Was There A Programmable PONG Chip?

Students of game console history will reach back into the 1970s for the primordial machines, tracing from the Magnavox Odyssey onwards, and thence via the Fairchild Channel F into the world of microprocessors and the chain of machines that lead us to those we enjoy today. In the early days there was a parallel evolution for a few years of dedicated video game consoles with no interchangeable cartridges or microprocessors, these took their inspiration from the legendary PONG arcade game and used dedicated non-programmable hardware in custom chips to create their video. But was there a programmable PONG chip lurking among all the others? [Old VCR] takes a look.

Many readers will be familiar with MOS Technology as the originator of the 6502 processor used in so many 8-bit home computers. But perhaps many of our attention spans will have passed over another of their products, the MOS 7600 and 7601. These were the chip company’s entry into the surprisingly congested mid-70s PONG-in-a-chip market, and the article investigates the question of whether they might in fact be mask-programmed microcontrollers masquerading as dedicated chips.

It’s a fascinating tour through the mid-70s in terms of games consoles, MOS, and through their eventual takeover, Commodore. The possibility of a mask-programmed PONG chip is explored in detail and discounted, though like [Old VCR], we’d love to see one decapped and reverse engineered. For us a stronger line of evidence comes in asking why MOS would stop at PONG if they had a mask-programmed microcontroller in their catalogue, and that our not having seen MOS microcontrollers appearing all over Commodore’s subsequent products suggests that it may be simply another dedicated PONG chip like all the others.

We’ve seen quite a few variants of this iconic game over the years, but few as impressive as one made from discrete components.

18 thoughts on “Was There A Programmable PONG Chip?

      1. It should be pretty easy to port the game to any system. All hardware is controlled through drivers. So it is only necessary to rewrite the drivers, the rest should be copy & paste.
        That said, i couldn’t find a lot of information about the ESP32’s DAC. 8-bit resolution is enought, but i couldn’d find the sample rate. I guess 6M/Sa to 8M/Sa are necessary (but only in bursts).

        If i read the datasheet correct, there should be a HAL sensor inside, so maybe it would be possible to control the game just with a magnet. Then even the potentiometer would be integrated to the Chip. ^^

    1. So was the original Pong arcade machine, and all arcade machines from 1971 to around 1975, when the first microprocessor-based arcade machines come out. Arcade machines based entirely on TTL+passives persisted until around 1980/1981, but from 1975 to that point, that number dropped off as rapidly as the number of microprocessor-based arcade games rose.

        1. A number of Atari games for the arcades were originally designed on the PDP-11 family using graphics enabled terminals as their display methods. It then followed that the computer himself tagged along. Sometimes supplying only his chip self, or perhaps a Micro-PDP-11 variety.

      1. I had a tabletop coin-operated pong machine I got cheap, broken, from the tennis club down the street. Not sure what brand it was, or even when it was made, but it had 4 gigantic circuit boards full of discreet 4000 series logic chips. No CPU. Problem with it was, yep, you guessed it, just a blown fuse. The fuse however was on the underside of the power supply and not immediately obvious.

      2. I worked as an arcade machine repairman starting when the arcade and video game madness was just starting to explode (1978-79). The company I worked for ended up having 10’s of thousands of machines including many of the earliest ones. As a side gig, I was buying up old pinball machines super cheap that were being displaced by video games. I would restore them and find new locations for them (and made a lot of money).
        I can assure you that other than the original pong (see my other post) and jukeboxes they all had a microprocessor in some form or another. Earliest ones based on intel 8008 or 8080 or 8051. Later ones on 6502 or 6800 or 6809, later moving into 68000, and 80186 and beyond. Even the really crude ones like whack-a-mole had a microprocessor. (Jukeboxes were mechanical marvels!)

  1. I was skimming kinda fast, but didn’t seem to see an awareness of what the “useless sliders” on the one machine were for. Some varieties had “useless dials” instead, and I think one particularly cheap one had a double row of holes and a couple of pegs. Got it yet? They were for the players to manually keep track of games won.

    Some 3+ decades back, I as young meddler looked inside a few machines. I guess it was at the point that these were the oldschool 70s stuff, and there were still scowls of dissaproval if I got near any of the newschool 8 bit computers with a screwdriver. Also the cheaper iterations were turning up for “50p” or so at rummage sales etc. Anyway, the “good” one was a Grandstand electronics 8 in 1…. and that one seemed like it might have had the 7600 type chip in it. My buddy’s was a cheaper unit with only “tennis” on it… and the chip was sanded off and the same shape… by referring to the selector knob on mine, we were able to make it play “football” by shorting some pins. (That was the only one we could get without lifting pins etc I think, we weren’t up to that level of hacking yet). Just with doing that, I thought we had the secrets of the ages down pat and I reconfigure any of them to play any game. So then I got that 50p well used, and cheapy cheap cheap version… and got into that one and bummer, not the same chip…that one only played tennis, and did not even have in game point scoring, just two lonnng “useless sliders” across the top of the machine to do it yourself. The chip in that one was a long, thinner 28pin IIRC, and that might have been a mask PROM 4 bit CPU. I was a pre-teen and had maybe one good screwdriver and https://en.wikipedia.org/wiki/Cow_Tools so not much was discovered at that point only that various states of corruption could be achieved with random pin shorting, no more games on that one, just “tennis” with manual points.

    1. I did the other way. 40+yrs ago i bought a kit for such a console at Hobbytronic electronic fair Dortmund Gernany i think it was branded Phillips. It had the 7600 kind chip plus a few TTLs and 8 Games.

  2. As always, the real content is in the comments. If you scroll down you’ll see that someone had in fact decapped a 7600C and posted a link to their high resolution photos of the metal layer and poly.

    Better yet, it seems Ken Shirriff has caught wind and plans a future blog post.

    From my untrained eye though, it sure looks like the structure in the lower left is a mask ROM, as evidenced by the address decoder along the top edge and the irregular filling of the poly below.

    There are two smaller ROM-looking structures, my guess is that they’re both LUTs by the sparser more regular pattern of poly bits and the lack of a decoder.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.