To this date, only a small number of linear circuits have been integrated. None of these circuits have had an impact equivalent to that of integrated logic circuits in computers. The reason lies in the difference in the types of components used for each class of circuit. In logic circuits there is hardly any need for close tolerances, so that they can be designed swith only diodes, transistors, and small value resistors. Thus, such circuits are readily adaptable to integrated circuit techniques. Linear circuits, on the other hand, not only require narrower tolerances, but also contain many types of components which are not available in integrated form; transformers, large capacitors, or high value resistors. This lack of suitable components has forced the designer to substitute transistor circuits for large value resistors, capacitors, and transformers wherever possible. In addition, the costs of the various components are entirely different from that in discrete circuits. In integrated circuits a component's cost is related directly to the area it occupies on the chip. The smaller the components of a circuit, the more circuits can be placed on one slice and therefore the lower the cost per circuit.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.