The human eye's detection system, which is in the retina, consists primarily of two types of light detectors: rod cells that capture light, dark, and shapes/figures, and the cone cells that detect color. A typical retina contains 120 million rods and 4.5 million to 6 million cones, which are divided among three groups that are sensitive to red, green, and blue light. This means that the eye has far more resolution in brightness, or "luminance", than in color. However, post-processing of the optic nerve and other portions of the human visual system combine the information from the rods and cones to re-create what appears to be a high-resolution color image.
The eye has limited bandwidth to the rest of the visual system, estimated at just under 8 Mbit/s. This manifests itself in a number of ways, but the most important in terms of producing moving images is the way that a series of still images displayed in quick succession will appear to be continuous smooth motion. This illusion starts to work at about 16 frame/s, and common motion pictures use 24 frame/s. Television, using power from the electrical grid, tunes its rate in order to avoid interference with the alternating current being supplied – in North America, some Central and South American countries, Taiwan, Korea, part of Japan, the Philippines, and a few other countries, this is 60 video fields per second to match the 60 Hz power, while in most other countries it is 50 fields per second to match the 50 Hz power.
In its most basic form, a color broadcast can be created by broadcasting three monochrome images, one each in the three colors of red, green, and blue (RGB). When displayed together or in rapid succession, these images will blend together to produce a full-color image as seen by the viewer. One of the great technical challenges of introducing color broadcast television was the desire to conserve bandwidth, potentially three times that of the existing black-and-white standards, and not use an excessive amount of radio spectrum. In the United States, after considerable research, the National Television Systems Committee approved an all-electronic system developed by RCA which encoded the color information separately from the brightness information and greatly reduced the resolution of the color information in order to conserve bandwidth. The brightness image remained compatible with existing black-and-white television sets at slightly reduced resolution, while color televisions could decode the extra information in the signal and produce a limited-resolution color display. The higher resolution black-and-white and lower resolution color images combine in the eye to produce a seemingly high-resolution color image. The NTSC standard represented a major technical achievement.
Experiments in television systems using radio broadcasts date to the 19th century, but it was not until the 20th century that advances in electronics and light detectors made development practical. A key problem was the need to convert a 2D image into a "1D" radio signal; some form of image scanning was needed to make this work. Early systems generally used a device known as a "Nipkow disk", which was a spinning disk with a series of holes punched in it that caused a spot to scan across and down the image. A single photodetector behind the disk captured the image brightness at any given spot, which was converted into a radio signal and broadcast. A similar disk was used at the receiver side, with a light source behind the disk instead of a detector.
A number of such systems were being used experimentally in the 1920s. The best-known was John Logie Baird's, which was actually used for regular public broadcasting in Britain for several years. Indeed, Baird's system was demonstrated to members of the Royal Institution in London in 1926 in what is generally recognized as the first demonstration of a true, working television system. In spite of these early successes, all mechanical television systems shared a number of serious problems. Being mechanically driven, perfect synchronization of the sending and receiving discs was not easy to ensure, and irregularities could result in major image distortion. Another problem was that the image was scanned within a small, roughly rectangular area of the disk's surface, so that larger, higher-resolution displays required increasingly unwieldy disks and smaller holes that produced increasingly dim images. Rotating drums bearing small mirrors set at progressively greater angles proved more practical than Nipkow discs for high-resolution mechanical scanning, allowing images of 240 lines and more to be produced, but such delicate, high-precision optical components were not commercially practical for home receivers.
It was clear to a number of developers that a completely electronic scanning system would be superior, and that the scanning could be achieved in a vacuum tube via electrostatic or magnetic means. Converting this concept into a usable system took years of development and several independent advances. The two key advances were Philo Farnsworth's electronic scanning system, and Vladimir Zworykin's Iconoscope camera. The Iconoscope, based on Kálmán Tihanyi's early patents, superseded the Farnsworth-system. With these systems, the BBC began regularly scheduled black-and-white television broadcasts in 1936, but these were shut down again with the start of World War II in 1939. In this time thousands of television sets had been sold. The receivers developed for this program, notably those from Pye Ltd., played a key role in the development of radar.
