What is the difference between analogue and digital computing?

Digital computers are based on binary logic, which require everything they process to be turned into a binary value – 0 / 1; true / false; on / off. They work by having a central processor handling endless strings of these binary values, one by one, at each step applying some logic of the form IF-THEN-ELSE to decide what to process next. Digital computers are ideal for handling data that comes in a predictable, limited form, such as written text or account details. They are also perfect for defining sequences of events that must be performed in a fixed order.

The form of analogue computer proposed by QQRONA is based on qcells, that handle infinitely variable values of data, all of which can be changing constantly in response to how other values are changing around them. These changes happen independently, without the need for any central processor.

Analogue computers are more suited for handling numerical values and dealing with events that occur in an unpredictable way. In other words, most real-world situations including the scientific, medical and engineering domains.

Is analogue better than digital?

For the types of application mentioned above, yes, analogue solutions will outperform digital in terms of space, speed, scalability and resource consumption, probably by several orders of magnitude. Analogue is inherently more efficient as, at its core, its basic unit of data is a variable value that a digital system would need many binary digits to process. Analogue is inherently faster as its basic architecture is parallel, not sequential, so it performs operations concurrently not consecutively. And because of this parallel architecture, there is no central processor creating a bottle neck, so that more analogue qcells can be added indefinitely to a model without scaling issues. The analogue proposal is also based primarily on passive components, while the fundamental unit in a digital system is the transistor, which is continually forcing a signal as fast as possible into one of its two allowed values. This active component needs a relatively high voltage to function (thousands of mV) and consequently uses a lot of electrical power and generates a lot of heat, which a large part of a digital system’s footprint is dedicated to dispersing.

However, the reality is that we need hybrid systems with digital providing management and control over analogue core systems, in the way that QST allows us to set up and manage QQRONASynth models.

How are analogue computers programmed?

They aren’t, at least not in the same way as digital. Analogue computers are not built to execute fixed instructions in a given order, but rather to respond to multiple, irregular events arriving at random times. They typically involve multiple nested feedback loops, as each qcell has to respond to activity in its connected qcells as well as any external connection it may have. The QQRONA analysis method is used to break down the desired behaviour into its basic component behaviours, which are then assigned to qcells. There follows a process of assigning initial values to the qcell configuration parameters, which will cause the model to approximately perform as needed, and then using the tuning tools included in QST to refine the model to the required accuracy and precision.

Do QQRONA analogue computers exist?

Only in the form of simulations in a digital environment, ie QQRONASynth, which clearly negates most of the advantages described above. However, we believe that a physical system can readily be designed by a suitably qualified engineering team. QQRONA provides the blue print and the operating system for such a development.