What does a mini moog synthesiser, a planimeter and a Kerrison Predictor have in common. Well they’re all forms of analogue computers.
The mini moog synthesiser uses continuously variable voltage controlled oscillators, filters, Amplifiers and envelope generators to create musical or non musical sounds. Analogue synthesiser technology was originally based in part on electronic analogue computer circuits.
The planimeter is a measuring instrument used to calculate the area of a two-dimensional shape just by tracing around it and a Kerrison Predictor was one of the first automated anti aircraft gun fire controllers.
All of these use no digital computing to perform their functions and represent how up until the advent of the 1st digital computers in the late 1940s all types of computations were performed.
Today if you ask what a computer is, most would say that it is a machine that uses binary arithmetic to solve almost any type of problem and performs tasks based on instructions in the form of a software program.
They can store, retrieve, and manipulate data for a wide range of applications from word processing, to Internet browsing, to game playing almost anything we can think of can be modelled by a modern computer, they are as Alan Turing predicted in the 1930s a universal computing machine.
And this is what makes digital computers so pervasive, they can be repurposed for almost any task, whereas analogue computers usually perform just one task and cannot be easily changed or reprogrammed to do something else.
Digital computers are now so embedded in our way of life that it’s almost impossible for some people to think that things like going to the moon with the Apollo missions in the 1960s could have been done with a million times less computing power than an original iPhone 1.
However, the most common device used for calculations in engineering and science from the late 19th century, up until 1973 was the slide rule, a hand-operated mechanical calculator consisting of slidable rulers that could do multiplication, division, exponents, roots, logarithms, and trigonometry and is one of the simplest analogue computers.
It wasn’t until the first pocket programmable calculator the HP-65 made by Hewlett-Packard and produced in 1973 did slide rules go the same way as the quill pen.
But even now there are devices like the E6B flight computer, a circular slide rule is used by pilots for determining fuel consumption, wind correction angles, ground speed, estimated time of arrival, and other critical flight parameters, often referred to as a “whiz wheel,” that remains a vital tool in aviation, especially for flight training and planning.
But mechanical devices using gears that performed calculations predicting the movement of the moon and stars into the future were developed in the time of the ancient Greeks over 2000 years ago.
The Antikythera mechanism is often quoted as the first analogue computer which was discovered in 1901 at a shipwreck off the Greek island of Antikythera. In 2005, using X-ray scans a team from Cardiff University were able to determine that the mechanism had 37 meshing bronze gears that allowed it to follow the movements of the moon and the sun through the Zodiac to predict eclipses and even the irregular orbit of the moon.
Although in recent times it’s thought that the device might not have been very accurate, this was not because of the way it was made but because of the limited understanding at the time of orbitals of the planets, something which wouldn’t be truly understood until Kepler’s Laws of planetary motion in the early 1600s.
This was a remarkable piece of engineering and the miniaturisation of the components was comparable to the 14th-century astronomical clocks almost 1500 years later.
Over the centuries many devices like the Astrolabe have been created to aid navigation by measuring the altitude of celestial objects, telling time, and calculating positions.
The Orrery is a mechanical model of the Solar System that illustrates or predicts the relative positions and motions of the planets and moons. Reports of similar devices come from ancient Rome but it wasn’t until the 1300s that the first clockwork versions were built.
And it is the use of clock like mechanisms using gears and cams in all of these devices that makes them so useful for mathematical operations because you can perform all the basic functions of adding, subtracting, multiplying and dividing.
With just two differently sized gears you can perform multiplication or division of both whole and fractional numbers. When combined, these can do many other mathematical functions which is best illustrated by Charles Babbage’s difference engine which was designed in the 1820s.
This was designed to tabulate logarithms and trigonometric functions to create Tables that would be used in engineering, astronomy and science cheaper and more quickly than by hand. This attracted the interest of the British government which funded the development of a machine for the next two decades.
But by then the increasing cost of the machine, the equivalent of £2.5 million in todays money and the fallout between Babbage and the government meant that it was never truly finished. But using the ideas gained from the differential engine, in 1837, Babbage proposed the Analytical Engine which incorporated an arithmetic logic unit and control flow in the form of conditional branching and loops and integrated memory.
This would make it the first design for a general-purpose computer that could be described as Turing complete, basically the same type of design as modern computers have now.