By 22 March 1935, 180-line black-and-white television programs were being broadcast from the Paul Nipkow TV station in Berlin. In 1936, under the guidance of "Minister of Public Enlightenment and Propaganda" Joseph Goebbels, direct transmissions from fifteen mobile units at the Olympic Games in Berlin were transmitted to selected small television houses (Fernsehstuben) in Berlin and Hamburg.
In 1941, the first NTSC meetings produced a single standard for US broadcasts. US television broadcasts began in earnest in the immediate post-war era, and by 1950 there were 6 million televisions in the United States.
The basic idea of using three monochrome images to produce a color image had been experimented with almost as soon as black-and-white televisions had first been built.
Among the earliest published proposals for television was one by Maurice Le Blanc in 1880 for a color system, including the first mentions in television literature of line and frame scanning, although he gave no practical details. Polish inventor Jan Szczepanik patented a color television system in 1897, using a selenium photoelectric cell at the transmitter and an electromagnet controlling an oscillating mirror and a moving prism at the receiver. But his system contained no means of analyzing the spectrum of colors at the transmitting end, and could not have worked as he described it. An Armenian inventor, Hovannes Adamian, also experimented with color television as early as 1907. The first color television project is claimed by him, and was patented in Germany on March 31, 1908, patent number 197183, then in Britain, on April 1, 1908, patent number 7219, in France (patent number 390326) and in Russia in 1910 (patent number 17912).
Scottish inventor John Logie Baird demonstrated the world's first color transmission on July 3, 1928, using scanning discs at the transmitting and receiving ends with three spirals of apertures, each spiral with filters of a different primary color; and three light sources, controlled by the signal, at the receiving end, with a commutator to alternate their illumination. The demonstration was of a young girl wearing different colored hats. The girl, Noele Gordon, later became a TV actress in the soap opera Crossroads. Baird also made the world's first color broadcast on February 4, 1938, sending a mechanically scanned 120-line image from Baird's Crystal Palace studios to a projection screen at London's Dominion Theatre.
Mechanically scanned color television was also demonstrated by Bell Laboratories in June 1929 using three complete systems of photoelectric cells, amplifiers, glow-tubes, and color filters, with a series of mirrors to superimpose the red, green, and blue images into one full color image.
This section needs additional citations for verification
. (September 2012)
As was the case with black-and-white television, an electronic means of scanning would be superior to the mechanical systems like Baird's. The obvious solution on the broadcast end would be to use three conventional Iconoscopes with colored filters in front of them to produce an RGB signal. Using three separate tubes each looking at the same scene would produce slight differences in parallax between the frames, so in practice a single lens was used with a mirror or prism system to separate the colors for the separate tubes. Each tube captured a complete frame and the signal was converted into radio in a fashion essentially identical to the existing black-and-white systems.
The problem with this approach was there was no simple way to recombine them on the receiver end. If each image was sent at the same time on different frequencies, the images would have to be "stacked" somehow on the display, in real time. The simplest way to do this would be to reverse the system used in the camera: arrange three separate black-and-white displays behind colored filters and then optically combine their images using mirrors or prisms onto a suitable screen, like frosted glass. RCA built just such a system in order to present the first electronically scanned color television demonstration on February 5, 1940, privately shown to members of the US Federal Communications Commission at the RCA plant in Camden, New Jersey. This system, however, suffered from the twin problems of costing at least three times as much as a conventional black-and-white set, as well as having very dim pictures, the result of the fairly low illumination given off by tubes of the era. Projection systems of this sort would become common decades later, however, with improvements in technology.
