Regular Scheduled HD Television is 75 Years Old

[Penthode]

On November 2 1936, the BBC began regular scheduled HD television, the first broadcaster to do so in the world. It was referred to as the "High Definition" service by virtue of it being (in the EMI System’s case) fully electronic and capable of 405 interlaced scanning line resolution. This was no small feat in 1936: the Baird system was 240 lines progressive and RCA was experimenting with 343 lines interlaced at this time.

The BBC studio and transmitter was located at Alexandra Palace in North London. The transmitter was constructed by Marconi and the studio systems were divided between the Baird Intermediate Film system and the EMI system which employed the "Emitron" tube camera. The Emitron was EMI's version of the Zworykin iconoscope.

The link between EMI and RCA is interesting. RCA merged with the Victor Talking Machine Company in January 1929. EMI was formed as an amalgamation of the Gramophone Company ("His Master's Voice") and Columbia Phonograph. The Gramophone Company was formed in 1898 from Emile Berliner's Gramophone. The same Emile Berliner provided the impetuous for Eldrige Johnson to form Victor in 1901. Victor actually acquired The Gramophone Company in 1920 and although the tie diluted by the late 20's, the link remained after the RCA merger, hence the sharing of early television patents. EMI must be credited with much of its own television development which resulted in a very advanced system for its day.

A Television Commission was formed in 1935 to help determine the future of the BBC Television service. However, with much apparent lobbying from Baird Television Limited, a clear decision as to what system to be used could not be made prior to the launch. Hence the Baird System ran alternate weeks with the EMI system until February 1937 when the EMI system was declared the best system. Baird did not employ interlace hence with a picture repetition rate of 25Hz, the flicker proved a considerable distraction. Less so maybe because of the earlier dim displays minimized the flicker to a degree. Baird's attempt to stay in the running was set back by the Crystal Palace fire later in November 1936 as that was where his company was located. (On a side note, I many years ago acquired a small machinist’s lathe which was purchased by an ex-Baird employee after the fire. I later acquired the lathe from his son. I now use it regularly in my workshop. It dates from the early thirties and I imagine it was used possibly to manufacture mechanical TV parts!)

Anyhow apart from the intervention of WWII and a three week hiatus in late summer 1937, the 405 line system in London (on the same "channel 1 frequency) continued running until January 3, 1985 to the final closedown at 12:10 am. The original BBC Television from 1936 (now known as BBC1) and ITV (Independent Television) from 1955 ran on 405 lines exclusively until 1969. BBC 2 began in 1964 as the first UK 625 line service and BBC1 and ITV began dual 625/405 transmissions from 1969 until 1985.

I have attached a few photos of 1936-7 paraphernalia.

1) You will see the first TV Listings supplement from the "Radio Times" from Monday November 2, 1936(The Radio Times is the BBC's magazine of listings and information first published by the BBC in 1923 and still published!) Note that the television service was for just two hours per day: 3 to 4pm and 9 to 10pm.

2) I have included an ad for EMI Televisions. Note the “His Master’s Voice” logo. The link extends before Victor. The original painting hangs today in the EMI Board Room. Strange as it may seem, the dog and the gramophone trademark was first used by the Berliner Gramophone Company in Canada in 1901. Eldridge Johnson used is a little later in 1902 and it did not appear on Gramophone Company products until 1909, which it displaced the “Recording Angel”. Since RCA and its successor BMG now have rights to the trademark usage in the US, EMI uses the recording angel in the US and Canada. JVC (the Japan Victor Company) has rights for its use in Japan.

3) I thought the January 1937 article on photographing television interesting. Note the photograph from the play “Marigold" which was broadcast the first week: you will see it in the listings.

4) You will find below the You Tube link to the BBC Film “Television Comes to London”. You will discover on the first evening's telecast at 9pm November 2, 1936. Look for Vladimir Zworykin's visit during the construction of the Alexandra Palace station.

http://www.youtube.com/watch?v=pps6aWcvReA (part 1)

http://www.youtube.com/watch?v=wbRPeRbZDIA (part 2)

[tubesrule]

Nice writeup and pictures Penthode. One comment, RCA was using 441 line as of June 1936, which was to become the pre-war American standard.

