Using the transistors and improving the machines and the programs, the calculation machine gets quicker and more economical and it diffuses in a tenth of thousand models all over the world. By the changed general economical situation, the continuous firm's increase, the introduction of new organisation 's techniques and the management of a firm, it passes from a prevalently book-keeping and statistical use to some more complex applications which concern all the assets' sectors.
The transistor was realised in 1948 by the Americans John Bardeen, Walter Brattain and William Shockley who shared the 1956 Nobel Prize in Physics for their invention. The transistor is an electronic device silicon crystal or germanium made in which different materials atoms are opportunely inserted.
For some values of the electric tension to which the transistor is exposed, it has the capacity to transmit or not the current, so to represent the 1 or the 0 which are recognised by the machine. Compared to the valves, the transistor have a lot of advantages: they have a smaller making price and a ten bigger times speed, passing from the position 1 to 0 in a few millionth of second. The sizes of a transistor are of some millimetre compared to the quite a lot centimetre of empty tube. The safety-operating directions is increased because the transistors, working "at cold", avoid the breaks because of the heating which were frequent in the valves. Machines are so built with ten thousand complex circuits which are included in a small space.
Among the systems of the second generation we mark the IBM 1401, which was installed from 1960 to 1964 in more than one hundred thousand models, monopolizing about a third of the world market. In this period there was also the only Italian attempt: the ELEA by Olivetti, produced in 110 models.
The remarkable development of the calculation machines and their applications in this period is not due only to the feature of the CPU (Central Processing Unit), but also to the continuos improvements done to the auxiliar memories and to the unities for the intake and emission of data.
The disks memories are able to set tens million of letters or digits. More unities can be connected at the same time to the calculation machine, taking so the total memory capacity to some hundreds million characters.
Near the disks which are firmly connected with the central unity it is introduced some unities in which the piles of disks are mobile and they can be easily replaced by another pile in few seconds.
Even if the mobile disks' capacity is smaller compared to the fixed ones, their interchangeability guarantees an unlimited capacity of data which are ready to the elaboration.
The second generation calculation machines, through a particular device made to sort the inside data, are able to superpose different operations, that means to read and punch the cards at the same time, to execute calculus and to take logical decisions, to write and read information on magnetic ribbons.
To guarantee the continuos change of information between the centre and the periphery, it rises the terminal unities which have to transmit the data to the central calculation machine which can be also at a distance of hundreds kilometres thanks to a telephonic connection.

Post-1960: third generation and beyond
Main article: History of computing hardware (1960s–present)
The explosion in the use of computers began with 'Third Generation' computers. These relied on Jack St. Clair Kilby's and Robert Noyce's independent invention of the integrated circuit (or microchip), which later led to the invention of the microprocessor, by Ted Hoff and Federico Faggin at Intel.
During the 1960s there was considerable overlap between second and third generation technologies. As late as 1975, Sperry Univac continued the manufacture of second-generation machines such as the UNIVAC 494.
The microprocessor led to the development of the microcomputer, small, low-cost computers that could be owned by individuals and small businesses. Microcomputers, the first of which appeared in the 1970s, became ubiquitous in the 1980s and beyond. Steve Wozniak, co-founder of Apple Computer, is credited with developing the first mass-market home computers. However, his first computer, the Apple I, came out some time after the KIM-1 and Altair 8800, and the first Apple computer with graphic and sound capabilities came out well after the Commodore PET. Computing has evolved with microcomputer architectures, with features added from their larger brethren, now dominant in most market segments.
An indication of the rapidity of development of this field can be inferred by the Burks, Goldstein, von Neuman, seminal article, documented in the Datamation September-October 1962 issue, which was written, as a preliminary version 15 years earlier. (See the references below.) By the time that anyone had time to write anything down, it was obsolete.