The Computer We All Wish We’d Had In The 8-Bit Era

The 8-bit home computers of yore that we all know and love, without exception as far as we are aware, had an off the shelf microprocessor at heart. In 1983 you were either in the Z80 camp or the 6502 camp, with only a relatively few outliers using processors with other architectures.

But what if you could have both at once, without resorting to a machine such as the Commodore 128 with both on board? How about a machine with retargetable microcode? No, not the DEC Alpha, but the Isetta from [RoelH]— a novel and extremely clever machine based upon 74-series logic, than can not only be a 6502 or a Z80, but can also run both ZX Spectrum games, and Apple 1 BASIC. We would have done anything to own one of these back in 1983.

If retargetable microcode is new to you, imagine the instruction set of a microprocessor. If you take a look at the die you’ll find what is in effect a ROM on board, a look-up table defining what each instruction does. A machine with said capability can change this ROM, and not merely emulate a different instruction set, but be that instruction set. This is the Isetta’s trick, it’s not a machine with a novel RISC architecture like the Gigatron, but a fairy conventional one for the day with the ability to select different microcode ROMs.

It’s a beautifully designed circuit if you’re a lover of 74 logic, and it’s implemented in all surface mount on a surprisingly compact PCB. The interfaces are relatively modern too, with VGA and a PS/2 keyboard. The write-up is comprehensive and easy to understand, and we certainly enjoyed digging through it to understand this remarkable machine. We were lucky enough to see an Isetta prototype in the flesh over the summer, and we really hope he thinks about making a product from it, we know a lot of you would be interested.

38 thoughts on “The Computer We All Wish We’d Had In The 8-Bit Era

    1. Apparently I reported this comment. Well by mistake I caught the ^button^ but this was in error as I left the page. Noticing the dots popup I came back. Seems it is not possible to un-report!!!.
      So here is the best I can do.
      The comment is fine, nothing wrong with it as such. The thing that is wrong is my mistake is unreversable it seems. Sorry; did I make up new words ??

    2. And the 6800 also. Got to see a real Ohio Scientific with all 3 CPUs. I had heard of the beast but never saw one when I was a young adult. I think Glitchworks might have one of those board (modern retro blank).

      And I was in the ‘other’ camp though I had an Atari 6502, I liked the 68xx family better.

    3. I guess my comment was rejected? Yes, it was OSI. II think the dual processor boards was part of the attempt at a business system including cool desk and computer cabinet and a hard drive with removable media. They tried several times to find a sweet spot and leverage their low manufacturing cost – due to a Molex pin backplane for up to 12 or 18 boards IIRC. And the PCB’s are uncoated. No solder mask or anything. They had a simple monitor/bootloader and BASIC and FIG-Forth that used their dual 8″ floppy. I have wondered what the sales numbers were.

      I have a Challenger II variation with the extra long backplane and 8″ drives in a separate case (not a use case). It ran a physical chemistry research setup for several decades. A UV optical system with cryogenics for samples. The entire system including analysis was written in 6502 Forth after trying BASIC, which was painfully slow. The Forth was about 20 times faster up to 100 times faster when 6502 assembly was sprinkled in for key operations – very easy to do in Forth.

      Did they also make a CPU board with 6502, Z80, and 6809?

    1. I had a Commodore 64 (it came with a 6510, which used the 6502 instruction set) with a Z80 coprocessor cartridge so I could run CP/M on it. I would have loved to have something like an Isetta.

    2. Yes, Indeed. And they made a “business” system with a desk and hard drive cabinet. They tried a lot of combinations of big and small and never hit a sweet spot people wanted. I have a Challenger II variant with dual 8″ floppies. It was the center of a physical chemistry instrument for a couple decades. 6502 only and the big card cage with 12 or 18 slots.

    1. Thank you for the information.
      I had no idea the Z80 was even more advanced than I had thought before.
      The part with the polysilicon wires, for example, gives an idea about complexity.
      In retrospect, we may often think that 8-Bit processors were simple, but they can’t be built on a kitchen table, either.
      Not at this scale, at least. An 100 KHz version (?) using discrete parts could maybe built at size of a wardwrobe. :)

  1. Hah! Back in the day I worked on a military system that inside was an Apple II with a commercial Z80 co-processor board advertised to run Z80 software with the 6502 just shifting data to peripherals. But we built a system (25kloc of Turbo Pascal) for terrain analysis that kept both pretty busy. Jamming functionality into that beast was lots of fun. Wonder if any of those co-processor boards still exist.

  2. That commodore was the computer we wanted, because it could do graphics and sound. The computer we wished for came later, the one that would run the tools and software to write this article, run the web site and the KiCad for this project.

    80’s wished for more ram, more bits, hardware multiply at an affordable price. Add 640×480 with 256 colors, non interlaced, to that dream list. Oh and 20MB of HD storage.

