Radofin 1292 advanced programmable video system

[Ricardo Cividanes da Silva]

Hi everyone.

  Guys, I'm researching for my book and I ran into this Radofin 1292 system. He is a very obscure video game system and would like to know if anyone has more detailed information about this system.

  First, I can't find the correct processor clock details. Everyone says that the Signetics 2650A-1 runs at 4.43 MHz, which is impossible since, because the specifications of this processor indicate that it reaches a maximum clock of 2 MHz. Another point is the amount of memory. Everybody all speak in 32. But, 32 Kilo bits or 32 Kilo Bytes?

  Most likely, by the time and the cost of memory at 70's, it would be 32 Kilo bits which would give about 4 Kilo Bytes.

  I am really grateful if someone has the right information.

 

Thanks!

[Rolo]

Hi,

 

Not so obscure, I think. Quite common, but it is true, not so much information available.

I opened up my ACETRONIC MPU-1000 and took some photos.

 

PAL-version: crystal 8.867 MHz, obviously divided.

 

Concerning clock, I read the figure 0,887 MHz, which would be something like a quarter of the NTSC color burst (~3,58 MHz).

Hmmm, PAL color burst is ~4,434 MHz, the crystal oscillates at twice this frequency. If we divide this by 4, we would end up at 1.108 MHz, which would seem plausible, too. Maximum Clock is 1.25 MHz. Maybe some measurement required. Does anybody know?

 

Microprocessor Signetics 2650AN (40 pin plastic):

 

RAM:

There is no real RAM onboard the console, at all. The console is using 37 bytes of scratch pad memory of the PVI-chip (Signetics 2636).

 

 

[CatPix]

I looked high and low and there doesn't seems to be any NTSC version of the system - which make sense since all clues point out for the system to have been designed originally by Philips -  so it make sense that the system use PAL-compatibles values from the get go.

RAM was included on several game carts, with amount of about "1Kb" which is probably bits and not octets, so 256 octets.:

 

http://amigan.1emu.net/igg/#archived

[carlsson]

There should be dozens of previous threads about the VC-4000 and 1292 APVS. Here is a previous one in which tech specs were discussed:

 

[Rolo]

There are only a few cartridges, which contain extra RAM inside the cartridge:

1. CHESS 1           4 kBytes ROM + 1 kByte RAM
2. CHESS 2           6 kBytes ROM + 1 kByte RAM
3. DRAUGHTS       4 kBytes ROM + 1 kByte RAM
4. BACKGAMMON  4 kBytes ROM + 1 kByte RAM
5. Come-Frutas     4 kBytes ROM + 1 kByte RAM
6. Hobby Module   2 kBytes ROM + 2 kBytes RAM

[Ricardo Cividanes da Silva]

35 minutos atrás, Rolo disse:

Hi,

 

Not so obscure, I think. Quite common, but it is true, not so much information available.

I opened up my ACETRONIC MPU-1000 and took some photos.

 

PAL-version: crystal 8.867 MHz, obviously divided.

 

Com relação ao clock, li a figura 0,887 MHz, que seria algo como um quarto do pico de cor NTSC (~ 3,58 MHz).

Hmmm, o burst de cor PAL é de ~ 4.434 MHz, o cristal oscila com o dobro dessa frequência. Se dividirmos isso por 4, acabaríamos em 1,108 MHz, o que também parece plausível. O relógio máximo é 1,25 MHz. Talvez seja necessária alguma medição. Alguém sabe?

 

Microprocessador Signetics 2650AN (40 pinos de plástico):

 

RAM:

Não há RAM real no console. O console está usando 37 bytes de memória de bloco de rascunho do chip PVI (Signetics 2636).

 

 

Yes, that makes sense. The 2650N has a maximum clock speed of 1.25 MHz. Signetics 2636, as I understand it, is a kind of video processor and is clocked at 3.58 MHz. It would make sense for the crystal to be 8,867 MHz.

