VAR logo English
Home page Services Past achievements Contact Site
map
Page d'accueil Services Réalisations précédentes Contact
Français


Discrete Transistor Computer

Science has a tendency to seem mystical to all but the few, and it appears to be getting worse. It was within living memory that anyone could approach new subjects through anecdotes and party tricks, thus at least giving an instinctive feeling for the subject. But, gone are the days of having a feel for how the music is stored on a revolving disk, for example (there is no equivalent for CDs of sticking a pin out of the corner of a matchbox, in order to reproduce the sound from a vinyl disk). Gone are the days of even seeing the spark plugs in an engine, let alone changing them. As for computer engineering, the workings, now, are not only hermetically sealed in integrated circuit packages, but would be too microscopic to see anyway, and buried deep under layers of silicon dioxide.

Alan Wilkinson's Computer Models (1968, Edward Arnold, SBN 7131 1515 X) continues to be of relevance as a first introduction to computer engineering, even today. Wilkinson describes the computer circuits, that he built with his secondary school class, in a way that brings home how straightforward the concepts really are; and this is applicable in any digital technology: such as pneumatic, hydraulic, optical or even DNA logic, not just the PNP transistor technology that he had as his vehicle).

Processing

The basic element, in Wilkinson's book, was the NOR gate, built with a single OC200 transistor and a few resistors. The circuits were soldered on metal pins nailed into peg-board, and mounted vertically so that the whole class could view the circuit from either side.

Nor Gate
RTL Nor Gate

The gate had a fan-out of three, and also had a fan-in of three, with one 10kΩ resistor per input (so-called resistor-transistor-logic, RTL). These input resistors were connected together at the base of the OC200 transistor, with the emitter connected to ground, and the collector connected to the output, and via 5.6kΩ resistor to the -12V supply (the OC200 is a PNP transistor, but the example might now be be converted to using a NPN transistor, as illustrated, or a FET).

Wilkinson described each of the usual computer engineering blocks built from NOR gates: combinatorial logic; astable, monostable and bistable multivibrators (using 0.01μF capacitors and OA81 diodes); binary counters; decimal counters; shift registers; serial adders/subtracters; parallel adders; repeated-addition multipliers.

Memory

One weakness is that there is no real reference to stored programs. Hence, the title of the book, "Computer Models", since the resulting circuits are limited to those of the arithmetic and logic unit (ALU).

However, it would be relatively simple to extend on this, such as in the processor example that I am fond of using as a simple introduction. Then, for the memory itself, perhaps a single static RAM integrated circuit could be included now (since memory is such a repetitive, albeit large-scale, part of the computer)

Communication

Input was made via switches and old, rotary, telephone dials that were RC-debounced (4.7kΩ and 1μF).

Output was made via 6V 0.05A torch bulbs. These were driven by an NKT224 (or OC70, or OC72) emitter-follower (with a 150Ω resistor from the emitter down to ground), which was in turn driven by the NOR gate that was giving the output. (Interestingly, the NKT224 was a germanium transistor).

Conclusion

Wilkinson's simple introduction to computer circuitry is still pertinent to the design of today's microprocessors. However, it goes deeper than that. Once the book has described how to build a logic gate out of a single transistor and a few resistors, the rest of the design can be conducted at the abstract, logic-gate level. At this level, it does not matter how the logic gates are implemented. It is easy for the reader to imagine the logic gates instead being implemented with Pneumatics, Hydraulics, Babbage-like mechanics, or laser-based photonics, or even with biological DNA. Even the concepts of quantum computer hardware design become approachable. This is the justification for my assertion that this book continues to be of relevance as a first introduction to computer engineering, even today.

Top of this page Home page Services Past achievements Contact Site
map
Page d'accueil Services Réalisations précédentes Contact
© Malcolm Shute, Valley d'Aigues Research, 2006-2015