    Not saying it wasn’t exiting, and I own my career to the VIC20, but it was very limited to what you could do.

    1. Ugh, Commodore. MOS 6510, 40×25 screens and broken 1541 floppy drives.. I’m afraid Commodore was the domain of football fans, alkoholics, CB radio operators and the cool kids sharing pirated games at schoolyard.
      Men of culture™ had worked with Z80, CP/M, serial terminals or mainframes at university. ;)

      1. So you were the guy in the cravat holding his cardboard coffee cup with his pink upriaised saying, “You! Operator Person! Fetch me my printout! The banner reads LITTLE LORD FAUNT.” Just kidding….

      2. Nevertheless I was personally aware of at least twenty businesses in my hometown that were using Apple, TRS-80, and Commodore computers to run their businesses. There was low-end but workable business software for all of those machines. They all also had 80-column text adapters available, although you had to use a monochrome monitor for that to work and the kids playing games wanted none of that. Still, the economy of scale for “toy” computers made them far cheaper than even the lowest-end “for business” systems, and for small businesses that couldn’t afford $40K for a “real” computer those machines were a workable way to get some automation in place. The business manufacturers completely missed this bus, and they were unprepared when the beams crossed and brought forth Doom. Suddenly the gamers wanted a PC because first person shooters were completely beyond any of the 8-bit machines, and low end IBM clone manufacturers suddenly had a huge market they had never realized existed. Even though IBM invented the architecture their attempt to keep it a business system and lock down features from the low-rent masses just made them dinosaurs in an industry that was leaking away from them. This all happened because computers truly are general purpose machines. One made for video games can still keep your inventory, and one made to keep your inventory make games possible that aren’t possible on the cheaper boxes.

        1. You are so right. This was something that frustrated me a lot, BITD, that so many people didn’t get this. I got a “game machine.” People would chuckle about it, but I liked playing games on it, but also doing serious stuff on it, too. What was ironic is so many of the people laughing at me liked to do the same thing on their “serious” PCs; serious stuff, and play games. Except, I thought their games kinda sucked. They had high-rez graphics, but only internal speaker noises for sound. Ugh. Once they got Ad Lib or Sound Blaster cards, then they started cooking with gas… but it was still a “serious” machine. Right, keep saying that… :)

          1. You’ve forgotten to mention Tandy 3-Voice, Game Blaster and the LAPC-I and IBM Music Feature card! :)

            And Covox/Disney Sound Source for the poor.
            Or the obscure SID-based soundcards or the lesser known Covox Sound Master series.

            So strictly speaking, IBM PC platform had excellent audio support a little bit before Sound Blaster. It was niche, though.

            And to be fair, merely Tandy 3-Voice (IBM PC Jr., Tandy 1000) and LAPC-I (MPU401+CM32L, synth based on MT-32) had a big games support.

            The Roland synths had blown the Amiga out of water, even.
            Many high-class games such as sophisticated text-adventures with graphics had Roland support next to AdLib/Sound Blaster (Gateway comes to mind).

            It even got a little revival in mid-90s when the AWE cards had featured basic MT-32 emulation using EMU-8000 synth.
            They competed with other cards like Gravis Ultrasound or Soundscape S-2000. And late OPL4 soundcards, maybe.

            So PC platform had a lot to offer, all in all.
            Except for AdLib/SB, it just wasn’t always mainstream, though.
            The Roland synths were real music equipment, for example.
            It was to no surprise it was a different class to an AdLib or SB 2.0 compatible.

            And the Tandy 3-Voice sound was limited to a few PC models from the 80s. From the US. In Europe, they were rare.
            It wasn’t until mid-90s that PC users could run TEMU to emulate it in software and play music/sfx on their homebrew Covox plugs.

        2. “Suddenly the gamers wanted a PC because first person shooters were completely beyond any of the 8-bit machines, and low end IBM clone manufacturers suddenly had a huge market they had never realized existed.”

          Yes, I think I remember. And the no-name Turbo XT mainboards in the mid-late 80s were the first wave of this phenomenon, maybe.
          They’ve sold very well and might be the reason why users got accustomed to DOS PCs early on.

          The users merely had to install RAM and ROM chips for PC BIOS/BASIC (unless a BIOS replacement was shipped with it).
          And the peripheral cards and drives. V20 upgrade optional.

          After this, they got the ability to run one of the many MS-DOS compatible OSes. MS-DOS, PC-DOS, DR-DOS, DOS Plus, Wendin DOS, MOS, Alpha DOS, Sigma DOS, DCP..

          Strictly speaking, though, the IBM PC 5150/5160 and its clones were 8-Bit systems, similar to 8080/Z80 systems.
          The PCs address bus was 20-Bit rather than 16-Bit, bus width was 8-Bit.
          Thus, the motherboard was an ordinary 8-Bit design.
          It merely had been prepared for accepting more than 64KB of memory.