[Ricardo Cividanes da Silva]

5 minutos atrás, carlsson disse:

Deve haver dezenas de tópicos anteriores sobre o VC-4000 e o 1292 APVS. Aqui está um anterior em que as especificações técnicas foram discutidas:

 

Hi Carlsson. I read this topic, but I still had doubts about the issue of processors and memory. Another thing, the VC4000 system is not a clone of 1292, right? They are, so to speak, cousins, as the project appears to have passed through the sieve of Philips.

[carlsson]

It is debatable which one came first, but software wise they're identical AFAIK. The cartridge pinouts differ, but I don't know if that is enough to differentiate them as "cousins".

[Ricardo Cividanes da Silva]

Many articles see them as the same system, but in my research they are similar systems. And this is because Interton and Radofin designed them at the same time with the help of Philips and, therefore, they are so similar. As we say here in Brazil: Children of the same mother, but of different parents.

[Rolo]

I checked the clock frequency of the 2650 processor:

 

It's 887 kHz.

 

[Ricardo Cividanes da Silva]

1 minute ago, Rolo said:

I checked the clock frequency of the 2650 processor:

 

It's 887 kHz.

 

 

Dude, it's amazing. That was my biggest doubt.

 

So, 1292 system processor has a clock of 887 KHz and the video processor 2636 has a clock of 3.54 MHz.  Now every make sense.

Here in Brazil we never had access to these video games which makes everyone here very curious.

 

Thank you for your help!

[Rolo]

The datasheet of the 2636 writes on page 2:

"PCK  |  input pin  |  Position Clock: Generated by the 2622 USG to synchronize the PVI's internal functions. (3.58 MHz, 227 pulses/line)."

 

There is a PAL USG 2621 and a NTSC USG 2622 by Signetics.

Datasheet:

PAL:

"The USG accepts a single 3.55MHz Input clock and generates various timing outputs including vertical, horizontal and composite blanking, composite sync and color burst flag. Several auxiliary clock outputs are also provided."

 

NTSC:

"The USG accepts a single 3.5796MHz Input clock and generates various timing outputs Including vertical, horizontal and composite blanking, composite sync and color burst flag. Several auxiliary clock outputs ara also provided."

 

8.867239 MHz (crystal) *2/5 = 3.546896 MHz (USG)

There must be a clock divider, probably the strange PE1X-chip, of which I do not find any information.

 

3.546896 MHz / 4 = 0.886724 MHz (CPU)

 

and

 

8.867239 MHz (crystal) / 2 = 4.4336195 MHz (European PAL color subcarrier) 

 

Edited January 11, 2021 by Rolo
additional information

[Ricardo Cividanes da Silva]

Em 11/01/2021 às 10h08, CatPix disse:

Eu olhei para cima e para baixo e não parece haver nenhuma versão NTSC do sistema - o que faz sentido, já que todas as pistas indicam que o sistema foi projetado originalmente pela Philips - então faz sentido que o sistema use compatíveis com PAL valores desde o início.

A RAM foi incluída em vários carrinhos de jogos, com uma quantidade de cerca de "1Kb" que provavelmente são bits e não octetos, então 256 octetos:

 

http://amigan.1emu.net/igg/#archived

Hi CatPix, I saw your others topic about 1292. Do you know if Philips projected this systems (VC4000 and 1292) and selled to Radofin and Interton separately?

[CatPix]

I look for clues often but unfortunately, both companies have been closed - and apparently, Radofin was so unimportant that no one even bothered buying the name. Yes, you can use the name Radofin as you see fit, no one is currently holding the rights for it.

 

Finding clues about who worked there it though. One programmer for the Interton VC 4000 posted here on AtariAge; I'l trying to find the topic back; if you can contact him he might have informations on the company or remember other people from the company.

Contacting Philips is almost certainly a lost cause, they are very silent about their past and I heard they are very uncaring about keeping archives on projects that did not made a reasonnable success.

 

What we can say is that adapters can sometime be found to play games from one system to another, that people have made such adapters, they dumped and compared ROM from Radofin and Interton games and most are 100% identical and work on both systems.

It also seems that the Radofin is probably closer to the original design than the Interton : one big difference between the two systems is that the Radofin have non-centering joysticks, unlike the Interton.