Babbage was unable to complete the construction of any of the engines due to conflicts with his chief engineer and inadequate funding. It would not be until 1941, that Konrad Zuse would build the world’s first general purpose computer the Z3 based on the same principles as the analytical engine of nearly 100 years before.
By then the world was at war and advances in computing became a national priority for all sides.
Even before the war, analogue computers had been developed by Navy’s to create naval fire control systems that could be used to aim naval guns at moving targets to allow for the speed and distance of a target be that another surface ship, a land based target or an aircraft.
They could also allow for surface wind velocity, roll and pitch of the firing ship, powder magazine temperature and drift of rifled projectiles. By the end of WW2 the British royal navy and the US Navy had perfected radar and when used with the fire control system, they created the first blind fire control system that could work at night and in times of poor or no visibility.
With the advancement of military aviation and long-range bombers, analogue computers were integrated into the fire control systems of anti-aircraft guns.
With the increased miniaturisation of electromechanical components, the Boeing B-29 Superfortress was the first aircraft to use computer controlled gun turrets to both aim and fire the guns and correct for the ballistic behaviour of projectiles, the movement of the target relative to the firing aircraft and the parallax difference from the Gunners sight to the remote control gun several metres away.
The General Electric analogue computer was based on a design by Vannevar Bush of a differential analyser which he developed at MIT between 1928 and 1931. Five were used, one for each of the turrets and meant that the gunners were no longer in the turret as they were in other aircraft.
Computers were also used for the analysis of the extremely maths-intensive and complicated physical processes, one of which was that of the development of the atomic bomb in the Manhattan Project.
One of the things they were used for was calculating the amount and shape of explosives used to detonate around the plutonium core to create a uniform pressure around it and start the nuclear chain reaction leading to the explosion.
The use of computers was essential as there were just too many variables to be calculated by hand and to be done in time to get the bomb ready.
By this time electronic components such as valves were replacing Mechanical gears making the operation of computers much faster and something else was noticed about the similarity of the mathematics of mechanical processes and electronic processes.
Although they were entirely different physically, an electronic circuit could be built using capacitors, resistors and inductors that would very accurately simulate mechanical properties like force, mass, springs and resistance and these were called analogical models.
Using this method, systems like suspension for vehicles and aircraft could be modelled electronically and achieve very high accuracies when the results were transferred to the mechanical domain.
This is where we get the name analogue computers from because the computer is an analogue of the system it is modelling. You could consider the observable properties of billiard balls to be an analogue to the theoretical activities of gas particles or the reflective quality of light.
By making small circuits which had particular electronic mathematical similarities to mechanical properties, it’s possible to build analogues of large and very complex interrelated systems entirely electronically and without using any digital processing.
These type of systems were used on large projects such as the Apollo programme and space shuttle for NASA and the Ariane rocket in Europe during the initial planning phase.
But analogue systems system usually only have an accuracy of 3 or 4 significant numbers, because tolerances in mechanical systems and noise in electronic systems become much great in proportion at very small levels.
A digital system, once it has the data can have much greater precision and is almost immune to noise. Most of the errors in a digital systems would be in the conversion stage from analogue to digital and back again
And here is the major difference between analogue computing and digital computing. In a digital computer, an analogue signal or input has to be converted into a binary representation.
An analogue system can do in one continuous process and orders of magnitude faster than digital processes, what it might take a digital system millions of operations to perform, the saving grace of modern digital systems is that they run incredibly quickly which is a reason why we don’t perceive the delay but they also use a lot more power to do it.
But if analogue systems are effectively hardwired, how could they compete with a digital system that can be reprogrammed on the fly.
Well with modern digital-analogue hybrids many thousands of simple electronic analogue circuits can be created on VSLI chips alongside the digital processing.
Using computer-controlled electronic switches, these analogue circuits can effectively be reconfigured to perform any functions that would be required which are only limited by the speed of the components themselves and do it in one continuous operation.
The analogue signals would fed through whatever configuration of analogue processing required which would be much faster and require much less power. It would then be converted back into the digital form to be passed on to other systems or out onto a network for example.
Using this method you could perform what would be information-intensive processing much more quickly for much less power compared to if it were done totally in the digital domain.
Hybrid analogue-digital computers are not a new thing, they have been around for over 50 years but this new way of matching the two allows possibilities that were just not possible before and while the digital world might seem like it taking over everything, analogue is having a renaissance in computing and not just in electronic music like the Moog and other synthesisers.
So thanks for watching and if you enjoyed this, please thumbs up, share and subscribe.
