Adifference engine is an automaticmechanical calculator designed to tabulatepolynomial functions. It was designed in the 1820s, and was created byCharles Babbage. The namedifference engine is derived from the method offinite differences, a way to interpolate or tabulate functions by using a small set of polynomial co-efficients. Some of the most commonmathematical functions used in engineering, science and navigation are built fromlogarithmic andtrigonometric functions, which can beapproximated by polynomials, so a difference engine can compute many usefultables.
The notion of amechanical calculator for mathematical functions can be traced back to theAntikythera mechanism of the 2nd century BC, while early modern examples are attributed toPascal andLeibniz in the 17th century.
In 1784J. H. Müller, an engineer in theHessian army, devised and built anadding machine and described the basic principles of a difference machine in a book published in 1786 (the first written reference to a difference machine is dated to 1784), but he was unable to obtain funding to progress with the idea.[1][2][3]
Charles Babbage began to construct a small difference engine inc. 1819[4] and had completed it by 1822 (Difference Engine 0).[5] He announced his invention on 14 June 1822, in a paper to theRoyal Astronomical Society, entitled "Note on the application of machinery to the computation of astronomical and mathematical tables".[6] This machine used the decimal number system and was powered by cranking a handle. TheBritish government was interested, since producing tables was time-consuming and expensive and they hoped the difference engine would make the task more economical.[7]
In 1823, the British government gave Babbage £1700 to start work on the project. Although Babbage's design was feasible, the metalworking techniques of the era could not economically make parts in the precision and quantity required. Thus the implementation proved to be much more expensive and doubtful of success than the government's initial estimate. According to the 1830 design for Difference Engine No. 1, it would have about 25,000 parts, weigh 4tons,[8] and operate on 20-digit numbers by sixth-order differences. In 1832, Babbage andJoseph Clement produced a small working model (one-seventh of the plan),[5] which operated on 6-digit numbers by second-order differences.[9][10]Lady Byron described seeing the working prototype in 1833: "We both went to see the thinking machine (or so it seems) last Monday. It raised several Nos. to the 2nd and 3rd powers, and extracted the root of a Quadratic equation."[11] Lady Byron's daughterAda Lovelace would later become fascinated with and work on creating the first computer program intended to solve Bernoulli's equation utilizing the difference engine. Work on the larger engine was suspended in 1833.
By the time the government abandoned the project in 1842,[10][12] Babbage had received and spent over £17,000 on development, which still fell short of achieving a working engine. The government valued only the machine's output (economically produced tables), not the development (at unpredictable cost) of the machine itself. Babbage refused to recognize that predicament.[7] Meanwhile, Babbage's attention had moved on to developing ananalytical engine, further undermining the government's confidence in the eventual success of the difference engine. By improving the concept as an analytical engine, Babbage had made the difference engine concept obsolete, and the project to implement it an utter failure in the view of the government.[7]
The incomplete Difference Engine No. 1 was put on display to the public at the1862 International Exhibition inSouth Kensington, London.[13][14]
Babbage went on to design his much more general analytical engine, but later designed an improved "Difference Engine No. 2" design (31-digit numbers and seventh-order differences),[9] between 1846 and 1849. Babbage was able to take advantage of ideas developed for the analytical engine to make the new difference engine calculate more quickly while using fewer parts.[15][16]
Inspired by Babbage's difference engine in 1834, the Swedish inventorPer Georg Scheutz built several experimental models. In 1837 his son Edward proposed to construct a working model in metal, and in 1840 finished the calculating part, capable of calculating series with 5-digit numbers and first-order differences, which was later extended to third-order (1842). In 1843, after adding the printing part, the model was completed.