Another solution would be to use a single screen, but break it up into a pattern of closely spaced colored phosphors instead of an even coating of white. Three receivers would be used, each sending its output to a separate electron gun, aimed at its colored phosphor. Although obvious, this solution was not practical. The electron guns used in monochrome televisions had limited resolution, and if one wanted to retain the resolution of existing monochrome displays, the guns would have to focus on individual dots three times smaller. This was beyond the state of the art.
Instead, a number of hybrid solutions were developed that combined a conventional monochrome display with a colored disk or mirror. In these systems the three colored images were sent one after each other, in either complete frames in the "field-sequential color system", or for each line in the "line-sequential" system. In both cases a colored filter was rotated in front of the display in sync with the broadcast. Since three separate images were being sent in sequence, if they used existing monochrome radio signaling standards they would have an effective refresh rate of only 20 fields, or 10 frames, a second, well into the region where flicker would become visible. In order to avoid this, these systems increased the frame rate considerably, making the signal incompatible with existing monochrome standards.
The first practical example of this sort of system was again pioneered by John Logie Baird. In 1940 he publicly demonstrated a color television combining a traditional black-and-white display with a rotating colored disk. This device was very "deep", but was later improved with a mirror folding the light path into an entirely practical device resembling a large conventional console. However, Baird was not happy with the design, and as early as 1944 had commented to a British government committee that a fully electronic device would be better.
In 1939, Hungarian engineer Peter Carl Goldmark introduced an electro-mechanical system while at CBS, which contained an Iconoscope sensor. The CBS field-sequential color system was partly mechanical, with a disc made of red, blue, and green filters spinning inside the television camera at 1,200 rpm, and a similar disc spinning in synchronization in front of the cathode ray tube inside the receiver set. The system was first demonstrated to the Federal Communications Commission (FCC) on August 29, 1940, and shown to the press on September 4.
CBS began experimental color field tests using film as early as August 28, 1940, and live cameras by November 12. NBC (owned by RCA) made its first field test of color television on February 20, 1941. CBS began daily color field tests on June 1, 1941. These color systems were not compatible with existing black-and-white television sets, and as no color television sets were available to the public at this time, viewing of the color field tests was restricted to RCA and CBS engineers and the invited press. The War Production Board halted the manufacture of television and radio equipment for civilian use from April 22, 1942 to August 20, 1945, limiting any opportunity to introduce color television to the general public.
This live image of Paddy Naismith was used to demonstrate John Logie Baird
's first all-electronic color television system, which used two projection CRTs. The two-color image would be similar to the basic telechrome system.
As early as 1940, Baird had started work on a fully electronic system he called the "Telechrome". Early Telechrome devices used two electron guns aimed at either side of a phosphor plate. The phosphor was patterned so the electrons from the guns only fell on one side of the patterning or the other. Using cyan and magenta phosphors, a reasonable limited-color image could be obtained. He also demonstrated the same system using monochrome signals to produce a 3D image (called "stereoscopic" at the time). Baird's demonstration on August 16, 1944 was the first example of a practical color television system. Work on the Telechrome continued and plans were made to introduce a three-gun version for full color. However, Baird's untimely death in 1946 ended development of the Telechrome system.
Similar concepts were common through the 1940s and 50s, differing primarily in the way they re-combined the colors generated by the three guns. The Geer tube was similar to Baird's concept, but used small pyramids with the phosphors deposited on their outside faces, instead of Baird's 3D patterning on a flat surface. The Penetron used three layers of phosphor on top of each other and increased the power of the beam to reach the upper layers when drawing those colors. The Chromatron used a set of focusing wires to select the colored phosphors arranged in vertical stripes on the tube.
In the immediate post-war era the Federal Communications Commission (FCC) was inundated with requests to set up new television stations. Worrying about congestion of the limited number of channels available, the FCC put a moratorium on all new licenses in 1948 while considering the problem. A solution was immediately forthcoming; rapid development of radio receiver electronics during the war had opened a wide band of higher frequencies to practical use, and the FCC set aside a large section of these new UHF bands for television broadcast. At the time, black and white television broadcasting was still in its infancy in the U.S., and the FCC started to look at ways of using this newly available bandwidth for color broadcasts. Since no existing television would be able to tune in these stations, they were free to pick an incompatible system and allow the older VHF channels to die off over time.