Darryl

[Penthode]

Darryl, that is a good question. I believe as of November 1936, RCA was still using 343 lines and 441 lines wasn't used until 1937/8.

Do you have a specific reference to indicate this? I shall meanwhile look into the RCA Reviews of the period.

Thanks.

[Penthode]

I found this on the ETF Website:


During late 1935 and early 1936 RCA manufactured a few nine-inch field test television receivers, including the RR-359 in our collection. The RR-359 started as a 9-inch mirror-in-the-lid set... Initially the set scanned 343 lines (interlaced) and had a continuous tuner capable of tuning 40 to 90 MHz which frequencies include the current low-band VHF channels 2 to 6 and below to the now defunct channel 1...

In 1937, RCA and other experimental broadcasters moved up to 441-lines, AM sound. This was the RMA (Radio Manufacturer’s Association) standard at the time. It is probable that RCA transferred its old call sign W2XBS for these broadcasts. Programming was extended to include pickups remote from the studio. NBC’s mobile television vans, then a great curiosity, appeared on the streets of New York for the first time on December 12, 1937.

[tubesrule]

Originally Posted by Penthode

I found this on the ETF Website:


During late 1935 and early 1936 RCA manufactured a few nine-inch field test television receivers, including the RR-359 in our collection. The RR-359 started as a 9-inch mirror-in-the-lid set... Initially the set scanned 343 lines (interlaced) and had a continuous tuner capable of tuning 40 to 90 MHz which frequencies include the current low-band VHF channels 2 to 6 and below to the now defunct channel 1...

In 1937, RCA and other experimental broadcasters moved up to 441-lines, AM sound. This was the RMA (Radio Manufacturer’s Association) standard at the time. It is probable that RCA transferred its old call sign W2XBS for these broadcasts. Programming was extended to include pickups remote from the studio. NBC’s mobile television vans, then a great curiosity, appeared on the streets of New York for the first time on December 12, 1937.

Penthode,
I actually provided much of the info for that article While the RMA didn't officially adopt the 441 standard until 1937, RCA was using it exclusively from mid 1936 on.

The original letter kicking off the RR359 project is dated November 25 1935. It states that the first sample should be completed no later than December 24, 1935 for evaluation. These first sets were 9" and used the then current 343i format. These 9" sets were only used for the first few months of testing and most were retrofitted with 12" crt's. The 12" versions of the sets, the RR359B's used the 441i standard. Because most of the 9" sets were retrofitted to 12", very few of the 9" sets exist today. The main chassis and power supply were all that was salvaged from the 9" sets. The cabinets and crts on the 12" sets were all new. The various RR359/B's went through many design changes like a continuous tuner to a detent tuner, higher bandwidth IF modules, a change from A306 prototype IF tubes (similar to 6D6) to communication style 1851's, dual 42 horizontal output tubes to a single 807 along with a change in the horizontal output transformer, seven knob control panel to three knob (they moved the position and size controls inside the set). these sets were built very modular so that sections could be easily updated or replaced.

From R.R.Beal's article in the Journal of the Society of Motion Picture Engineers July 1936: "These receivers have been used to produce two sizes of pictures. For the first few months of the tests, the picture size was 5 1/4 by 7 1/2 inches. At the present time most of the receivers have Kinescopes that produce pictures 7 1/2 by 10 inches in size. Fig. 18 shows a 9-inch Kinescope that produces a 5 1/4 by 7 1/2-inch picture. A Kinescope of 12 1/2 inches in diameter is required to produce a 7 1/2 by 10 inch picture."

The first formal display of these sets to the general press took place on November 6 1936. The sets used in this demonstration were all of the newer 12" variety. This was done both because the 12" crt and 441i format represented the state of the art at the time for RCA, but also because the earlier 9" versions used the 1800 crt which produced a yellowish-green image which had been noted as less than optimal for use for the general public.

