Tape Recording Technology

The date: April 14, 1956

The place: Normandy Lounge, Chicago

The event: Annual CBS-TV affiliates meeting

As the impact of what was happening before them sank in, the silence was broken by an equally sudden roar of applause, cheers, whistles and stamping of feet. The curtain behind the podium opened to reveal a large gray console that looked like a gargantuan audio tape recorder, with overgrown reels of tape on its horizontal top plate. Hovering over this whirring magnetic monster, with looks of obvious relief, were a few of the engineers responsible for developing it.

The race to develop television in the 1920s and '30s by Baird, Zworykin, Farnsworth and many others, was re-enacted in the 1950s by Bing Crosby Enterprises, Ampex, RCA and firms in Germany, France and Japan. All were struggling to develop a means of storing complex video signals on magnetic tape.

The concept of magnetic storage was developed by Vladimir Poulsen (Denmark) in 1899. But the technology remained a laboratory concept until the Germans developed their Magnetophone tape recorder. Starting in 1931, AEG engineers in Berlin began working on the Magnetrophone (a dc-bias machine) while BASF chemists produced the first modern iron-oxide tape formula. The discovery of ac-bias recording (in the U.S.) in 1939-40 led to regular on-air use of the Magnetophone on German radio by 1941.

John T. (Jack) Mullin is credited for bringing the Magnetophone to the United States after the war. Mullin served in the U.S. Army Signal Corps and was assigned to the RAF as a liaison officer concerned with the interchange of technical information. After the invasion of France, Mullin and his group set up shop in Paris with the objective to ferret out developments in which the Germans may have been active during the war and at the time of their retreat. (See also Jack Mullin.)

Mullin's Discovery

About 6 of the Magnetophone machines had been captured and tested as the war drew to a close, but Mullin found them to be quite poor in dynamic range. The background noise, in fact, was not even as good as a 78 RPM shellac record. The distortion was high as well. A British army officer, however, told Mullin about a machine being used at the studios of Radio Frankfurt that provided remarkable dynamic range and low distortion. Mullin, while skeptical, ventured to the station. He later wrote the following account of his visit:

"Radio Frankfurt had vacated the city during heavy bombing raids and had relocated to a large house at a resort spa north of the city in a small town called Bra Nauheim. The station was being directed by the U.S. Armed Forces Radio Service, but the German staff was still operating and maintaining the equipment.

I asked if I might hear one of the tape machines they were using. An order was directed to one of the technicians and I was taken into a room in which there was a large loudspeaker and two of the Magnetophones. The mechanism appeared to be the same as the ones we had in Paris, but there was an obvious difference in the electronics.

The technician placed a roll of tape on one of the machines and started it. Suddenly, out of complete silence, an orchestra bloomed into being with fidelity such as I have never heard in my life. From deep resonant bass to the shimmering of the flute it was all there. It was clean! It was free from any noticeable distortion. And if that were not enough, the dynamic range was fantastic compared with anything I had ever experienced."

Mullin's assistant photographed all of the schematic diagrams and instruction manuals (which were in German), and talked the officer in charge of the Frankfurt station out of a few rolls of tape. Mulling later brought two of the Magnetophones back with him to the United States. After some modifications and tests, he was ready to demonstrate the capabilities of the machines. Through a stroke of good fortune, Mullin was asked to demonstrate the capabilities of his Magnetophones to singer Bing Crosby. (See the related article, The Magic of Magnetic Tape.)

The demonstration was a huge success, and led to the development of the first audio recorder produced in the United States, the Ampex model 200. The machine was first used by ABC in 1948. The rugged design of the model 200, and the financial and political support for Crosby, turned industry skepticism about magnetic recording into widespread confidence -- almost overnight. Crosby later founded Bing Crosby Enterprises to conduct a variety of broadcast and recording research.

Recording Pictures

Although many in broadcasting take video recording for granted, it was not always with us. The early days of television were solidly intertwined with film and film processing, and live programming. By the late 1940s, TV engineers had already developed a whole new field within broadcasting and teleproduction: TV recording. Today, there are thousands of video recording specialists in the world, but back in the late '40s only a few could lay claim to that remarkable title. To understand the impact -- the revolution -- of the videotape recorder on the TV industry, we must look at life before the VTR, the days when video recording meant kinescope.

The kinescope dominated TV recording for time delay in the early 1950s. A kinescope recorder was basically a special 16mm or 35mm film camera mounted in a large box aimed at a high quality monochrome video monitor (a kinescope). All things considered, the kinescope made respectable and reliable TV recordings. Most engineers called the process "kine" (pronounced "kinney") for short.

The kinescope was quite a clever device. Its film camera ran at a speed of 24 frames per second. Because the TV image repeated at 60 fields (or 30 frames) per second, the film had to move intermittently between video frames, and then be rock-steady during exposure. Theoretically, the pulldown period for the film frame should be during the video vertical interval of less than 2ms, something no mechanical contraption could do.

Several kine makers, such as RCA, Acme, General Precision, Eastman Kodak and Palmer Films found various ways around the problem. They developed novel shutter mechanisms that used the extra six frames of the 30 frame video signal to move the film. This action integrated the video half-images into what seemed like smooth 24-frame-per-second film pictures. Of course, the kines were played back on the air using regular 24-frame-per-second film chains, so the conversion to film was complete.

The toughest operational problem with kines was the shutter bar. Although recording engineers tried to get precise synchronization between the camera shutter and the video image, problems often surfaced. The various sources of sync were usually locked to their own primary power references, causing a dark line or shutter bar to drift through the filmed image. One solution involved driving the kine film camera from signals locked to the sync that was coming in with the video source, and then stripping the sync a moment later.

