Physics

In some ways we would not have computers today were it not for physics. Furthermore, the needs of physics have stimulated computer development at every step. This all started due to one man's desire to eliminate needless work by transferring it to a machine.

Charles Babbage (1791–1871) was a well-to-do Englishman attending Cambridge University in the early 1800s. One day he was nodding off over a book containing tables of astronomical phenomena. He fancied that he would become an astronomical mathematician. The motion of heavenly bodies was, of course, governed by the laws of physics. For a moment, he thought of having the tables calculated automatically. This idea came up several times in succeeding years until he finally designed a calculator, the Difference Engine, that could figure the numbers and print the tables. A version of the Difference Engine made by someone else found its way to the Dudley Observatory in Albany, New York, where it merrily cranked out numbers until the 1920s. Babbage followed this machine with a programmable version, the Analytical Engine, which was never built. The Analytical Engine, planned as a more robust successor to the Difference Engine, is considered by many to be the first example of a modern computer.

In the late 1800s, mathematician and scientist Lord Kelvin (William Thomson) (1824–1907) tried to understand wave phenomena by building a mechanical analog computer that modeled the waves on beaches in England. This was a continuation of the thread of mechanical computation applied to understand physical phenomena in the 1800s.

In the 1920s, physicist Vannevar Bush (1890–1974) of the Massachusetts Institute of Technology built a Differential Analyzer that used a combination of mechanical and electrical parts to create an analog computer useful for many problems. The Differential Analyzer was especially suited for physics calculations, as its output was a smooth curve showing the results of mathematical modeling. This curve was very accurate, more so than the slide rules that were the ubiquitous calculators in physics and engineering in the first seven decades of the twentieth century.

Beginning during World War II and finishing just after the war ended, the Moore School of the University of Pennsylvania built an electronic digital computer for the U.S. Army. One of the first problems run on it was a model of a nuclear explosion. The advent of digital computers opened up whole new realms of research for physicists.

Physicists like digital computers because they are fast. Thus, big problems can be figured out, and calculations that are boring and repetitious by hand can be transferred to computers. Some of the first subroutines, blocks of computer code executed many times during the run of a program, were inspired by the needs of physics.

Even though digital computers were fast with repetitious tasks, the use of approximation and visualization has the largest effect on physicists using electronic computers. Analog machines, both mechanical and electronic, have output that models real world curves and other shapes representing certain kinds of mathematics. To calculate the mathematical solution of physical problems on digital computers meant the use of approximation. For example, the area under a curve (the integral) is approximated by dividing the space below the curve into rectangles, figuring out their area, and adding the small areas to find the one big area. As computers got faster, such approximations were made up of an ever-increasing number of smaller rectangles.

Visualization is probably the physicist's task most aided by computers. The outputs of Lord Kelvin's machine and the Differential Analyzer were drawn by pens connected to the computational components of the machine. The early digital computers could print rough curves, supplemented by cleaner curves done on a larger scale by big plotters. Interestingly, the plotters drew what appeared to be smooth lines by drawing numerous tiny straight lines, just like a newspaper photograph is really a large number of gray points with different shades. Even these primitive drawing tools were a significant advance. They permitted physicists to see much more than could be calculated by hand.

In the 1960s, physicists took millions of photographs of sub-atomic particle collisions. These were then processed with human intervention. A "scanner" (usually a student) using a special machine would have the photographs of the collisions brought up one by one. The scanner would use a trackball to place a cursor over a sub-atomic particle track. At each point the scanner would press a button, which then allowed the machine to punch the coordinates on a card. These thousands upon thousands of cards were processed to calculate the mass and velocity of the various known and newly discovered particles. These were such big jobs that they were often run on a computer overnight. Physicists could use the printed output of batch-type computer systems to visualize mentally what was really happening. This is one of the first examples of truly large-scale computing. In fact, most of the big calculations done over the first decades of electronic digital computing had some relationship to physics, including atomic bomb models, satellite orbits, and cyclotron experiments.

The advent of powerful workstations and desktop systems with color displays ended the roughness and guessing of early forms of visualization. Now, many invisible phenomena, such as fields, waves, and quantum mechanics, can be modeled accurately in full color. This is helping to eliminate erroneous ideas inspired by the poor visualizations of years past. Also, these computer game–quality images can be used to train the next generation of physics students and their counterparts in chemistry and biology classes, making tangible what was invisible before.

Finally, the latest and perhaps most pervasive of physics-inspired computer developments is the World Wide Web. It was first developed as a way of easily sharing data, including graphics, among researchers in the European cyclotron community and also for those outside of it with appropriate interests. So whenever a browser is launched, 200 years of physics-driving computer development is commemorated.