The Mother of Invention

The Henschel aircraft works in Berlin offered Zuse a job as a stress analyst. The work proved boring; it involved repetitious calcula­tions for which, thought Zuse, there must be a better way – a machine, perhaps. It was not the first time he had entertained such thoughts because his degree course had exposed him to equally tedious work with a slide rule[11].

It was not only the calculations that bothered him but also the "traffic control": noting intermediate solutions, transferring them to other parts of the problem, and so on. His first thoughts (around 1933–34) had been to devise pre–printed forms to control and record the flow of work in a standardized way for some common problems. This was followed by ideas for punched cards and mechanical calculation. In fact, whilst still a university student, Zuse had already arrived at fundamental ideas for information control, the reduction of problems to a sequence of simple opera­tions, and the concept that a machine could be built to carry out that sequ­ence. By 1934 he was using the terms "memory unit", "seleclor" and "control device". When work at the Henschel factory reinforced his thoughts he set about building a machine in his spare time using the living room of his parents’ home in Berlin as his workshop.

Necessity was not the mother of in­vention, says Zuse, it was laziness and boredom: the desire to rid himself 3 of those tedious calculations.

Launching the V1

One of his first decisions proved crucial to success: to use binary arithmetic instead of decimal). One of the friends whose help was enlisted, Walther Buttmann, was asked to research the pub­lished work of Gottfried Liebniz in the Berlin University library. It was Liebniz who had first studied binary arithmetic in the I7th century.

So in l936 Zuse started making the component parts of his first all-mechanical machine: using metal pins and slotted metal plates, the ends of the slots representing ones and zeroes. The memory was to hold 64 binary numbers of 16 bits each and he successfully completed it with help from friends who laboured to make the thousands of parts by hand. However, the more complex arithmetic unit required greater manufacturing precision than they could achieve. Programs were coded by punching series of up to eight holes into discarded 35mm movie film, which was far cheaper than the commercially–available paper tape.

This machine was named the Versuchsmodell-1 (experimental model 1) or V1 for short. It was followed by a V2, both of which were later renamed the Z1 and Z2 to avoid confusion with the V1 - flying bomb and the V2 - rocket.

The Z2 re-used the successful memory of the Z1 but with an arithmetic unit made from second-hand telephone relays. Here another friend, Helmut Schreyer, came into his own. Like others, Schreyer had done his share of cutting out metal plates for the Z1. Now he suggested using electromechanical relays instead of the mechanical pins and slots.

New relays were expensive, and since funding was coming out of their own pockets and those of friends, every penny counted. A fully mechanical computer had proved im­practicable and a full-sized relay machine would need thousands of re­lays; so a test model was built using just 200 second-hand relays.

By this time, Zuse had developed the design of his future computer to the stage where he had achieved the yes–no (binary) logical structure for the machine and recognized that it was independent of the physical methods used to build it.

Date: 2015-12-24; view: 822