Another problem with kine quality was the gray scale. It was difficult to match the non-linear gray scale characteristics of the video system. Often when engineers were able to get a good film look on a kine recording, the image just didn't reproduce well on the Iconoscope cameras used in film chains. Engineers were continually fighting problems with kinescope phosphors, blooming on white-colored objects, gamma transfer characteristics, variable film sensitivity and chemical processing.

The kine image and optical audio quality often left something to be desired, but the broadcast industry had no choice. All three networks were using kines for time delay to the West Coast.

Hot Kines

During the early- and mid-1950s, the late afternoon hours at the recording departments of the three network West Coast operations centers were a beehive of activity. The fun started early, with shows that were scheduled to air in California at 7:30 p.m. coming down the microwave link from New York at 4:30 p.m. PST, the same shows that affiliates in other time zones were broadcasting live. The only way for the West Coast to accomplish the necessary time delay was through the use of kinescopes.

Using CBS as the example, here's how kine transcription worked: CBS engineers at Television City in Hollywood recorded the show on their Acme 35mm kinescope recorders, with a 16mm Eastman running as backup. As soon as the first 34-minute reel was empty and the switch was make to the second kine, a courier grabbed the first 35mm reel and rushed it over to the film processor on Santa Monica Boulevard, about two miles away. Meanwhile, a second courier, taking a different street route, transported the 16mm backup to the same lab.

As the film emerged from the dryers, it was spooled onto reels and stuffed into cans, which the waiting couriers rushed back across town to the CBS projection room. At the same time, the third and fourth couriers were taking the second reels of 35mm and 16mm film from the studio to the processor. Engineers of the day called these hot kines, because at air time, they sometimes still seemed warm from the film dryer, and they really were hot off the press.

Video historian and veteran video engineer Albert Abramson remembers a night at CBS Television City when the second 35mm reel of a program had not yet returned from the processor and it was almost air time. The breathless courier showed up with only a minute to go. Abramson says he never saw anyone thread and cue up a film chain so fast! To his knowledge, nobody at any of the network operations in Hollywood ever lost a show because of a kine processing foul-up, although they all certainly used the 16mm backup on occasion.

By 1954, the cost of kinescoping for time delay had gone sky-high. American TV operations used more raw film for kines than all of the Hollywood film studios combined. It was estimated that NBC used more than one million feet of film a month in its New York facility alone to time-shift programs. The networks would have gladly paid a king's ransom for an alternative TV recording method. Enter magnetic recording.

Video Magnetic Recording

The earliest known proposal for the use of magnetic recording to store pictures dates back to the late 1920s and a British patent office grant to Boris Ritcheouluff of London. Ritcheouluff designed a picture recorder of considerable ingenuity based on the Poulsen machine, developed in Denmark many years before.

A 1932 German technical publication described a series of schemes for picture telegraphy proposed by Dr. Fritz Schroeter, professor at the Berlin Technischen Hochschule and a director of Telefunken. The illustrations in the book quite clearly resemble present day transverse and helical scanning formats.

In 1938, an Italian inventor by the name of Luigi Marzocci filed a patent application for a variety of rotary head recorders that were clearly intended for sound recording, and so labeled in the patent. Nevertheless, the drawings are amazingly similar to the arcuate and transverse video recording concepts that came later with the first Ampex VTR. In fact, the company that held the rights to Marzocci's invention were prompted to consider legal action against Ampex after they commercialized their VTR.

Even as the Ampex group was developing the transverse recorder, German engineer Eduard Schueller, working in Hamburg for Telefunken, filed for a patent covering a two head helical recorder that is almost identical in concept to some of the machines later produced in the U.S., Japan, and Germany.

The Race Begins

As early as 1950, Jack Mullin, then Bing Crosby's recordist and chief engineer, began working at the newly established electronics division of Crosby Enterprises to develop a magnetic TV recorder. Jack and colleague Wayne Johnson developed some interesting prototype recorders that used fixed heads and high tape speeds to achieve the high head-to-tape velocities needed for TV recording.

The Mullin-Crosby machine started out with 12 tracks at 120-inches/second (ips) on 1-inch tape. Ten tracks carried the monochrome video information, a clock track provided control synchronization and an FM audio track completed the format. The basic theory was to use frequency division multiplexing with the 10 channels covering the desired video range. A bandwidth of 17kHz per channel produced a cumulative spectrum of 1.7MHz, adequate for monochrome reproduction. Sync was recorded on a separate track to avoid the distortion effects of the band separation filters.

By 1952 Crosby Enterprises was inviting broadcasters in to look at credible results of off-air black and white recordings. There were, however, some severe picture impairments that would require much improvement before the recorder could be considered acceptable to broadcasters.

Three years later Mullin had the machine recording color video using only five tracks total: Three for recording RGB information, one for vertical and horizontal sync, and one for the FM audio track. The 1/2-inch wide tape ran at 240ips. People who saw the Crosby video prototype say the picture quality was pretty good, except for a fuzzy screen-door look over the image.

After transverse video recording was introduced, the Crosby project was acquired by 3M Company as their Mincom Division, and was put to other recording tasks in the instrumentation field.

Meanwhile, David Sarnoff, the head of RCA, expected his engineers to come up with a magnetic TV recorder. Sarnoff reportedly wanted the machine and a few other innovations from RCA as a present for the upcoming anniversary of his 50th year in the radio business. The development team was headed by Dr. Harold Olson. The effort resulted in a longitudinal, high speed VTR that made monochrome and color pictures.

The first laboratory prototype was shown publicly in 1953. It used 1/2-inch wide tape and ran at 30-feet/second (ft/s). There were five tracks on the tape. A 7,000 foot reel ran for just 4 minutes. Subsequently, the tape speed was lowered to 20ft/s and the playing time was increased to 15 minutes.

