Wrong UK 625 line TV system?

[colly0410]

A friend of mine (who's 87) used to be a TV engineer & worked all over the world. He told me we (the UK) could have got a perfectly good 625 line system on the cheap by using a modified Argentinian system N.

Jacks system (that's his name) would use positive video modulation up to about 4 megs & AM sound at 4.5 megs, almost the same as system N. The advantages of this system (according to Jack) is that 405 line broadcast equipment could be adapted to it, duel standard TV's would be cheaper as no positive/negative or AM/FM gubbins needed, the lower bandwidth channels could be shoehorned into band 3 so saving cost of UHF transmitters & aerials until extra channels needed. He has seen system N pictures in Argentina & claims they're as good as our system I. He also says Australia & New Zealand should have gone for system N instead of system B. What do others think?

[Cathovisor]

In other words, all the faults of system A with added colour. So, white spotting with interference, susceptible to atmospheric disturbance, soft pictures on larger tubes, and no interchange with Europe. The only advantage would have been a closed market for TV manufacturers.

[Michael Watterson]

PAL-N would have been pointless.
Very little saving to consumer and poor above 17" size screen. A 26" screen only for very large rooms.

There is good reason few adopted it.

You'd still want UHF for more than 2 channels.

[colly0410]

Interesting comments.... Jack loves his modified system N, could have saved millions he claims. He swears that you wouldn't be able to tell the difference between system N & I using normal TV's in a side by side test, I've never seen system N but should imagine it's like a slightly miss-tuned system I. I have seen system M (525 lines NTSC) when I was in America & Bahamas & was surprised at how good their pictures looked even with 100 less lines..

[Michael Watterson]

Perhaps.

But I doubt it would have saved much.

[Mrhoover]

Argentina,Paraguay and Uruguay all use system N..This was because the local mains frequency is 50Hz and not 60.

I seem to recall mention somewhere that Cuba did Secam M tests,just to get away from US NTSC but there were problems with the Secam FM colour difference carriers interfering with the 4.5 MHz audio and so it was abandoned.

Hugh

[Mrhoover]

When these countries started TV in the 1950's system M (US etc standard) monochrome TV's would have been easily imported from say neighbouring Brazil or assembled locally...Apart from the input voltage maybe being 220 instead of 110 a system M mono valve set would immediately work on system N with basically a tweak to the vertical hold & height adjustment.Importing system B sets from Europe would have been more expensive no doubt.

No thought would have been given to colour in the early 50's but the problem would have reared its head 20 years later (colour was introduced in Brazil in 1972 so presumably similar in Argentina etc).

[Mrhoover]

I was in Brazil in 1974 and colour TV set marketing etc really pushed the PAL aspect as being better than
NTSC.

It's possible one reason was to protect the local large TV market from imports.

[Michael Watterson]

They speak Portuguese and the the USA and rest of South America does Spanish maybe is something too.

[Synchrodyne]

As I understand it, the System N parameters were developed in Japan during its initial debate about what TV system to use. The choice was apparently between NTSC (System M) and CCIR/Gerber (System B), and to some extent the debate was characterized as one between 6 MHz and 7 MHz channels. Thus the CCIR/Gerber proponents demonstrated that the 625/50 system could be fitted into a 6 MHz channel if required. If that indeed were the case, then reconciliation of the timing might be an issue, as Argentina started broadcasting with System N in 1951, whereas Japan did not start (with System M) until 1953. So it might have been the other way around, with the early System N work done for Argentina, where the requirement was for a 50 Hz system that fitted the standard American 6 MHz channelling pattern.

