The Most Notable Obsolete Computer Platforms

Geek!This is Brian Patrie’s submission for the HP Magic Giveaway. Feel free to leave comments for this article as you see fit – your feedback is certainly welcomed! If you’d like to submit your own how-to, what-is, or top-five list, you can send it to me. Views and opinions of this writer are not necessarily my own:

Before 64-bit. Before 32-bit. Before even those old archaic 16-bit machines of the mid to late ’80s. We used 8-bit microcomputers. Okay, before that there was 4-bit; but I’m not that old. By 8-bit, I mean fully 8-bit. The 8088 CPU had an 8-bit external data bus; but internally it was the 16-bit 8086.

Here is a hastily compiled list of a few noteworthy machines from the heyday of 8-bit computing in the mid 1970s through mid 1980s.

I’ll start with the Apple II—mainly because it was the first micro that I ever knew. I was immediately enthralled with the idea of being able to make it into anything that I could programme it to be. Electronic Lego-blocks, I called it. Eventually I got my hands on the Apple ][ / ][ Plus Technical Reference Manual, and read of an electronic logic ecosystem that didn’t waste a single gate. Its wide open architecture made it a hit with hardware hackers (and, no doubt, made hardware hackers of many of its initiates). Its handling of addresse decoding on the motherboard made it a simple matter of checking one of two lines for a card to detect that its memory or i/o space was being addressed. It did have some limitations compared to some of its later contemporaries, such as monochrome text, and some bizarre graphics idiosyncracies (things that Apple were slow to improve, due to being distracted by other major projects, like the Apple III, Lisa, and Macintosh). But it was a great machine, nonetheless.

It’s fairly predictable that the Commodore 64 would turn up in the list. It’s most obvious strength was its price—which made it one of the most popular personal computers of its time. For a few-hundred bucks, you could have a working system. Its design didn’t match the austere elegance of the Apple II; but it had some attractive features beyond it, including chroma/luma video output (which is what s-video is), multicolourable text (limited to one background colour for the whole screen), and sprites. It made for an attractive game platform. (I gather that Commodore did attempt to market a C-64 based game console; but it failed miserably.) The worst criticism that I have of the C-64 is the disk-unfriendly initial user interface. Starting a programme from disk required a rather cumbersome LOAD “*”,8 followed by RUN (versus the short and sweet 6 ctrl-p of the Apple II, or nothing for the auto-starting II Plus). And when developing a BASIC programme, one had to specify the device number for every stinking SAVE. (I was spoiled by the Apple II’s feature of remembering the last drive you used.) It was also impossible (without 3rd party software) to list a disk directory without stepping on the BASIC programme in memory. (I confess, though, that the directory-as-a-BASIC-programme was a cute trick.) These things may seem trivial; but they were enough of an annoyance to turn me off to an otherwise reasonably attractive platform.

Most people never heard of the Panasonic JR-200U. A friend of mine found one at a yard sale, and it ended up in my hands. Apparently it had a bit of a following in Japan and Europe (where it was usually sold under the National brand). It sported many of the features of the C-64, plus the ability to colour the background on a per character basis. Unfortunately its display was only 32×24 characters (versus the 40×24 of the C-64 and early Apple II,II+ and 80×24 of the later Commodore models and Apple IIs). It also had an 8 colour 2×2 per character block graphics mode that allowed any text character cell to display 4 independently coloured blocks, along with some higher resolution modes that acted like colour custom character sets. One of the little things that made me smile was the ability to use hexadecimal numbers simply by prefixing them with a $ (the old Berkley convention). This could also be used with the val function to convert a string to decimal within a programme; and there was a hex$ function that worked like str$ to convert the other way.

The Kyocera Kyotronic (best known in the guise of the TRS-80 Model 100) was one of the first notebook computers. The Epson HC-20 has the honour of being THE first; but its calculator-sized display was a bane to its popularity). The Kyotronic sported a 40×8 character display which made it practical for basic word processing. It’s ability to run for several hours on a handful of AA cells made it particularly popular with field journalists. it came in 16 and 32kB RAM configurations—the rest of its 64kB addresse space being reserved for firmware modules that contained application software. (It is also noteworthy as the last project to which Bill Gates personally contributed the majority of code.)

I should probably put the Sinclair ZX Spectrum (ZX-82) here; but I know almost nothing about it. Instead I’ll mention its predecessor, the Sinclair ZX-81. This was an extremely frugal machine that came with a whopping 1kB of RAM (expandable to 16 or 64k with one of two rather cantankerous modules, that you’d better not bump whilst in use). It was available assembled or in kit form. It had a 32 column monochrome text display with the curious feature of being firmware generated—which allowed software to take it over and produce highish resolution graphics. It’s not a machine that I would be terribly interested in using for anything anymore; but it warrants mention by virtue of being interesting and weird. (Maybe I should have included the TRS-80 or Atari 800 instead. But I’m not very up on those, either.)

9 thoughts on “The Most Notable Obsolete Computer Platforms”

  1. No talk or mention of the Commodore Amiga platform? The Atari 2600? What gives? These were both highly notable (the Amiga sold more games than all other platforms combined until Doom was released). And they are both very obselete. No mention of any of the Tandy computers? And even the C64 was not Commodore’s first big hit.