RCA felt confident enough with this advanced color machine to make some program demonstrations at NBC in late 1955. It was even used for about two minutes on the air. The RCA system reduced the incoming color video signal into its R-G-B elements, each of which went on a separate track, as did the synchronizing information. The 5th track carried FM audio information. The stationary magnetic heads of one micron gap width could handle frequencies up to 1.5MHz.

The RCA machine introduced many innovations that became fairly standard on subsequent VTRs. Tape tension servos, eddy current brakes, luminance/chrominance separation and sync reinsertion -- familiar techniques today -- were all part of those early recorders.

The BBC also saw the potential advantages of a vision magnetic recorder, and in 1952 assigned Dr. Peter Axon to lead a long-term effort in that direction. The machine was called VERA (Vision Electronic Recording Apparatus) and was quite different from its contemporaries. Many of the developments achieved in this recorder served as departure points for subsequent helical VTRs that were developed later.

The 1/2-inch tape moved at 16ft/s; a 21-inch diameter reel gave 15 minutes of playing time. VERA was capable of 3MHz bandwidth, a significant achievement for its era in longitudinal recording. The fundamental approach involved the separation of the video signal into two bands: the first for low frequencies up to 100kHz, and the second for frequencies from 100kHz to 3MHz. Low frequencies were recorded using FM while the high frequencies were applied to the tape using AM.

Upon playback, the two frequency bands were recombined to produce an image good enough to be put on the air in 1958. VERA had many unique features, among the most notable being heads made of ferrite cores clad with permalloy with a gap width of 0.5 micron, and a closed-loop tape drive system that held tape speed to within 0.04% by locking to an external sync generator and comparing the sync off-tape with the reference source.

The BBC terminated the project when it became evident that the transverse approach, introduced by Ampex, would become the standard of the industry. As a matter of fact, Dr. Axon also recognized this potential and later became managing director of the Ampex Reading (England) operation.

While little information was coming out of Japan in those days, it later became evident that several Japanese organizations were also experimenting with videotape recording. In 1953, Toshiba was working on a crude helical scan system, which five years later evolved into both single- and dual-head prototypes.

None of the early efforts by Crosby, RCA, the BBC or Toshiba resulted in a product that was ready for the marketplace. It wasn't until a group of young upstarts from Redwood City, CA, perfected their concept of the VTR that the industry's desire for a videotape recorder would be satisfied.

The Ampex Mark IV

In the early 1950s, Ampex Corporation engineers began work that would eventually result in the first practical broadcast quality videotape recorder. In less than a decade, recording technology would advanced from a high frequency limit of 20kHz to video magnetic frequencies ranging to 5MHz or higher.

The development of the videotape recorder was the product of years of hard -- and at times inspired -- work by a 6 man engineering team at Ampex. At times progress was slow, and the project was put on the shelf more than once. The key members of the engineering team that eventually made the electromechanical monster work included:

  • Charles Ginsburg, the team leader
  • Charles Anderson, the "father of FM recording"
  • Ray Dolby, the ingenious engineering student
  • Alex Maxey, the vacuum tension experimenter
  • Fred Pfost, the master of tape head design
  • Shelby Henderson, the model maker

In retrospect, all members of the small team that developed the Ampex VTR credit one individual for its eventual emergence. The accolade is not so much for his technical skills, which were considerable, as for his tenacity and purposeful single-mindedness in the face of every possible adversity that lack of time, money and technical precedent can impose. That man was Charles Ginsburg, a young obscure engineer who joined Ampex in late 1951 specifically to make pictures on tape, and did so to the utter amazement of his peers in the rest of the industry.

At the beginning of the effort in 1951, the concept of the video recorder system involved the use of three tape heads mounted on the surface of a drum, scanning in an arcuate fashion the surface of a 2-inch-wide tape. The tape-to-head speed was to be about 2,500ips (with the tape moving at 30ips). This speed proved to be impractical, however, and eventually was set at 1500ips. Designers hoped the system would give dependable recording frequencies as high as 2.5MHz. Progress was slow in the early months of the project, but in October 1952 engineers were able to demonstrate an almost unrecognizable picture that still was promising enough to maintain management's enthusiasm in the video recording project.

By March 1953, a second machine had been built using four heads mounted on the plane face of the drum (instead of three). An amplitude modulation system was used in the recorder and a switching arrangement selected the proper heads during rotation of the drum. The capstan motor was driven directly from the 60Hz line frequency and the high speed drum motor was driven by a power amplifier whose input was the fifth harmonic of that signal (300Hz). A reference drum motor speed signal was recorded on the tape along with the video information, and used during playback to provide control for the drum motor power amplifier.

In September 1954, after several periods of inactivity, the VTR project was recommenced in earnest, with several significant technical changes. Instead of the arcuate sweep configuration, engineers changed to a geometry that became standard in the late 1950s in which the tape wrapped transversely around the rotating head drum. Consequently, the video information was written across the moving tape in straight lines. An automatic gain control also was developed to compensate for amplitude fluctuations of the rotating head recording system.

Creating the head assembly was a monumental task. Engineers had to develop a head unit that met stringent mechanical and electrical specifications. Complicating the process were the high centrifugal forces that the head assembly faced during operation.

In December 1954, the first picture was demonstrated using the new geometry and modified electronics. In the words of Charles Ginsburg, leader of the Ampex research team, "It took a great deal of faith and understanding to be optimistic in the face of some rather gross shortcomings in the reproduced picture." The decision then was made to attack the video recording process from a different perspective.

A recording system using vestigial sideband frequency modulation was proposed to replace the amplitude modulation technique. In January 1955, the first pictures were seen using the FM video recording system. A few months later, the engineering team gave a convincing demonstration to the Ampex board of directors. Although the resolution was extremely low (the system bandwidth was somewhat less than 1.5MHz) and the video monitor had to be modified to compensate for horizontal instabilities generated by the system, the images produced by the recorder were good enough to persuade management that work should continue.