The CCIR/Gerber system was already something of a compromise. The Russians had developed the 625-line system for an 8 MHz channel with 6 MHz video bandwidth, the latter allegedly to provide the same horizontal definition as 16 mm film. Otherwise they had generally followed NTSC (I) precepts. It would appear that an 8 MHz channel was more than most Western European countries were prepared to accept, particularly knowing that the USA had chosen a 6 MHz channel. Philips had developed its 567-line system as the best-fit for a 50 Hz system in a 6 MHz channel, but that was going nowhere. Thus came about the Gerber compromise, 625/50 in a 7 MHz channel with 5 MHz video bandwidth. One assumes that 5 MHz video bandwidth was considered to be as low as it was prudent to go with the 625/50 system. As such it provided horizontal definition that was only barely greater than that provided by the British 405 line system with 3 MHz video bandwidth, and about the same as that of the French 441 line system with 3.5 MHz video bandwidth.

In the UK, the BBC started its experimental UHF 625-line transmissions in 1958 May using the System B parameters (as recorded in Wireless World 1958 May). One may surmise that soon thereafter it must have become apparent that Europe was heading towards adopting a standard 8 MHz channel, with standardized vision carrier frequencies, for UHF transmissions. In turn this prompted the BBC to look at the optimum use of the 8 MHz channel. The obvious choice might have been simply to follow the Russian (OIRT) precedent, but what came out of this study were the System I parameters, as embodied in the 1960 TAC report. Evidently the best overall performance increment, as compared with System B, was the combination of a bigger vestigial sideband with a video bandwidth of 5.5 MHz. The latter also allowed NTSC (and derivative) colour subcarrier sidebands to be accommodated with asymmetry.

One may question the wisdom of introducing yet another 625-line transmission system, rather than picking the best of those already available. Perhaps a defence here could have been that firstly, the underlying system was the same as used across Europe, thus was not inimical to easy program exchange, and secondly, that as future British receivers would necessarily be dual-standard 405/625, they were unique anyway, so that differences in 625-line transmission parameters would not be of major consequence.

Whether the modulation question was seriously considered in the run-up to System I is unknown. Perhaps the BBC and the TAC simply followed the established majority choice of negative/FM. Or perhaps there was concern about oscillator drift at UHF, which would have pointed to negative/FM because it allowed intercarrier sound. But given that System I was a new transmission system, there was no fundamental reason why positive/AM could not have been chosen for it, just as the French did with System L.

NTSC (I) in 1941 chose negative vision modulation primarily because it on its face it allowed simpler black-level referenced (i.e. sync tip) agc. Intercarrier sound was not a factor because it was not invented until the late 1940s. In practice simple agc did not work very well at VHF because of impulsive interference, so that line-gated (and preferably also noise-gated) agc was required. This reduced the advantage of negative over positive, although negative still required less agc system gain (exacerbated by the fact that whereas sharp cutoff valves were often used in negative/FM receivers, there was a greater tendency to use remote cutoff types in positive/AM receivers).

Intercarrier sound as typically executed was in and of itself a compromise, and where high quality sound reception was required, particularly multi-channel, it was necessary to revert to split sound, or use the quasi-split technique, or “true intercarrier” as used in some BBC receivers. Once economic PLL synchronous vision demodulation techniques became available, intercarrier sound could be “clean” enough that it could also be used for positive/AM systems (e.g. look up the Motorola MC44302A IC).

Allegedly the French chose positive/AM for System L to simplify dual-standard receiver design. As part of this there was also a common sound IF (39.2 MHz) for Systems E and L. This required System L UHF receivers to be oscillator-low. (System E receivers were a mix of oscillator-low and oscillator-high.) When System L was extended to VHF, whilst oscillator-low was possible for Band III, it was not for Band I, so that the Band I channels were inverted (as System L’), with vision carrier high, so that oscillator-high could be used with them. As anyone who has been involved with corporate complexity reduction exercises will likely know, complexities can be moved around and hidden from immediate sight, but they usually turn up somewhere else in another form. It has been said that the French chose positive/AM as a non-tariff import barrier, but the counter there is that for the Outré-Mer territories, where there had been no use of System E and therefore no need for dual-standard receivers, the French chose System K’, which was negative/FM. (Although Algeria had started out with System E, including bilingual sound, but somewhere along the line that was abandoned.)