In 1851, funded by the government, construction of the larger and improved (15-digit numbers and fourth-order differences) machine began, and finished in 1853. The machine was demonstrated at theWorld's Fair in Paris, 1855 and then sold in 1856 to theDudley Observatory inAlbany, New York. Delivered in 1857, it was the first printing calculator sold.[17][18][19] In 1857 the British government ordered the nextScheutz's difference machine, which was built in 1859.[20][21] It had the same basic construction as the previous one, weighing about 10 cwt (1,100 lb; 510 kg).[19]
Martin Wiberg improved Scheutz's construction (c. 1859, his machine has the same capacity as Scheutz's: 30-digit and sixth-order) but used his device only for producing and publishing printed tables (interest tables in 1860, andlogarithmic tables in 1875).[22]
Alfred Deacon of London inc. 1862 produced a small difference engine (20-digit numbers and third-order differences).[17][23]
AmericanGeorge B. Grant started working on his calculating machine in 1869, unaware of the works of Babbage and Scheutz (Schentz). One year later (1870) he learned about difference engines and proceeded to design one himself, describing his construction in 1871. In 1874 the Boston Thursday Club raised a subscription for the construction of a large-scale model, which was built in 1876. It could be expanded to enhance precision and weighed about 2,000 pounds (910 kg).[23][24][25]
Christel Hamann built one machine (16-digit numbers and second-order differences) in 1909 for the "Tables ofBauschinger and Peters" ("Logarithmic-Trigonometrical Tables with eight decimal places"), which was first published in Leipzig in 1910. It weighed about 40 kilograms (88 lb).[23][26][27]
Burroughs Corporation in about 1912 built a machine for theNautical Almanac Office which was used as a difference engine of second-order.[28]: 451 [29] It was later replaced in 1929 by a Burroughs Class 11 (13-digit numbers and second-order differences, or 11-digit numbers and [at least up to] fifth-order differences).[30]
Alexander John Thompson about 1927 builtintegrating and differencing machine (13-digit numbers and fifth-order differences) for his table of logarithms "Logarithmetica britannica". This machine was composed of four modified Triumphator calculators.[31][32][33]
Leslie Comrie in 1928 described how to use theBrunsviga-Dupla calculating machine as a difference engine of second-order (15-digit numbers).[28] He also noted in 1931 that National Accounting Machine Class 3000 could be used as a difference engine of sixth-order.[23]: 137–138
During the 1980s,Allan G. Bromley, an associate professor at theUniversity of Sydney,Australia, studied Babbage's original drawings for the Difference and Analytical Engines at theScience Museum library in London.[34] This work led the Science Museum to construct a working calculating section of difference engine No. 2 from 1985 to 1991, underDoron Swade, the then Curator of Computing. This was to celebrate the 200th anniversary of Babbage's birth in 1991. In 2002, theprinter which Babbage originally designed for the difference engine was also completed.[35] The conversion of the original design drawings into drawings suitable for engineering manufacturers' use revealed some minor errors in Babbage's design (possibly introduced as a protection in case the plans were stolen),[36] which had to be corrected. The difference engine and printer were constructed to tolerances achievable with 19th-century technology, resolving a long-standing debate as to whether Babbage's design could have worked using Georgian-era engineering methods. The machine contains 8,000 parts and weighs about 5 tons.[37]
The printer's primary purpose is to producestereotype plates for use in printing presses, which it does by pressing type into soft plaster to create aflong. Babbage intended that the Engine's results be conveyed directly to mass printing, having recognized that many errors in previous tables were not the result of human calculating mistakes but from slips in the manualtypesetting process.[7] The printer's paper output is mainly a means of checking the engine's performance.
In addition to funding the construction of the output mechanism for the Science Museum's difference engine,Nathan Myhrvold commissioned the construction of a second complete Difference Engine No. 2, which was on exhibit at theComputer History Museum inMountain View, California, from May 2008 to January 2016.[37][38][39][40] It has since been transferred toIntellectual Ventures inSeattle where it is on display just outside the main lobby.[41][42][43]
 | This sectiondoes notcite anysources. Please helpimprove this section byadding citations to reliable sources. Unsourced material may be challenged andremoved.(June 2017) (Learn how and when to remove this message) |
The difference engine consists of a number of columns, numbered from1 toN. The machine is able to store one decimal number in each column. The machine can only add the value of a columnn + 1 to columnn to produce the new value ofn. ColumnN can only store a constant, column 1 displays (and possiblyprints) the value of the calculation on the currentiteration.
The engine is programmed by setting initial values to the columns. Column 1 is set to the value of the polynomial at the start of computation. Column 2 is set to a value derived from the first and higherderivatives of the polynomial at the same value ofX. Each of the columns from 3 toN is set to a value derived from the first and higher derivatives of the polynomial.[44]
In the Babbage design, one iteration (i.e. one full set of addition andcarry operations) happens for each rotation of the main shaft. Odd and even columns alternately perform an addition in one cycle. The sequence of operations for column is thus:[44]
Steps 1,2,3,4 occur for every odd column, while steps 3,4,1,2 occur for every even column.