And several games were programmed with non-centered joystick in mind; Bowling for example plays very easily on the Radofin system, but it near impossible on the Interton because positionning the bowling ball is not at the "zero" of the joystick, so placing it and hitting it become a struggle on the Interton (especially with the Interton's very strong springs).

The Interton machine was however certainly more succesfull since several games only appears on the Interton; some of those games bear mention of a programmer in Hong Kong :

 

https://tcrf.net/Shoot_Out_(Interton_VC_4000)

 

It also prove that Interton was still active as of 1981 to promote the system.

[CatPix]

Interestingly, I found a scan of a French magazine from November/December, 1982. The Radofin is annonced as "just arrived on the market". No mention of the anteriority of the system elsewhere, but the writers didn't paid much attention to the system. The late date may just be due to the need to convert the console to SECAM, which, from the console I own, called for a board redesign, the (RGB?) to SECAM conversion being done with a bunch of discrete components unlike PAL one which was made by an IC.

You can see that even as "new" the system was sold cheaper than the aging "Philips" (Videopac/Odissey2) and Atari (2600). Amusingly, the Prestige is a variant of the Arcadia 2001, the "evolved" version of the "Signetic system" with a more powerful graphic chip! (it's amusing that also the "Video System" get two stars and the Prestige only one for "sound", since it's the very same sound chip... but to be fair, the testers most likely expected "more" from the more advanced-looking system and gave a note accordingly)

 

This doesn't help much about finding the earliest possible release date... but we can also tell that at least the 1292 was still sold as of Christmas 1982.

Edited January 12, 2021 by CatPix

[carlsson]

The old 1292 FAQ has a lot of apocryphal dates:

• Interton VC-4000 designed in 1974, but not launched until 1978
• Radofin 1292 APVS launched in 1976 (which many sources still recite)

What we know is that the Signetics 2650 CPU was not launched until July 1975. They begun work on the 2650A in March 1976 and finally manufactured it in 1977. This is an important date because both the VC-4000 and 1292 systems use the 2650A variant, not the older 2650 variant.

 

The ideas of ROM based consoles were already going on at Fairchild, RCA and Atari so the principle is not unthinkable but Interton must've been sketching on a different architecture if they begun designing the 4000 already in 1974.

 

Radofin was formed in July 1974 and released their Pong-style Tele-Sports in 1976. It uses a Mostek MPS-7600-001 (same as Commodore 3000H from 1977). I doubt a such young company would have come up with two systems right away, in particular as we see from above that the 2650A was not even available in 1976.

 

While it doesn't prove who was first out of Interton and Radofin, we know that 1978-79 is far more likely dating than 1974-76.

[CatPix]

I strongly suspect that people mentionning 1974 for Interton are mixing up the VC 4000 and the Video-2000; the Video 2000 is a discrete-based system using carts containing either jumpers or logic gates (If you're thinking that it sounds like what Ralph Baer wanted to do with the Odyssey, you'd be right, it's exactly the same kind of tech. Given that Germany saw the release of the "ITT Odyssee" in 1973, it's not even hard to argue Interton at least knew the design would work).

 

As for the confusion, I've seen the Video 2000 name used for the VC-4000, and vice versa. The confusion probably came from the earliest ages of Internet.

The Video-2000 's release date appears to be 1975. 1975 or 1976, both dates appears reasonnable for such a "simple" system.

 

Also, good information on the chip's specific release date. Now we have a strong argument to date those systems later than the mentionned dates.

 

On Radofin's side, their Pong system used the same general design, and a similar name, which might also cause confusion :

 

Edited January 12, 2021 by CatPix

[carlsson]

It also struck me we have that other reference, the Swedish TV program Tekniskt Magasin broadcast in April 1979 where they had visited the Philips research centre in Eindhoven and where we concluded they were playing on an Interton VC-4000. Sure that is a fairly late date, but the fact that Philips would host an Interton console, not a Radofin or one of the other brands in the family tells me something... not sure exactly what. Given the lead times to film something like that, I would imagine it was filmed in the beginning of 1979.