Development progressed with a number of changes and improvements in the recording system. Vacuum-controlled tape tension was introduced to the unit, and a radical redesign of the tape heads (going to a sandwich-type of construction) was made.

The Final Push

In early February 1956, the engineering team put on a demonstration for what was supposed to be a small Ampex management group. About 30 company people showed up, however, for the historic event. Ginsburg remembers the demonstration this way:

"For all of us on the engineering project, this was the most dramatic demonstration we were to make. The guests arrived, were seated, a few words were spoken to the effect that we would show them what we had produced and the machine was put in the playback mode to reproduce a program we had recorded an hour earlier.

"We then announced that we would record a sequence and immediately play it back. We recorded for about two minutes, rewound the tape and pushed the playback button. Completely silent up to this point, the entire group rose to its feet and shook the building with hand clapping and shouting."

A crash program followed the demonstration, with introduction of the machine to the industry scheduled for the National Association of Radio and Television Broadcasters (later shortened to and hereafter referred to as NAB) Convention in Chicago, just six weeks away. In the weeks that followed the demonstration to Ampex management, a number of visitors were shown the video recorder. Engineering executives from CBS, ABC, CBC and the BBC were sworn to absolute secrecy and ushered in and out separately so they would not see each other.

As the result of a visit by Bill Lodge of CBS, arrangements were made to use a demonstration model, the Mark IV machine, which had not yet been assembled, for a surprise showing to the annual CBS affiliates meeting that was to occur the day before the formal opening of the NAB.

With just six weeks to go before the show, working hours were extended considerably to complete construction of the Mark IV unit, and at the same time continue development work so the picture to be demonstrated in Chicago would be as good as CBS was expecting it to be. The activity became furious. The administrative engineer for the now considerably enlarged group shed his business suit in favor of work shirt and jeans. He spent most of his regular time plus nights and weekends modifying mounting brackets for the new Mark IV console, making cable assemblies and building re-designed electronic assemblies.

A three-year old idea of placing the switching transients in the horizontal blanking interval was rushed into hardware form as the blanking switcher and integrated into the basic system as a toggle switch option. An automatic rotary head deguassing system came into being to eliminate the necessity of manually demagnetizing the video heads after a recording operation and before playback.

Meanwhile, it has been decided that the Mark III machine used for the Ampex management demonstration should be used for a press demo on the same day that the NAB was to begin. Therefore, while Mark III was being used for development work, it also had to be prepared for the press appearance. The true orphan of the project, the audio, which had been sadly neglected up to this point, was made at least to approach professional standards.

Ampex management also gently suggested that a piece of equipment that was to cost upwards of $50,000 should look substantially more polished than the engineering models used for demonstration purposes to date. The prototype machines were housed in crude looking consoles and partially filled racks. A more compact console and rack mounting system was developed to accommodate the wishes of the Ampex brass and the expectations of potential customers.

The long days and weeks paid off. The Mark IV was complete and broken down into a number of pieces for shipment to Chicago. Three days before the press demonstration, however, the Redwood City machine was having severe mechanical trouble. Some of the team left for Chicago and the rest stayed in Redwood City to patch the Mark III back together.

The demonstrations were scheduled for Saturday. By Thursday afternoon the Mark IV was assembled in Chicago and was making the best pictures the team had yet seen on tape. A predictable situation then occurred. The CBS engineering staff said the pictures were not good enough. The signal-to-noise ratio was too low and the noise banding was intolerable.

Between Thursday night and Friday night, the team trimmed and adjusted the machine, aided by the last-minute delivery of new tape samples that greatly exceeded the performance of anything they had used before. The crew in Redwood City, meanwhile, had solved their mechanical problems and was ready for the press demonstration.

The Big Day

There was an air of normal conviviality among the 200 or so station owners and managers attending the CBS television affiliates meeting that preceded the opening of the 1956 NAB convention. As the guests filed into the Normandy Lounge to listen to their headquarters manager, William Lodge, deliver the annual report, they had no inkling that this April 14, 1956 meeting would reveal to the world a technical development that would change television forever.

It was not out of the ordinary at such gatherings to have a camera pointed at the podium and a few monitors set up in the room so as to convey an aura of television. If any of those in attendance noticed an unfamiliar high pitched whine, they probably assumed that it was due to a faulty fluorescent lamp or some external machinery.

Lodge made his prepared presentation with what seemed like a pregnant pause at the end. Suddenly, the monitors in the room were showing what must have seemed to the audience as an impossibility, for they were looking at an instant playback of the Lodge speech, with an image clarity indistinguishable from the original they had seen a few minutes earlier. There was a hushed silence as those in the room tried to relate this assault on their senses, with their prior knowledge that TV images could not be immediately repeated by any known device. Cheers and applause then broke out.

There was no available monitor for the Ampex team behind the curtain to check the playback before punching it up on the TV screens in the other part of the room. As a result, they had to operate on the blind faith that everything was working well. That short silence at the beginning of the playback seemed like an eternity to Ginsburg, Dolby, Anderson and Pfost.

The station executives crowded in around the VTR, trapping the operating personnel against the machine, while they pushed, elbowed and stood on chairs to get a glimpse of this latest video marvel. The VTR crew, who just a few minutes earlier were holding their breath in the fond hope that his hastily assembled contraption would perform on cue, were now busily answering questions from excited interrogators who were naturally curious about performance, price and availability.

After news of the breakthrough spread throughout the convention, pandemonium broke loose and Ampex was flooded with orders. The era of videotape recording had arrived. Engineers who witnessed the event still remark about its historical significance. The demonstration machines were far from finished products, however.