The FCC called for technical demonstrations of color systems in 1948, and the Joint Technical Advisory Committee (JTAC) was formed to study them. CBS displayed improved versions of its original design, now using a single 6 MHz channel (like the existing black-and-white signals) at 144 fields per second and 405 lines of resolution. Color Television Inc. demonstrated its line-sequential system, while Philco demonstrated a dot-sequential system based on its beam-index tube-based "Apple" tube technology. Of the entrants, the CBS system was by far the best-developed, and won head-to-head testing every time.
While the meetings were taking place it was widely known within the industry that RCA was working on a dot-sequential system that was compatible with existing black-and-white broadcasts, but RCA declined to demonstrate it during the first series of meetings. Just before the JTAC presented its findings, on August 25, 1949, RCA broke its silence and introduced its system as well. The JTAC still recommended the CBS system, and after the resolution of an ensuing RCA lawsuit, color broadcasts using the CBS system started on June 25, 1951. By this point the market had changed dramatically; when color was first being considered in 1948 there were fewer than a million television sets in the U.S., but by 1951 there were well over 10 million. The idea that the VHF band could be allowed to "die" was no longer practical.
During its campaign for FCC approval, CBS gave the first demonstrations of color television to the general public, showing an hour of color programs daily Mondays through Saturdays, beginning January 12, 1950, and running for the remainder of the month, over WOIC in Washington, D.C., where the programs could be viewed on eight 16-inch color receivers in a public building. Due to high public demand, the broadcasts were resumed February 13–21, with several evening programs added. CBS initiated a limited schedule of color broadcasts from its New York station WCBS-TV Mondays to Saturdays beginning November 14, 1950, making ten color receivers available for the viewing public. All were broadcast using the single color camera that CBS owned. The New York broadcasts were extended by coaxial cable to Philadelphia's WCAU-TV beginning December 13, and to Chicago on January 10, making them the first network color broadcasts.
After a series of hearings beginning in September 1949, the FCC found the RCA and CTI systems fraught with technical problems, inaccurate color reproduction, and expensive equipment, and so formally approved the CBS system as the U.S. color broadcasting standard on October 11, 1950. An unsuccessful lawsuit by RCA delayed the first commercial network broadcast in color until June 25, 1951, when a musical variety special titled simply Premiere was shown over a network of five East Coast CBS affiliates. Viewing was again restricted: the program could not be seen on black-and-white sets, and Variety estimated that only thirty prototype color receivers were available in the New York area. Regular color broadcasts began that same week with the daytime series The World Is Yours and Modern Homemakers.
While the CBS color broadcasting schedule gradually expanded to twelve hours per week (but never into prime time), and the color network expanded to eleven affiliates as far west as Chicago, its commercial success was doomed by the lack of color receivers necessary to watch the programs, the refusal of television manufacturers to create adapter mechanisms for their existing black-and-white sets, and the unwillingness of advertisers to sponsor broadcasts seen by almost no one. CBS had bought a television manufacturer in April, and in September 1951, production began on the only CBS-Columbia color television model, with the first color sets reaching retail stores on September 28. But it was too little, too late. Only 200 sets had been shipped, and only 100 sold, when CBS discontinued its color television system on October 20, 1951, ostensibly by request of the National Production Authority for the duration of the Korean War, and bought back all the CBS color sets it could to prevent lawsuits by disappointed customers. RCA chairman David Sarnoff later charged that the NPA's order had come "out of a situation artificially created by one company to solve its own perplexing problems" because CBS had been unsuccessful in its color venture.