 

(sure, Philips in the Netherlands could have stronger bonds with Interton in Germany than with Radofin in the UK)

 

We should also not forget the Magnavox Odyssey^2 which supposedly was first released in Europe as Philips Videopac G7000 in December 1978. Sure it probably was developed completely separately from what the Signetics division was offering and Philips was and still is such a huge company that they could have several branches into video games at the same time, but as I noted before, the video from Eindhoven didn't showcase this G7000 at all despite Internet sources claim it should have been brand new at the time of filming.

Edited January 12, 2021 by carlsson

[CatPix]

Well from the game listing, it seems that Interton was the most interested in the system, so I can't see as impossible that Interton approached Philips to get the system upgraded with a newer video chip. (when was the 2637 designed?)

As I said, Bowling and another game common on both systems are clearly designed with the non-centering joysticks of the Radofin. Of course, it's not far-fetched to imagine that prototype of the Interton also had non-centered sticks and the decision to make them auto-centered came too late to reprogram the few games not optimized for it.

[carlsson]

I looked it up a month ago and concluded that the 2637 UVI was available at least by January 1982 when it was renumbered (!) so it must have been out before. As I noted then, some sources claim that the Soundic/Hanimex MPT-03 which is an Arcadia 2001 "clone" also was out by 1979, thus at the same time as everything points the Radofin 1292 but again I doubt that dating. Excatly how that MPT numbering works is a mystery given that 02 is a RCA Studio III, 03 is an Arcadia 2001, 04 and 05 both appear to be 1292 systems.

 

https://atariage.com/forums/topic/314534-arcadia-2001-fan-appreciation-thread/?do=findComment&comment=4699167

 

[carlsson]

The Signetics 2650 CPU manual from 1975 mentions a number of peripherals, but the 2636 and 2637 are absent from the document, which further delays the video game systems. I mean if neither the processor or the graphics chip were on the market, exactly what would you develop in terms of hardware and software and if you sketched on something else, how much would you have to rework when you finally had chosen the hardware platform?

 

The Signetics integrated circuits catalog from 1978-79 mentions the 2650 but also lacks the 2636 PVI and 2637 UVI chips. Sure the 2636 could have been ready by then but not included in the catalog of parts available to order.

 

The Radofin 1292 manual says it is printed in 1979 so there goes... the 1392 manual btw is printed in 1980.

Edited January 12, 2021 by carlsson

[Ricardo Cividanes da Silva]

I think with the information you gave me, I think it’s possible to put together a story with a lot of speculation:

  - Philps bought Signetics in 1975

  - To help in the sale of the processor family Philips Europe decided to make a project using the processor family from Signetics.

  - Radofin and Interton entered this project at some point, but it seems that Philips treated them as separate customers helping them with details of the project. Taking both manufacturers to launch their consoles in 1978 and 1979.

  - Other manufacturers made clones of both the Interton set and the Radofin set. Philips must find everything very interesting, since everyone bought chips from Signetics.

  - At one point a Hong Kong company, Universal Appliances Limited (UAL), took a more current version of Signetics-based hardware and licensed it to Emerson and more.

 

I don't know, but I'm seeing this puzzle take a form.

[Ricardo Cividanes da Silva]

 

About what you talked about Odyssey2 / Videopac I raised what happened. Philips had bought Magnavox in the USA, but it does not seem to have been very interested in the Odyssey. Magnavox had a project for a new video game, but he was almost dead when the people at Magnavox decided to take everything to Ralph Baer to check it out. In reality it was Baer who saved the Odyssey2 project by consulting the project and leading to its launch in 1980. But the Magnavox project was all based on Intel chips. As far as I researched Philips, it hardly interfered in the American project, but it found it so interesting that it took it to Europe and here to Brazil, where it was a huge success. The European Philips project, it seems, aimed at having the base project for video games and computers with Signetics chips to boost Signetics sales. That is why it makes sense that the basic design of the VC 4000 and 1292 was born at Philips and both Interton and Radofin took it on with little or no change (apparently they changed the pinout of the cartridges) in the hardware and none in the software . It's speculation, but due to the difficulty of getting information, I think it makes more sense.