The "Real Work" Begins

The next several months were just hard work for the entire engineering team. Ampex had originally expected to deliver 5 prototype machines beginning in late summer or early fall to customers in government agencies for evaluation, along with a gradual program leading to delivery of machines for TV use starting in 1957. Now, however, the engineers were faced with the pressure of producing 16 hand-built machines, most of them going to broadcasters for immediate on-air use. At the same, they had to gear up for full-scale production of VTRs for which the industry was eager.

Despite the subjectively good pictures demonstrated in April 1956, much work remained to be accomplished. A review of the tasks facing the team before releasing the first machine for air use revealed how shaky the technical ground really was. Some of the key tasks involved:

  • Tape evaluation. Until then, neither the personnel nor machine facilities nor technical advances were sufficient to properly evaluate magnetic tape for video recording. The tape program consumed many hundreds of man-hours and was the cause of severe headaches to the tape manufacturer, Ampex engineering staff and early network customers.
  • Tape head life. The predicted head life for the early machines was just 100 hours. Worse yet, the heads were made in a tedious one-at-a-time fashion. The many parameters in head construction, several of which had been varied madly in cut-and-try fashion to squeeze out a few precious decibels of signal-to-noise ratio for the April demonstrations, had to be frozen to establish some standard before delivery of the first machines to customers. At the same time, head construction had to evolve into a semi-production process, rather than a one-at-a-time handcrafting technique.
  • Processing hardware. A processing unit had to be designed and built capable of providing blanking and sync in the reproduced signal that was sufficiently free of noise to allow the video to be handled without difficulty by conventional stabilization amplifiers and clamps anywhere along the transmission line.
  • Picture improvements. The picture reproduced from tape had to be greatly improved over that shown in April with respect to noise, resolution, overshoot and ringing, and horizontal stability.
  • Mechanical considerations. The entire machine and all individual chassis had to be repackaged and tested. The mechanical design details were endless. The top plate components had to be not only reliable, but completely interchangeable.

On the Air!

The historic first broadcast via videotape was the CBS airing of the Douglas Edward and the News program on November 30, 1956, from New York. CBS Television City in Hollywood replayed the broadcast three hours after it was received on the West Coast. Confidence at CBS in the new machines was not all that high, and for a month, the network ran a backup kinescope just in case of a breakdown.

"Headhunting," as it was known at the time, was a major problem in the early machines. Headhunting caused picture jitter for home viewers with receivers that had horizontal AFC circuits designed for fringe area reception. Viewer complaints instigated a crash program to alleviate the defect before CBS could make good on its threat to take the machines off the air. Ampex engineers literally slept in the VTR room to have access to the Mark IVs during non-broadcast hours, between 1:00 and 5:00 a.m. A lot of midnight oil was (literally) burned to solve the problem.

In the following months, NBC and ABC followed the CBS lead. Ampex built 16 Mark IV production prototypes in 1956-57 -- calling them the VRX-1000 -- before beginning production of the VR-1000 in 1957.

RCA unveiled its version of the VTR, the TRT-1A, in 1957. The machine used the same head and video systems as the Ampex VR-1000. RCA began marketing the VTR process as quadruplex recording, based on the fact that the machine had four video heads. The industry soon adopted the shortened version of the name, terming it quad.

The VTR Grows Up

The introduction of the VTR was only the beginning. Over the following few years, Ampex and RCA continued to modify and enhance their recorders. By 1957 many stations had received their first VTRs, VR-1000s, with KING-TV in Seattle getting serial Number 1. (CBS in Hollywood had earlier received the first production prototype, the VRX-1000.) Both ABC and NBC also received early VRX recorders.

It took a while for the VTR to overtake the kinescope. Stations did not completely trust the new technology and most engineers were unfamiliar with its complicated circuits. The kinescope, although crude, was simple and got the job done.

The early VTRs suffered from a number of problems. Skewing, scalloping, venetian blind effect and incorrect quadrature became common terms among the new breed of video engineers. As the industry became aware of the problems, solutions were developed, one at a time.

One year after the official unveiling of the VR-1000, a rotary head was shown by Ampex on which the quadrature could be adjusted and the heads spaced precisely 90 degrees apart on the head wheel. This cured the major geometric problem of the early quad machines and allowed tape interchange. It did nothing, however, for the jitter (timebase instability) problem.

The lack of interchangeability among the very early VTRs was a serious problem. The same head assembly used to record a program had to be used for playback. Therefore, the record machine head assembly had to be shipped along with the tape just so the program could be properly reproduced.

By 1957, most of the TV industry was using 3M's Number 179 tape. It was the first commercially produced videotape. Ampex actually used a prototype of 179 for its 1956 NAB demonstration. The early tapes suffered from dropouts. At some stations, all incoming reels were checked prior to use. An engineer would load a reel of new tape and record black video for the entire length of it, then watch the entire tape on playback, looking for dropouts.

Compared to kines, videotape was amazingly cheap. A 1-hour reel of 3M 179 cost approximately $300. Compared to the many millions spent each year on kinescope film and processing, the networks would have gladly paid four or five times the original $50,000 price of the VRX-1000. KING-TV in Seattle, the first local station on the air with its own VTR, estimated that videotape cut the cost of recording a program hour from $88 when using film to less than $3.

Color Recorders

Ampex unveiled a modified VR-1000 that produced color pictures at the 1958 NAB convention. Actors Ronald Reagan and Red Skelton were the demonstration hosts. RCA followed later that year with a modification for its VTR to permit color recording. RCA's first color VTR was called the TRT-1AC. Neither recorder, however, provided very good color reproduction.