    I think the title should be changed, as this does not even scratch the surface of obsolete computers available today. Even the PCJr. should be mentioned.

  2. I admit that i neglected platforms that i’m simply not familiar with (e.g. that Atari 2600—which i tend to think of as mainly a game console, anyway—and the Atari 800/400).

    I’m glancingly familiar with the TRS-80 Model III; but it never struck me as a remarkable machine. But i’ll grant that it was one of the early popular personal computers.

    As for the likes of the Amiga, NeXT, and IBM-oid Tandys, i was focusing on 8 bit platforms, and these are 16 and 32 bit systems (though the 8088s were shoehorned into an 8 bit motherboard architecture).

    My original title was “Brian Patrie’s Top Five Obsolete Computer Platforms (in no particular order).” I suspect that Chris felt that it was too conditional. I have to agree. I was attempting to excuse the narrow scope of my list. I should have just gone with more than five. Still, i don’t think it’s too shabby for a couple hours of off the cuff rambling. ☺

    Maybe i’ll develop it more some day, and broaden the scope. Or, someone else could write one. ☺

  3. Myself, I have some fond memories of the Atari 800 and 1200 computers. My first Atari was a 1200, believe it or not. Over the 800, the one thing I really liked about the 1200 is how the keyboard felt. It did not feel as cheap as the 800 did.. It had more of the feeling of the TRS-80 Model 3 and others at the time. For a period of time, I ran a FoREM BBS using a 300 baud modem. I don’t remember the brand of the modem but it was a third party brand modem that interfaced with (IIRC) the Joystick port. The Atari was not my first computer that I used for modeming and calling boards (honestly, that was an old TI data terminal with no screen, just a thermal printer)… the Atari 800 I had would live a portion of its life working as a “controller” for the old REC telephone conference (party-line) system from the late 80’s. Of course, the biggest joke of Atari was “Atari DOS3” which was a completely dumbed down version of DOS2 which was MUCH better.

    Before I owned the Atari, I had a Sinclair ZX-81… I had the RAM expansion. The only thing I really remember about that computer was that every time you pressed a key, the screen would clear and slowly redraw and that RAM module got EXTREMELY hot. It was cute.

    Also, don’t forget the Atari 2600 and their fairly rare BASIC Programming cartridge. This was one of the few “game programs” for the Atari 2600 that used the keyboard controllers (which I would assume these days is just as rare). The display would be a split screen where your “source” would appear in one screen, your “stack” of variables would appear in another part of the screen and then your output would appear elsewhere. You can toggle each one on and off. The functions (print, goto, etc.) were preset and involved having to make two keystrokes to generate. Overall, it was cute and had the potential of introducing many “geek-newbs” at the time to computer programming.

    I also have very fond memories of the TRS-80 Models I and III. I never owned one, but the Rat Shack store was just down the street and I had a bicycle at the time. Those were the days!

    Michelle A. Eyre, K7REC
    founder, REC Networks

  4. To some of us this isn’t really ancient history. I was the engineer behind the 8-inch floppy drive in the TRS-80 line mentioned above.

    While I was not directly involved, the company I worked for was party to the first successful 8-bit microprocessor, the Intel 8008. This was originally a joint venture between Intel and Datapoint, with the design work being done by Intel specifically to replace the TTL logic Datapoint was using in our early “desktop” computers. When Datapoint management decided that committing to a single chip was too risky (which sounds silly now), Datapoint released its rights to the design in return for not having to pay Intel what it owed for Intel’s design effort. Intel finished the design, went commercial, and the rest is history. One artifact of this relationship is that Intel’s instruction set even today still bears some resemblance to the Datapoint instruction set of the early ’70s.

    Datapoint kind of backed into making 8-bit desktop computers. The company began as Computer Terminal Corporation in 1968, making what was originally called a “glass teletype” to replace electro-mechanical, hardcopy terminals. The first one was called the “Datapoint 3300” and was a dumb terminal.

    Our next major product was a “smart” terminal, intended still to connect to a mainframe, but with enough smarts to have some functionality if the line to the mainframe went down. The intent was that it would be used in remote offices. It was our customers who figured out that the Datapoint 2200 was the first desktop computer and started using it by itself. IIRC, the 2200 came out about 1972.

    This led to a desire for a much more powerful desktop machine, and Datapoint responded with the 5500, which is the one originally intended to have the Intel 8008.

    Datapoint later did succumb to the lure of the microprocessor, and developed the 1500, which used a Zilog Z-80, another 8-bit chip.

    I can remember when we first started offering hard drives with the 5500. The first one was a Wangco 5 MB (that’s not a typo), 14-inch drive about the size of a trash compactor. I remember at the time that we all speculated that no one would ever figure out how to fill up something so huge. IIRC, our first floppy drives, used on the 1500 and 1800, had a capacity of 92 KB per disk.

    Fun stuff.

  5. I had a 16k RAM module for my ZX-81. It never got noticably hot; but that expansion bus sure was touchy. One bump the wrong way and -crash-!

    Well, surely i’m not going to come even close to winning the contest. But i’m getting good comments! ☺

    Interesting stuff about the DataPoint terminals. I hitherto knew nothing about them. Reminds me a little bit of the DEC VK-100 (from much later), which had a BASIC interpreter built into it (and could save and load via ASCII to and from the serial port).

Comments are closed.