While the FCC was holding its JTAC meetings, development was taking place on a number of systems allowing true simultaneous color broadcasts, "
dot-sequential color systems". Unlike the hybrid systems, dot-sequential televisions used a signal very similar to existing black-and-white broadcasts, with the intensity of every dot on the screen being sent in succession.
In 1938 Georges Valensi demonstrated an encoding scheme that would allow color broadcasts to be encoded so they could be picked up on existing black-and-white sets as well. In his system the output of the three camera tubes were re-combined to produce a single "luminance" value that was very similar to a monochrome signal and could be broadcast on the existing VHF frequencies. The color information was encoded in a separate "chrominance" signal, consisting of two separate signals, the original blue signal minus the luminance (B'–Y'), and red-luma (R'–Y'). These signals could then be broadcast separately on a different frequency; a monochrome set would tune in only the luminance signal on the VHF band, while color televisions would tune in both the luminance and chrominance on two different frequencies, and apply the reverse transforms to retrieve the original RGB signal. The downside to this approach is that it required a major boost in bandwidth use, something the FCC was interested in avoiding.
RCA used Valensi's concept as the basis of all of its developments, believing it to be the only proper solution to the broadcast problem. However, RCA's early sets using mirrors and other projection systems all suffered from image and color quality problems, and were easily bested by CBS's hybrid system. But solutions to these problems were in the pipeline, and RCA in particular was investing massive sums (later estimated at $100 million) to develop a usable dot-sequential tube. RCA was beaten to the punch by the Geer tube, which used three B&W tubes aimed at different faces of colored pyramids to produce a color image. All-electronic systems included the Chromatron, Penetron and beam-index tube that were being developed by various companies. While investigating all of these, RCA's teams quickly started focusing on the shadow mask system.
In July 1938 the shadow mask color television was patented by Werner Flechsig (1900–1981) in Germany, and was demonstrated at the International radio exhibition Berlin in 1939. Most CRT color televisions used today are based on this technology. His solution to the problem of focusing the electron guns on the tiny colored dots was one of brute-force; a metal sheet with holes punched in it allowed the beams to reach the screen only when they were properly aligned over the dots. Three separate guns were aimed at the holes from slightly different angles, and when their beams passed through the holes the angles caused them to separate again and hit the individual spots a short distance away on the back of the screen. The downside to this approach was that the mask cut off the vast majority of the beam energy, allowing it to hit the screen only 15% of the time, requiring a massive increase in beam power to produce acceptable image brightness.
In spite of these problems in both the broadcast and display systems, RCA pressed ahead with development and was ready for a second assault on the standards by 1950.
The possibility of a compatible color broadcast system was so compelling that the NTSC decided to re-form, and held a second series of meetings starting in January 1950. Having only recently selected the CBS system, the FCC heavily opposed the NTSC's efforts. One of the FCC Commissioners, R. F. Jones, went so far as to assert that the engineers testifying in favor of a compatible system were "in a conspiracy against the public interest".
Unlike the FCC approach where a standard was simply selected from the existing candidates, the NTSC would produce a board that was considerably more pro-active in development.
Starting before CBS color even got on the air, the U.S. television industry, represented by the National Television System Committee, worked in 1950–1953 to develop a color system that was compatible with existing black-and-white sets and would pass FCC quality standards, with RCA developing the hardware elements. ("Compatible color," a phrase from advertisements for early sets, appears in the song "America" of West Side Story, 1957.) RCA first made publicly announced field tests of the dot sequential color system over its New York station WNBT in July 1951. When CBS testified before Congress in March 1953 that it had no further plans for its own color system, the National Production Authority dropped its ban on the manufacture of color television receivers, and the path was open for the NTSC to submit its petition for FCC approval in July 1953, which was granted on December 17. The first publicly announced network demonstration of a program using the NTSC "compatible color" system was an episode of NBC's Kukla, Fran and Ollie on August 30, 1953, although it was viewable in color only at the network's headquarters. The first network broadcast to go out over the air in NTSC color was a performance of the opera Carmen on October 31, 1953.