While Babbage's original design placed the crank directly on the main shaft, it was later realized that the force required to crank the machine would have been too great for a human to handle comfortably. Therefore, the two models that were built incorporate a 4:1 reduction gear at the crank, and four revolutions of the crank are required to perform one full cycle.
Each iteration creates a new result, and is accomplished in four steps corresponding to four complete turns of the handle shown at the far right in the picture below. The four steps are:
The engine represents negative numbers asten's complements. Subtraction amounts to addition of a negative number. This works in the same manner that modern computers perform subtraction, known astwo's complement.
The principle of a difference engine isNewton's method ofdivided differences. If the initial value of a polynomial (and of itsfinite differences) is calculated by some means for some value ofX, the difference engine can calculate any number of nearby values, using the method generally known as themethod of finite differences. For example, consider the quadraticpolynomial
with the goal of tabulating the valuesp(0),p(1),p(2),p(3),p(4), and so forth. The table below is constructed as follows: the second column contains the values of the polynomial, the third column contains the differences of the two left neighbors in the second column, and the fourth column contains the differences of the two neighbors in the third column:
| x | p(x) = 2x2 − 3x + 2 | diff1(x) = ( p(x + 1) − p(x) ) | diff2(x) = ( diff1(x + 1) − diff1(x) ) |
|---|---|---|---|
| 0 | 2 | −1 | 4 |
| 1 | 1 | 3 | 4 |
| 2 | 4 | 7 | 4 |
| 3 | 11 | 11 | |
| 4 | 22 |
The numbers in the third values-column are constant. In fact, by starting with any polynomial of degreen, the column numbern + 1 will always be constant. This is the crucial fact behind the success of the method.
This table was built from left to right, but it is possible to continue building it from right to left down a diagonal in order to compute more values. To calculatep(5) use the values from the lowest diagonal. Start with the fourth column constant value of 4 and copy it down the column. Then continue the third column by adding 4 to 11 to get 15. Next continue the second column by taking its previous value, 22 and adding the 15 from the third column. Thusp(5) is 22 + 15 = 37. In order to computep(6), we iterate the same algorithm on thep(5) values: take 4 from the fourth column, add that to the third column's value 15 to get 19, then add that to the second column's value 37 to get 56, which isp(6). This process may be continuedad infinitum. The values of the polynomial are produced without ever having to multiply. A difference engine only needs to be able to add. From one loop to the next, it needs to store 2 numbers—in this example (the last elements in the first and second columns). To tabulate polynomials of degreen, one needs sufficient storage to holdn numbers.
Babbage's difference engine No. 2, finally built in 1991, can hold 8 numbers of 31 decimal digits each and can thus tabulate 7th degree polynomials to that precision. The best machines from Scheutz could store 4 numbers with 15 digits each.[45]
The initial values of columns can be calculated by first manually calculating N consecutive values of the function and bybacktracking (i.e. calculating the required differences).
Col gets the value of the function at the start of computation. Col is the difference between and...[46]
If the function to be calculated is apolynomial function, expressed as
the initial values can be calculated directly from the constant coefficientsa0,a1,a2, ...,an without calculating any data points. The initial values are thus:
Many commonly used functions areanalytic functions, which can be expressed aspower series, for example as aTaylor series. The initial values can be calculated to any degree of accuracy; if done correctly the engine will give exact results for first N steps. After that, the engine will only give anapproximation of the function.
The Taylor series expresses the function as a sum obtained from itsderivatives at one point. For many functions the higher derivatives are trivial to obtain; for instance, thesine function at 0 has values of 0 or for all derivatives. Setting 0 as the start of computation we get the simplifiedMaclaurin series
The same method of calculating the initial values from the coefficients can be used as for polynomial functions. The polynomial constant coefficients will now have the value
The problem with the methods described above is that errors will accumulate and the series will tend to diverge from the true function. A solution which guarantees a constant maximum error is to usecurve fitting. A minimum ofN values are calculated evenly spaced along the range of the desired calculations. Using a curve fitting technique likeGaussian reduction anN−1th degreepolynomial interpolation of the function is found.[46] With the optimized polynomial, the initial values can be calculated as above.
{{cite book}}:ISBN / Date incompatibility (help)Alt URLIn WikiSource and also reprinted inThe works of Charles Babbage, Vol 2, p.119ff
Calculator | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Manual | |||||||||||||
| Logarithm |
| ||||||||||||
| Mechanical |
| ||||||||||||
| Electronic |
| ||||||||||||
| Software | |||||||||||||
| Historical | |||||||||||||