Darryl

[zenith2134]

I wish the analog HD system being developed in the 80s caught on in USA

[wa2ise]

Originally Posted by Penthode

In 1937, RCA and other experimental broadcasters moved up to 441-lines, AM sound. This was the RMA (Radio Manufacturer’s Association) standard at the time.

I can imagine how AM sound would have sounded with the intercarrier system developed later on.

You could do a reasonable simulation of 441i using a DTV converter box set to letterbox mode, and a TV set, and assuming that there's enough oomph in the vertical jack up the height to make the black bars at top and bottom disappear to vertically fill the screen with picture, and increase the horizontal width to make the picture look reasonably correct. Or mask off the CRT to shrink the screen size. You could (assuming you have the woodworking skills for furnature quality work) build a prewar style console cabinet (or convert a common radio console cabinet that's missing its radio chassis, herisey!), and insert a decent B&W solid state TV or video monitor set inside, masked off to have it show approz 441i lines. And build a tube audio amp to drive a large speaker mounted below.

[Penthode]

It is interesting you should mention intercarrier. All the 405 line receivers were split sound as intercarrier would not work for AM sound.

In an American set with FM split sound design, one would have to carefully tune to optimize the sound. Intercarrier sets would maintain the sound quality over a wider range of fine tuning.

Later, fully superhet UK 405 sets with channel selectors had wide bandwidth sound IF channels. The VHF reduced noise and combined with the wide bandwidth ensured that the sound quality remained high over a wider range of fine tuning than American split sound sets.

Further, the wide bandwidth provided high fidelity sound. In other words, AM negated the need for intercarrier sound.

Note that the immediate prewar UK TV sets were almost all TRF video receivers. There was only one channel and so from the antenna through to the detector, the amplifier would be generally stagger tuned to the incoming 45 MHz video carrier with double sideband AM. The audio was generally tuned from the built-in separate AM superhet radio with a band covering the 41.5MHz AM TV audio.

Perhaps the later adoption of FM sound in 1941 for American TV was a compromise to the E H Armstrong interests?

[earlyfilm]

Originally Posted by Penthode

Perhaps the later adoption of FM sound in 1941 for American TV was a compromise to the E H Armstrong interests?

I'd guess that RCA and Armstrong were too busy suing each other to do that.

FM can live within the NTSC video sideband, but AM cannot as it would pick up the video highs. The AM sound carrier would have to be located outside of the video sideband and that would require a wider band width.

This is like phase modulation used in NTSC color can live within the video sideband and (with a little luck) not be messed up by the AM video.

James.

[Penthode]

I'd guess that RCA and Armstrong were too busy suing each other to do that.

Actually the acrimony was just beginning. The law suits came later.

I am only surmising but it does seem credible that RCA traded acceptance of FM for TV in trade for frequency band allocations (42MHz to 50MHz) for FM Radio broadcast in 1941. Remember, early 40's technology dictated lower frequencies and RCA and the NTSC gave away a chunk of bandwidth which could have been used for an extra TV channel.

FM can live within the NTSC video sideband, but AM cannot as it would pick up the video highs. The AM sound carrier would have to be located outside of the video sideband and that would require a wider band width.This is like phase modulation used in NTSC color can live within the video sideband and (with a little luck) not be messed up by the AM video.

I do not believe this to be so. The BBC experimented with NTSC color in the fifties and sixties for the 405 line system with no change in channel or audio parameters. The audio carrier is located entirely separate from the video in both 525 line with FM sound and 405 with AM sound. Adding color is a matter of using suppressed carrier quadrature modulation and frequency interleaving in the video band only.

[Penthode]

Originally Posted by tubesrule

Penthode,
I actually provided much of the info for that article While the RMA didn't officially adopt the 441 standard until 1937, RCA was using it exclusively from mid 1936 on.

Thanks for the information. I thought the date was September 1938 for adoption of the 441 line standard by the RMA and even then Philco broke ranks. The feeling was that 441 was going to become rapidly obsolete. This led to the final settlement on 525 in 1941.