Work continued on ways to tame the rotational instability of the head, but it was not until the 1960 NAB convention that a solution to the problem was found. At the show, Ampex introduced the Intersync head wheel servo that brought timebase errors down to a peak-to-peak value of 0.15us from the previous peak-to-peak errors of up to 60-70us. This represented a vast improvement, but still was not good enough to record and play back color signals directly. Intersync also allowed several VTRs to be interlocked.

The breakthrough came from Charles Coleman of WBBM-TV, Chicago, who devised a voltage variable delay line whose delay could be varied +/-0.5us by means of an error voltage derived from a comparison of off-tape sync against house (reference) sync. As a result, timebase errors were reduced to 30ns. Rights to the device were acquired by Ampex, and Coleman was hired by the company as well.

Thirty nanoseconds was still not good enough for the long-sought goal of reproducing the color signal directly, so Coleman and associate Peter Jensen developed time error corrector that was similar to the Amtec device, introduced earlier. Amtec was a corrector that removed the jitter and instabilities common to early machines. The new system had a shorter delay line and compared off-tape color burst with reference burst to generate a correction voltage to drive the delay line. Direct color playback became a reality when the new Colortec system was demonstrated at the 1961 NAB show in Washington, D.C.

Further Refinements

With color recording well in hand, engineers then turned their attention to improving signal quality. The original FM carrier frequencies used for recording monochrome signals were totally inadequate for color. Initially, different frequencies and greatly reduced deviation were chosen to reduce offensive moire patterns, however, the reduced deviation made for a noisy picture and so was not totally satisfactory.

The engineers pushed the state-of-the-art again and managed to move the carrier frequencies upward from a range of 5.5 to 6.5MHz to the range of 7 to 10MHz, yielding markedly superior performance.

The next bombshell development came from RCA with the 1962 release of the TR-22, the first all-transistor VTR. The original VTRs required three or more 19-inch racks full of vacuum tube electronics. A typical machine would employ 160 tubes and weighed about 900 pounds. From a maintenance point of view, the solid-state TR-22 was a dream come true. From that point forward, the technology began to resemble what we see today. Solid-state components permitted smaller recorders with more and advanced features. Although the early transistorized VTRs continued to be major maintenance headaches, the lack of tubes was a real plus.

Helical Format Development

Despite industry acceptance of quad in the late 1950s, Toshiba, in Japan, continued to work on its helical scan approach to video recording. A company engineer presented a paper on the concept at the 1960 SMPTE conference in Los Angels, creating a controversy over the relative merits of basic VTR formats. It also caused Ampex to rush to completion an ill-fated 2-inch helical recorder (the VR-8000), which was shipped in heavily constructed, locked boxes to the following NAB convention in Chicago. Flying different aircraft, Ginsburg and Maxey had the only keys to the locks. The VR-8000 was to be demonstrated only if a competitive broadcast helical recorder was shown. There were none and the black crates were returned to Redwood City unopened.

In 1975 Fernseh (Bosch) introduced its BCN series of 1-inch helical recorders at Montreux. Ampex and Sony introduced 1-inch recorders of their own a year later at NAB. All were claimed by their respective proponents to be suitable for broadcast service, but all were different and interchange of tape between the formats was impossible. The Sony and Ampex formats were, however, tantalizingly similar. All three formats worked well, and it was clear at least that helical recording had matured.

Several users of recording equipment asked SMPTE to see if some sort of common 1-inch helical format could be worked out. During 1977 many SMPTE engineering committee meetings were held to see what could be done. The Fernseh format was different enough so that it was defined separately as the B format. Excellent results were obtained, however, in hammering out the C format, which was similar to -- but not identical to -- either the Sony or Ampex machines. In December 1977 SMPTE announced agreement in principle on the C format, and several manufacturers announced plans to supply equipment conforming to it. This effort by SMPTE is a shining example of how the standardization process can work.

Editing Videotape

By 1958, networks and stations were still using VTRs primarily for time-delay broadcasts. True editing capability was not yet a reality. Simple mechanical edits could be performed, but the process was complex and difficult. The edits were often jumpy. Even so, CBS aired the first totally VTR-produced program, Playhouse 90, in 1958.

The early editing process required steady hands and special tools. The video and audio signals were recorded on different sections of the tape. Therefore, if the audio was important or dialogue was involved, extra time was required to layback and resync the audio track.

The required tools included a special microscope or magnifying glass, a cutting block, magnetic developing fluid and degaussed razor blades. The tape had to be cut during the video vertical interval between frames. The early machines often added special edit pulses in the control track along the bottom edge of the tape to help identify the possible edit points.

The first step in making an edit was to locate the proper video point by viewing the tape. Then, the tape oxide was painted with a special developing solution (Ampex called it Edivue). The material contained carbonyl iron. The iron powder of the solution lined up with the magnetic scan lines and the edit pulses. Through this process, the engineer/editor could locate the vertical intervals on the tape.

The editing process caused the loss of one-half second of audio. The missing audio then had to be replaced by re-recording the audio back on the spliced video tape using a second machine. This editing process was far from foolproof. Such work was often referred to as "Kamakazi" editing by the cutters of the day.

To speed the editing process, NBC in Burbank developed a method of editing tape by first using a kinescope film as a work print and later conforming the tape master to the edited film work print. One of the first television shows to utilize this technique was the Fred Astaire Special, edited in 1958. The technique was an immediate success and a flood of television shows were edited that first year. It was really the first attempt at what we now call off-line editing.

Physical splicing of tape continued through the early 1960s, until timecode editing appeared in 1967. Computerized editing was introduced to the industry in 1971.

The More Things Change...