For Systems K’ and L, the French chose a video bandwidth of 6 MHz (as in the Russian System D/K) coupled with a 1.25 MHz vestigial sideband. By the usual reckoning, this combination would have required a 8.5 MHz channel, but was used with 8 MHz channels. There was a System E precedent for this kind of “squeezing”, though. Originally System E was designed for a 14 MHz channel, as evidenced by the original channel that became F8A (174 to 188 MHz). But the definitive tête-bêche VHF channelling system actually spaced the vision carriers at 13.15 MHz.

Certainly dual- and multi-standard receivers that covered both negative/FM and positive/AM systems involved some complexities, but they were hardly new when they became the norm in the UK. They had been used in Belgium since the beginning of TV there in 1953, and also in parts of France (around Strasbourg). Not only that, but some of those receivers covered four standards and managed to have line-gated black-level agc on both negative and positive systems, a precedent that the UK setmakers seemed unwilling to follow. The BREMA standard IFs might have encouraged some compromise on 625, as well. The simplest approach was probably a basic IF curve (before 405 traps) that was of double-Nyquist form, with -6 dB points at 34.65 and 39.5 MHz. This would have limited the 625 video bandwidth to 4.85 MHz maximum at -6dB, maybe around 4.2 MHz at -3 dB, well below the 5.5 MHz transmitted.

Now turning to Australia, System B, as used in most of Western Europe, was the logical choice in the mid-1950s, at which time System I was still in the future. There was no good reason for it to use a rather compromised Latin American version of the 625-line system. By so doing it may have gained one extra VHF channel, but it solved the channel count issue by using Band II, and later on some of the space just below Band III. By the time it needed more channels, UHF was a reality, and uniquely it adopted 7 MHz channelling in bands IV and V.

New Zealand adopted System B on the cusp of availability of System I, but it would have been a brave move for a small country, and one of the world’s most remote, to have been the first to use that system. The available Band I and Band III channels were initially sufficient for a single program, and in fact the seven first program main transmitters were all accommodated in Band I. Not only that but some of the second program main transmitters were also fitted into Band I. Eventually a third program was accommodated in Band III to which two channels were added at the top end, but for this exercise co-siting throughout had to be abandoned. Bear in mind too that in the NZ case, reasonable population coverage does not require anything like full geographical coverage, unlike the UK case where the two parameters are closer together. Band I would still have been limited to three channels whether the latter were 6 or 7 MHz. NZ did adopt European 8 MHz channelling for UHF; that may have been a plus when digital came along.

Now back to the UK case, adoption of a positive/AM version of System N, and taking advantage of its lower bandwidth, would have meant using 6 MHz channels at UHF rather than the European 8 MHz standard. Whilst such unilateral action was not outruled, it would have been somewhat counter-current at the time. System N, with its 4.2 MHz video bandwidth, would have meant that at best, 625-line pictures would have had 90% of the horizontal definition of 405-line pictures, and in practice probably not as good as this. One may see the case (just) for System N where the use of 6 MHz channelling was seen as an imperative, but not where 625-lines was to be introduced as a step-improvement over the existing 405-line system. Back in those days broadcasters usually adopted future-looking standards that might initially be a stretch for deluxe consumer receivers, but they did not pare back to match the assumed “modal” receiver or even further to match the lowest common denominator. At least System I was a potential improvement over System A, both in terms of line-count and horizontal definition.

Agreed that Band III in the UK was probably not fully utilized, whereas Band I was likely oversubscribed. But whether, even with 5 MHz 405-line channels, there was enough room in Band III to provide national coverage for two programs is debatable. Some use of UHF likely would have been required. And mixing 5 and 6 MHz channels in Band III might have brought its own complications and set of “taboos”, perhaps even more so once colour subcarriers were in the picture, so as to speak. Moving to a national chain of UHF transmitters at the time was logical and allowed for future expansion, and consistent with what was happening elsewhere. Starting a third programme by shoehorning it into Band III with 6 MHz channels would have delayed the need to use UHF by only a handful of years, but would have saddled the country with a compromised transmission standard.