[CatPix]

Going off-topic a bit, but at least for Europe, the success of the Videopac is easily explained by several factors :

 

- lack of serious competition : the Videopac was released in Europe in 1978/79. The main opponent, the Atari 2600, was released in 1980, the Intellivision in 1981 or 82.

- local manufacturing : the Videopac systems (maybe not all of them. It seems early ones were shipped from the US) were manufactured in France and in the Netherland; as such they didn't suffered from extra taxes, costly shipping, etc.

- better knowledge of the local market and better network. Philips being a major electronics seller in Europe since decades, they had a huge network and a deep knowledge of their markets. It was easy for them to put their name and weight in balance to market the Videopac, as opposed to Atari. (Funnily mirroring how things happened in the USA with Atari and Magnavox)

- "standard" : Videopac were sold under several brands in different countries. Usually those were just Philips brands but at least Brandt (part of Thomson ) sold the Videopac too; it made the Videopac appears as a largely adopted standard of video games. (Most likely, this is the reason why Philips would abandon their in-house designed computers to sell MSX standard computers.)

 

 

We will probably never know why Philips didn't released the Signetics systems under their own name. My best guess is that Philips wanted to cash on the Odyssey name, and they probably figured that they could produce games more quickly for more-know Intel chips than for Signetic chips. The influence of Ralph Baer might have played a role too; also, if the system was ready for production, it was also more clever for Philips to release the Odyssey2 in the USA in 1978 than delaying a potential Signetic game machine back to 1979 or 1980, with Atari already filling the US market, and Mattel intending to step in too.

 

 

[carlsson]

18 minutes ago, CatPix said:

The main opponent, the Atari 2600, was released in 1980

Not entirely true. Last spring the Royal Swedish Library opened their newspaper collections for everyone to search due to Covid-19 and people were bored at home. I took advantage of this and made quite a bit of research, mainly on Commodore and Atari but also a little on Intellivision. What I found out was this:

• January 28, 1977: The Atari Pong etc games have already been on the market for a while due to direct import by Beckman Innovation (many years later agent for Sinclair computers) and resellers are starting to have discount sales. At this point, there didn't seem to be a Swedish agent for Atari though.
• February 27, 1978: Cherryföretagen advertise that they're the Swedish agent for Atari and intend to set up a new division for home electronics and are looking for a product manager for the Atari VCS and similar products. Cherryföretagen are cooperating with Gösta Brännström Consulting.
• November 12, 1978: The Atari Video Computer System is advertised for sale at 1495 SEK including one game, or paid off 12 months x 116 SEK + initial payment 399 SEK (total 1675 SEK).
• November 24, 1978: The low price store Ramsells in Gothenburg advertise the Atari VCS and will have a special demonstration the following weekend.
• December 8, 1978: Cherry advertise about "the most entertaining technology breakthrough of the 70's", i.e. the Atari VCS which was demonstrated with applause (!) at the expo Videomässan in Stockholm.
• December 9, 1978: The first classifieds appear, someone selling their lightly used Atari VCS for 1200 SEK...
• December 14-16, 1978: Another two stores advertise the Atari VCS at 1495 SEK. The price remains unchanged until August 1980 (!!) when a mail order company advertises the VCS at 995 SEK while others still sell it at 1295 SEK.
• August 1981: Bonniers owned board game/toy company Alga have had a few rough years, but have decided to switch to video games and thus bought Cherry Home Electronics and with that acquisition obtained the agenture to Atari video games (and subsequently home computers in January 1982). Supposedly Alga/Cherry during 1981 sold video games at a value of 8 million SEK, with the aim to sell for 20 million in 1982/83 and 40 million in 1983/84. At this point, it is said that 8% of all households in USA who has a TV set also own an Atari video game, and the aim for Sweden is at least 1%.

Rest of the history line is irrelevant over here, but it goes to tell that at least on the Swedish market, sales of the Atari VCS begun in November 1978 which may coincide with when the Videopac G7000 begun to sell. I don't know about the rest, surely Luxor VES/VEC (Channel F) was already on the market, more doubtful about any RCA Studio III machines or for that matter the VC-4000 etc.