It is amazing to consider the longevity of 525 considering the changes over the past 70 years. It certainly proved to be the correct choice as it is still strong. 625 I believe only began in Germany in 1952 and French 441 and 819 along with 405 gave way to 625 in the end.

The remarkable thing about 405 is that it emerged from EMI so early in May 1934, was the UK industry standard from February 1937 and remained so until the Pilkington Committee recommended the introduction of 625 lines for the second BBC channel in 1964. Nevertheless, 405 continued until January 3 1985.

[ppppenguin]

There were substantial patent cross licensing arrangments between M-EMI and RCA at the time. It is still a matter of research as to how much M-EMI's work owes to RCA and vice versa. The terms of the pre-war competition in the UK stipulated that HD meant a minimum of 240 lines. Baird offered exactly this while M-EMI worked on 243 lines interlaced. It has often been stated that the move from 243 to 405 was a leap of faith. However it has been pointed out by Paul Marshall that 405/50i and 343/60i are essentially equivalent in terms of required bandwidth. In modern parlance the pixel clock rate would have been almost equal. (The 525/60 and 625/50 standards are likewise equivalent)

After WW2 the 405 system was continued. The reasons for this are debated and I'm sure there is further research to be done. In my opinion 405 was retained because we wanted to restart the TV service ASAP and there was no other affordable option for a country that was victorious but nearly bankrupt.

The 405 services were duplicated on 625 starting in 1969 so the changeover period was inordinately long, lasting until 1985. It would have been less expensive to shut down the 405 service a few years earlier and compensate the owners of 405 only sets. Even allowing for spurious claims (imagine somebody handing over a long defunct 1950s set to get the compensation) this could have saved a lot of money, though possibly at the expense of destroying a lot of older sets.

[Penthode]

You could do a reasonable simulation of 441i using a DTV converter box set to letterbox mode, and a TV set, and assuming that there's enough oomph in the vertical jack up the height to make the black bars at top and bottom disappear to vertically fill the screen with picture, and increase the horizontal width to make the picture look reasonably correct.

I thought about this and you have to be careful here. Picture resolution is dependent on a number of factors.

An important factor which is being overlooked is the Kell Factor. Kell attempted to define the required channel bandwidth (affecting horizontal resolution) against the number of lines. He asserted that because of the gaps between the lines, vertical resolution is lost which means that the horizontal resolution hence video channel bandwidth can be reduced. The Kell factor is the ratio or vertical resolution against horizontal resolution and its maximum value is unity or "1".

When vestigial sideband transmission was introduced for 441 television in the late '30's, the same 6 MHz channel containing the same 4.2 MHz video channel was used for the later 525 line system. The only thing that changed was the reduction of the Kell Factor to about 0.6. This means that the horizontal resolution of 525 video was less than 441 line video to compensate for the increase in the number of lines. Even in those days, it was felt that artificial picture "sharpness" could help compensate for the loss of resolution with the switch from 441 to 525. And the the horizontal resolution of 525 was further reduced with the introduction of color. The Kell factor with color broadcasts on US sets made after 1953, color and monochrome was now less tthan 0.5. However 405 line had a Kell factor of unity hence had considerably higher than expected horizontal resolution.

This may explain why when CBS engineers went to England in the late 50's and examined 405 line pictures first hand, they exclaimed that in many instances the 405 line pictures looked sharper than 525 line pictures.

[newhallone]

Ok everyone...restrain yourselves! hahaha Welcome Jessica.

[ppppenguin]

Originally Posted by Penthode

An important factor which is being overlooked is the Kell Factor.

The Kell factor is largely an empirical fudge. Nothing wrong with that because it reflects how we actually perceive H and V resolution. I don't have the evidence to prove it but I strongly suspect that the work that Kell did applies only to displays with an approximately gaussian spot shape. This is more or less true for CRTs and totally wrong for LCDs. Not sure what happens to Kell if you use spot wobble on a CRT.

It was widely observed in the UK that 625 pictures seemed less sharp than 405. This may have been due to relatively poor HF response in some sets of the time. It may also be a psycho-optical effect of a visible line structure being perceived as sharper.