One thinks of the Belgian System F case, where the French 819-line system was shoehorned into a 7 MHz CCIR channel. Perhaps justified in 1953, before standards conversion was an everyday event, where Wallonia needed to be able to access French origin programs, and where transmitters needed to be able to relay both 625-line (System C) or 819-line (System F) programs according to whatever was incoming. But as far as I know the System F transmitters were all on System C by the end of the 1960s, and Belgium had adopted System H for its UHF transmissions.

So one might say that whatever the short-term advantages that might be ascribed to System N, its basic parameters were just too much of a compromise.

Negative/FM or positive/AM was really a separate question, and the UK would appear to have had a free choice in this aspect.

PAL-N colour did not arrive until 1978 as far as I know, although no doubt it would have been developed earlier had it been required. An interesting question, had a positive/AM version of System N been adopted, is whether the advent of colour would have required an offset for the sound carrier as it did in the 405-line NTSC case. That would have been another complication, perhaps the more so if afc had become widespread in domestic receivers when colour arrived, with some receivers using vision channel-derived afc and others sound-channel derived. And positive/AM at UHF probably did required good afc.

Finally Donald Fink, who was an eminent member of NTSC (I) and the person who chose the 525-line number, is on record as saying that it would have been better to have adopted an 8 MHz channel than the 6 MHz actually used. (Although 6 MHz was a "given" that NTSC (I) was not free to change.)

In summation, I think that the foregoing amounts to a “no” vote on the modified System N proposition. It definitely would have been the wrong 625-line system.

Cheers,

Steve

[Amrad]

Hello Steve,

In the UK, the BBC started its experimental UHF 625-line transmissions in 1958 May using the System B parameters (as recorded in Wireless World 1958 May).

Are you sure about that? The BBC carried out experimental colour television transmissions at that time on 405/50 using NTSC. I don't recall any 625/50 transmissions then.

Regards,

Dave.

[Mrhoover]

Yes,625 tests were also mentioned in Practical Television Magazine...They ran for about a year or less
though according to the PT report with + 6.5 MHz audio on approximately Ch 44(European Uhf channels hadn't been assigned then) from Crystal Palace.

After the video tests were completed,a lot of propagation tests were done,no video being used.

Some info can be found in the old old BBC online R&D reports from that era.

[Mrhoover]

They speak Portuguese and the the USA and rest of South America does Spanish maybe is something too.

The Brazilians were/are quite protective of their national industries and I'm sure didn't want a lot of NTSC imported receivers on the market.

I remember seeing newspaper ads for NTSC set "conversions" to PAL when staying there.

[Synchrodyne]

Here is the WW item re the BBC experimental 625-line broadcasts.

Cheers,

Steve

[Mrhoover]

Here are a couple of BBC reports on the tests

http://www.bbc.co.uk/rd/publications/rdreport_1959_25

http://www.bbc.co.uk/rd/publications/rdreport_1958_14

It's an interesting site with lots of information

http://www.bbc.co.uk/rd/search?Type=Pub ... ecade=1950

[Michael Watterson]

Here is the WW item re the BBC experimental 625-line broadcasts.

Cheers,

Steve

What year was that?

[Synchrodyne]

1958. That was from the 1958 May issue of WW.

Cheers,

Steve

[colly0410]

Very interesting comments, I love to read how & why things evolved differently in other Countries. I was very surprised at how good the M-NTSC pictures in USA/Bahamas looked, SWMBO has epilepsy & is affected by 50 hz flicker on CRT's over here & can't get too near to them. In USA/Bahamas she didn't seem to be bothered by the 60 hz flicker although she didn't push it by going near TV's. Our 32 inchs LCD TV doesn't affect her at all so I don't have to worry about her so much, worth the 400 quid I paid for it..

[Michael Watterson]

LCD sets inherently don't flicker whatever the refresh rate. The LCD cells hold the last brightness until they are refreshed.

Edited to add: Except when the programme material contains flicker.