Few products in this industry have changed more than video tape recording. The following list contains most of the video recording formats produced for professional or consumer uses through the late 1980s:

Quad (2-inch)

Sony 2-inch helical

Ampex 2-inch helical

Sony 1-inch helical

Ampex 1-inch helical (type A format)

EIAJ 1/2-inch

1-inch IVC (International Video Corp.) format

2-inch IVC segmented field helical scan

3/4-inch U-Matic

1-inch type B segmented field (Bosch)

1-inch helical type C

Cartivision consumer cassette format

1/2-inch Betamax consumer (Sony)

V-Cord I and V-Cord II 1/2-inch consumer (Toshiba)

VCR and VCR-LP from Philips (those were the actual names)

InstaVision/InstaVideo (Ampex and Toshiba)

AutoVision VX-100 and VX-2000 (Sanyo)

VHS 1/2-inch consumer

8mm consumer

4mm consumer (Samsung)

M format

Betacam format

3/4-inch U-Matic SP

M-II format

Betacam SP

ED-Beta 1/2-inch professional/consumer

Super-VHS 1/2-inch professional/consumer

D1 (component digital, 3/4-inch videotape)

D2 (composite digital, 3/4-inch videotape)

Audio Cartridge Machine

Video products generally capture the attention of most attendees at the annual NAB convention. However, in 1959 the main attraction at the convention was the first audio cartridge recorder. Until then, the playback of program material was handled by discs and reel-to-reel tape decks.

The late '50s was a time of dynamic change in radio as the industry moved from a medium offering long, discrete programs to one offering a single, continuous program made of many short pieces. The rapid-fire, music-and-news format came to be known as Top 40 radio.

Acetate discs could handle the pace of a Top 40 format, but offered poor fidelity. The lead-in grooves tended to sound scratchy after several cueings and, after many plays, the stylus often would jump grooves. Reel-to-reel, on the other hand, offered excellent fidelity but suffered from loading, cueing and tape breakage problems. The need existed for a playback system offering the fidelity of reel-to-reel decks combined with the handling ease, rapid cueing and instant start of acetate discs.

The audio cartridge machine as we know it was developed by two stations working in the late 1950s toward the common goal of devising a better method of playing back commercial announcements. The stations were WJBC-AM, Bloomington, IL, and WWDC-AM, Washington, D.C. The WJBC project began as an idea cooked up by Vern Nolte, general manager of the station, Ted Bailey, chief engineer, and Bailey's assistant, Jack Jenkins. The three had come up with the idea of using short-length cartridges for commercial announcements and having them recue automatically. Design of the electronics to accomplish the scheme belonged to Bailey and Jenkins.

On January 13, 1959, Nolte wrote to Parker Gates, president of the Gates division of Harris Intertype, proposing that the company manufacture and market their cartridge machine. Three days later, Gates wrote back to Nolte to thank him for the offer, but Gates explained that his company was already working on a spot playback device of its own. Gates was, therefore, not interested in developing a product that would duplicate something they already had.

The product to which Parker Gates was referring was the ST-101, better known as the 101 Spotter. Unlike cartridge designs that used an endless loop of 1/4-inch tape, the 101 Spotter employed a 13-inch wide band of tape on which spots were recorded side-by-side. An indexed, sliding head arrangement allowed the user to select any one of 101 spots. Rights to the machine had been purchased earlier by Gates from Nelson Rupard, manager of KIND-AM, Independence, KS.

After having been turned down by Gates, Nolte pitched the idea to Collins Radio. Gene Randolph, then a sales representative for Collins, paid a visit on WJBC early in 1959. Bailey and Jenkins told him about their cartridge tape system, which was then being used on the air to play commercials. As a former control board operator who had handled countless small reels of tape for commercial announcements, Randolph saw what a boon such a machine would be to radio broadcasting.

Randolph returned to the Collins factory in Cedar Rapids, IA, and told his boss about the cartridge system. Within 24 hours, John Haerle -- then product line manager at Collins -- and a company attorney were in Bloomington to work out a marketing agreement with Nolte.

In a crash program, Collins arranged with Nolte's newly-formed Automatic Tape Company (ATC) to produce enough cartridge decks to show at the NAB convention some 6 weeks away. The original models were designated P-150 (for those having a 15-inch mounting panel for desk-top use) and the P-190 (for 19-inch rack mounting). The record amplifier, with a single cue tone, was a separate assembly. More than a thousand units were sold from the floor of the convention that year. The purchase price of a record/play system was $945.

A parallel development effort was, meanwhile, underway in Washington D.C. at station WWDC to develop an endless loop, automatic re-cueing tape cartridge system. Principals in the effort were Ross Beville, chief engineer of the station, and Austin Knox, who had built a crude endless loop machine for background music applications. The two men joined forces and formed Broadcast Electronics on June 18, 1959. The company's product was the Spotmaster cartridge machine. Ben Strouse, who owned WWDC, provided most of the financial support for the company in the early days.

Although Broadcast Electronics was not incorporated until after the Collins/ATC machine was introduced at the 1959 NAB convention, company records show that the first Spotmaster system had been shipped months earlier. So, who was first? It depends upon who you ask.

The two cart machines were similar in design, and certainly similar in function. The primary difference was the automatic release mechanism used on the Collins/ATC system, versus the manual release mechanism used on the Spotmaster.

Tape Takes Over

After tape cartridge equipment was shown at NAB in 1959, it became obvious to Gates/Harris Intertype that the ST-101 was not going to make any significant marketing inroads. There were several limitations to the system: Announcements could not exceed 90 seconds in length and, before the next spot could be selected and played, the tape had to be rewound, similar to a player piano roll.

The machine was set to automatically recue when the 90 second limit of tape travel had been reached. Alternately, the user could initiate a rewind by pressing a front panel switch. Full rewind of the tape could be accomplished in 22 seconds. A spot tape index scale was placed on the front panel to allow selection of the desired announcement. A tape speed of 5 1/2ips was used. Tape cylinders could be replaced in about 15 minutes, if needed. Routine exchange of tapes was not recommended, however.