Also 8ms etc claims are nonsense. At higher refresh rates if you do nothing special any movement between frames results in ghosts. If you do some special stuff you get artefacts and a little less ghosting.

But they don't either have a LONG enough "ghost" time (persistence) to do visual interlacing like a CRT phosphor does. So even if it's a "natively interlaced*" display then source video needs to be converted to progressive.

(* Some older 1600 x 1200 60Hz "progressive" LCDs have internal "native interlace" of some strange nature. If you put a 100% Black and white horizontal alternate line image, i.e. 600 x one pixel white lines and black lines the lines flicker madly. Exactly what this is I'm not sure as a single horizontal 1 pixel line (say a frame or box has no obvious flicker )

Right Click and View Image too.

Click on Test image to see if your LCD does strange stuff internally?

refreshInterlace.png (3.53 KiB) Viewed 6254 times

Boxes

refreshtest2.png (7.11 KiB) Viewed 6254 times

[Michael Watterson]

BTW,

Thanks Steve for some very interesting posts. It does seem a shame there wasn't just one 625 based on the Russian 8MHz version.

It's a bit like today many DTT are using only 544 rather 720 for DTT* and really WS SD surely should have been a a universal 1024 x 576 @ 48 fps progressive. We now have far more digital resolutions and digital transmission formats for Terrestrial (it's reasonable for Satellite, Cable and Terrestrial to use different modulation schemes) that there ever was for analogue.

(* No wonder that even upscalled SD by broadcaster on the 1440 x 1080 Terrestrial HD looks better than SD. Also if there is such a surplus of spectrum such that 790 to 862 was sold off and now they want to sell 690 to 790 why is DTT only 544 or 1440 and not 720 & 1920 and not 50p rather than 25i?)

[Synchrodyne]

Well, there were certainly no shortage of 625/50 transmission variants. In approximate order of first use, they appear to have been D/K, B/G, N, C, I, L, H, K’. Plus there were variations within some of the letter categories as well.

Possibly H should be grouped with B/G, as – neglecting any group delay issues - it could be resolved on B/G receivers, and likewise K’ could be resolved on D/K receivers.

As already said, N was 625/50 squeezed into an M channel with some M-like parameters as well. Its FM sound was ±25 kHz with 75 µs pre-emphasis, and as far as I know it retained 6 (each of pre and post) equalizing pulses (as for System M) when the other extant 625/50 systems were changed to 5. M video bandwidth was originally 4.0 MHz, later increased to 4.2 MHz (I am not sure when, but perhaps with the advent of colour), so N probably went through the same transition.

C was the positive/AM counterpart to B. Unusually for AM sound at the time, which was usaully not pre-mephasized, it had the same 50µs pre-emphasis as was used for System B FM sound. (Later, much later I think, as hoc pre-emphasis curves were used by MF broadcasters in the USA until the NRSC 75 µs curve arrived in the 1990s. And NHK Japan had apparently standardized on 100 µs for MF broadcasting before then.)

Not 625/50, but a fellow-traveller, F was 819/50 squeezed into a CCIR channel, with parameters otherwise similar to those of System C, including pre-emphasized AM sound and the use of equalizing pulses (7 of).

K’ was the negative/FM counterpart of L, which was positive/AM. L AM sound was not pre-emphasized; no doubt keeping it the same as System E AM sound was a requirement.

On the subjective side, my own viewing experience (of a more than passing nature) has been with B/G (New Zealand and Australia), I (UK and South Africa) and M (USA). Also, during my travels I have also seen, in passing, D/K (on visits to China), L and M-PAL.

With regard to M-NTSC, during my time in the USA I seldom saw the hue errors for which M-NTSC was notorious, and generally it wasn’t necessary to re-adjust after channel changing. My initial impression after coming from a 625/50 area was that the pictures were slightly coarser. The freedom from flicker is not something that strikes one at first. But after a time in the USA, the first sight of a TV screen when travelling to a 625/50 area created the impression of excessive flicker, and it took a week or more before one didn’t notice it too much. Then when returning to the USA, the freedom from flicker of 525/60 was more apparent. Between B/G and I, I had the impression that System I pictures were usually a bit better, I’d say somehow “cleaner” overall. Maybe that was simply because I was “seeing” what I expected to see; hard to say, although I noticed the same difference between Singapore (B) and Hong Kong (I) even when I wasn’t looking for it; more a case of – this looks better than the last one – oh, that’s right, it’s System I here. My first good look at Chinese TV back in 1987 was a bit similar; nice clean picture – then remembering that it was D/K there. When one cannot understand the soundtrack, maybe one looks more closely at the picture, as well. System L was also in the same league as far as I could tell. I can recall back in 1996 staring for quite some time at a B&O TV in the hotel in Paris trying to find something about the picture that would confirm that it was positive not negative modulation, but without any luck. At the time I was living in Cape Town, so accustomed to good 625/50. M-PAL looked just like M-NTSC, but I had only one week in Rio, and not much time for TV watching, so I shouldn’t say that it was a definitive comparison.

Cheers,

Steve

[colly0410]

I remember being in a Boulogne TV shop in 1986, they had telly's showing BBC & French channels side by side, I tried to see if I could spot any difference between PAL-I & SECAM-L, they both looked the same = good.

The worst picture I've ever seen was in Majorca in 1978, the mains was out of sync with the incoming TVE 1, there was wavy S distortion traveling up the screen, it would slow down & stop & it would look OK for a few seconds, it would then start going down for a while before repeating, it was very annoying. I presume Majorca has a separate grid with poor frequency response.

On the subject of flicker, I've got used to our LCD telly, I went to a friends house the other day who had a CRT telly & it flickered away, I think my brain must have changed. We must have got that used to flicker that our brains must have developed a flicker filter, but my flicker filter's shut down through non use.

[Mrhoover]

I remember being in a Boulogne TV shop in 1986, they had telly's showing BBC & French channels side by side, I tried to see if I could spot any difference between PAL-I & SECAM-L, they both looked the same = good.

Years ago I used to have off air French TV in Sussex from Boulogne.I found French TV always had what's best described as a "pinky blue" look,flesh tones rather pink and backgrounds rather blue.compared with UK TV.

You can see the effect on old youtube French TV clips

Strangely now they've all gone digital the same colouring is still there to my eyes...Maybe it's something they got used to and are reluctant to change...It's not unpleasant and one soon gets used to it.

Hugh

[Crusty Burke]

One point I'd like to make to Steve's excellent and comprehensive post was that Ireland was actually the first country to introduce System I in the summer of 1962, when Telefis Eireann introduced 625 line broadcasts on Channel H from Kippure, following the opening of T.E. on System A Channel 7 from Kippure in 1961. This means that the Irish did have System I transmissions nearly two years before BBC-2 opened on UHF in London.

System I with full 8 MHz bandwidth channels on Bands I & III did co-exist with System A transmissions from T.E. and BBC/UTV/Teledu Cymru. One effect was that UTV transmissions on Channel 9 suffered from a buzz on the sound in Dublin, possibly due to Channels F & H being active. Perhaps Michael can elaborate. Ireland stayed with System I on VHF (and UHF after 1978) until analogue shutdown nearly two years ago.

Richard

[Michael Watterson]

A bit more than two years? 2.25 years

The PUBLIC launch of RTE was 31st Dec 1961.

BBC 2 PUBLIC launch was 20th & 12st April 1964.

But BBC had transmitted 625 in 1958. I don't know when RTE first 625 test transmission was.

Curiously:

1 July 1967, BBC2 became the first channel in Europe to begin regular broadcasts in colour, using the PAL system

Because SECAM had been tested I thought before PAL invented and also Public 626 Line started somewhere in 1948 I think, with tests in 1945/1946 (not sure if on a transmitter or Lab, I presume Russian?).

Analogue HDTV was used in 1940s for closed circuit German Rocket Ranges.