Faced with the competition of the tape cartridge, Gates began to look for a way to enter the cartridge business. Between the first showing of the Collins/ATC machine in 1959 and 1966, the partnership of the two companies had been dissolved. ATC was manufacturing cart machines on its own in Bloomington. Gates struck a deal with the owners of the company to buy ATC in 1967, and three years later moved the operations to Quincy, IL. At that point, Jack Jenkins and several other former ATC employees founded International Tapetronics Corporation (ITC) in Bloomington. The Gates/Harris Intertype cart machines were marketed as the Criterian line.

The majority of endless-loop tape designs used the Fidelipac cartridge, whose inventor-patents covered the movable pinch roller, which eliminated the need to have a roller mounted inside each cartridge. The Fidelipac drive arrangement was used in some Viking tape cartridge audio players designed for point-of-purchase displays.

Despite widespread enthusiasm for the new cartridge system, there were growing pains. One of the major problems was pressure pad adjustment. If not properly set, the tape audio could sound muffled or the cue tone could be missed. Any early user of cart machines can recall countless times when a tape ran past its cue tone and started playing again, usually over other program material.

Testimony to the classic design of the cartridge machine pioneered by Nolte, Bailey and Jenkins at WJBC, and Beville, Knox and Strauss at WWDC, is the long life it enjoyed.

Bibliography

  • Anderson, Charles, "The Progression of Videotape Recording," Broadcast Engineering, Intertec Publishing, Overland Park, KS, May 1979.
  • Ginsberg, Charles, "The Birth of Videotape Recording", Ampex Corporation, Redwood City, CA.
  • Ginsburg, Charles, "The First VTR: A Historical Perspective," Broadcast Engineering, Intertec Publishing, Overland Park, KS, May 1981.
  • Graubart, Lawrence, "Magnetic Tape Trends," Ampex Corporation Magnetic Tape Division, Redwood City, CA.
  • Hammar, Peter, "The Birth of the VTR," Broadcast Engineering, Intertec Publishing, Overland Park, KS, June 1986.
  • Mullin, Jack, "Discovering Magnetic Tape," Broadcast Engineering, Intertec Publishing, Overland Park, KS, May 1979.
  • Randolph, Gene, "The History of the Cart Machine," Broadcast Engineering, Intertec Publishing, Overland Park, KS, August 1984.
  • "Retrospective: The History of Video Developments," Video Systems, Intertec Publishing, Overland Park, KS, March 1985.
  • Roizen, Joseph, "Video Tape Recorders: A Never Ending Revolution," Broadcast Engineering, Intertec Publishing, Overland Park, KS, April and May 1976.
  • Schneider, Arthur, "The Growth of Videotape Editing," Broadcast Engineering, Intertec Publishing, Overland Park, KS, May 1981.
  • Schow, Edison, "A Review of Television Systems and the Systems for Recording Television," Sound and Video Contractor, Intertec Publishing, Overland Park, KS, May 1989.
  • Schubin, Mark, "From Tiny Tubes to Giant Screens," Video Review, April 1989.
  • Ziff, Richard, "The 25th Anniversary of the Videotape," Broadcast Engineering, Intertec Publishing, Overland Park, KS, April 1981.

Historical Photos

This Magnetophone "model K-4" tape deck is similar to the two machines brought back to San Francisco by Jack Mullin in 1946. Mullin's machines inspired Ampex to produce the first successful U.S. professional audiotape recorder in 1948. (Courtesy of AEG.)

The Great Race to Video Recording

 

Members of the Ampex videotape recorder R&D team pose around the prototype Mark IV VTR that changed TV broadcasting. Shown (from the left) are: Fred Pfost, Shelby Henderson, Ray Dolby, Alex Maxey, Charles Ginsburg (team leader) and Charles Anderson. (Courtesy of Ampex.)

The world's first practical videotape recorder, the Ampex-Mark IV prototype VTR, was unveiled in April 1956 at the National Association of Radio and Television Broadcasters (NAB) convention. (Courtesy of Ampex.)

The first broadcast via videotape on November 30, 1956. CBS Television City in Hollywood, shown, replayed the Douglas Edwards and the News program three hours after it was received on the West Coast. CBS engineer John Radis stands at the controls of the Ampex VRX-1000. Note the racks of support electronics (on the left) for the tape transport. (Courtesy of Ampex.)

The progress of the videotape recording from 1953 to 1956. Clockwise from top left: Shirley Temple, recorded late 1953 using AM signal processing; the Bob Crosby show, 1954, using FM signal processing; an FM videotape picture of a kinescope broadcast, mid-1954; and FM recording, early 1956. (Courtesy of Ampex.)

The early Ampex machines required three racks of tube-filled chassis -- and a lot of maintenance. (Courtesy of Ampex.)

The RCA TR-22 was the first fully transistorized VTR. Solid-state technology allowed the complete machine to be compacted into a single unit 55-inches wide and 72-inches high. (Courtesy of Ampex.)

Audio Spot Recording

 

There is some dispute over which company built the first audio cartridge machine. This Spotmaster deck is at least one of the world's first cart machines, built in 1959. (Courtesy of Broadcast Electronics.)

The Gates ST-101 spot tape recorder. The selector bar on the front panel determined which pre-recorded announcement was played. The spot playback system was introduced at the 1959 NAB convention. (Courtesy of Harris.)

Internal view of the ST-101 showing the magnetic tape cylinder. The roll could be replaced in the field in about 15 minutes, if necessary. (Courtesy of Harris.)

Side view of the magnetic tape cylinder of the Gates ST-101 spot recorder/player. (Courtesy of Harris.)

Return to Top of Page.