          Computers like Olivetti M24/AT&T 6300 and Amstrad PC-1512 or 1640 were real 16-Bit systems.
          They had an 8086 or V30 processor and an RAM that used 16-Bit rather than 8-Bit wide memory bandwidth. The M24/6300 even had 16-Bit slots.

          The IBM AT was IBM’s first true 16-Bit PC, I would say. The 80286 was an 16-Bit processor that deserved its name.

    2. “80’s wished for more ram, more bits, hardware multiply at an affordable price. Add 640×480 with 256 colors, non interlaced, to that dream list. Oh and 20MB of HD storage.”

      256 colors in high resolutions were a proper problem.
      On PC, the IBM PGC had 640×480 at 256c, but was very expensive.
      Clones followed, but it remained niche. Just like other graphics standards, which remained limited to CAD and DTP use.

      The only exception was Super EGA in 800×600, maybe.
      Multiple EGA compatibles had extra resolutions beyond normal 640×350 pixel EGA..
      They still had 16 colors merely, but dithering in that resolution produced good results on small multisync monitors of the day.
      Super EGA was in use before VGA got mainstream in 1988 (VGA was from ’87, clones followed early ’88).

      The other alternative that comes to mind was Amiga, which was labeled “Commodore” but really was a bit if its own. Amiga Inc had done most work, I think.

      Amiga ICS and OCS chipsets could do resolutions close to VGA (Standard VGA was 640×480), but topped out at 32 colors (VGA had 16, not counting mode 13h).
      And it had to use interlacing and overscan here.

      VGA was progressive scan and worked without challenging limits of TV technology (pro users used VGA or multisync monitors on A2000, through flicker-fixers or scan doublers).

      In HAM mode, Amiga could produce 4096 pseudo colors which was okay for still pictures.

      Unfortunatelly, only a few Amiga users had hard disks, which were standard on business PCs.
      And it wasn’t until 1989 that Kick Start 1.3 with HDD support was released.

      That’s why bridgeboards had been popular, maybe, I suppose.
      They allowed using PC side of an A2000 to handle communication with cheap MFM/RLL harddisks and ISA bus HDD controllers.
      Or ISA bus filecards, early AT-Bus HDDs.

      On PC, Super EGA or early VGA cards (all being SVGA cards, actually) were about only choice to get many colors.
      That’s when more than 16 colors could be drawn on screen.
      Either real or through dithering.

      Again, niche graphics adapters had existed for years, too, but had limited software support. Windows 1 or 2, GEM, AutoCAD, Wordperfect and so on.

      IBM EGA could also switch palettes on the fly, so an EGA monitor could be made to display 64 colors same time. But that was a cludge, a hack.
      If only IBM EGA had full 64c support right from start. Sigh.

      In professional field graphics workstations had existed, of course.
      SGI, to name an example.

      And UK had Acorn Archimedes, which also had impressive specs for its time.

      Last but not least, there had been MSX2, which was sort of niche in Europe, but well supported in Japan.
      Some MSX2 or MSX2+ models had better specs than that offered by mode 13h on PC (PS/2 video, VGA).

      That being said, graphics aren’t everything. Sound and connectivity were equally being important.

    3. “80’s wished for more ram, more bits, hardware multiply at an affordable price. Add 640×480 with 256 colors, non interlaced, to that dream list. Oh and 20MB of HD storage.”

      I forgot something. Too bad HaD has no edit function.

      Early OS/2, – version 1.1 was graphical and around in ’88 -, had shipped with support for IBM’s 8514/A video adapter (intelligent, accelerated).

      It could do 640×480 in 256c or 1024×768 in 16c, I think. Clones had more combinations.

      So it was technically possible that users of PC workstations had access to 640×480 and 256 colors in the 80s. Using official IBM hardware.
      – Before VGA (SVGA) cards had been popular and had offered proprietary video modes (VESA VBE fixed it, but was late).

      XGA and XGA-2 also were around, but never catched attention of the public anymore.

      8514/A had a chance, but to users Super VGA was cheaper and built on beloved VGA technology.

  3. Before the Apple ][E came out, I picked up for personal use a Franklin Ace 1200 at Children’s palace for about $300. Franklin ran afoul of Apple because they basically cloned the Apple ROMs. The Ace 1200 was actually a lot nicer than the Apple ][ because it had a better keyboard with upper/lower case, had support for lower case characters, beefier power supply with a fan and the disc controller on a chip with IDC ribbon sockets on the motherboard for two discs. I read a lot of Usenet articles back in the day on that system.

    After Franklin lost to Apple, they de-appleized the remaining Ace 1200s to get rid of the ROMs via a weird mezzanine card and a bunch of hack wires. A pre boot disc could allow it to do a bad job of running ProDOS. We bought a bunch of the de-appleized Ace 1200s from Children’s Palace re-Aceified them by removing the hacks and burning 2716s with the original ACE images – which were basically purloined Apple ROM images.

    We installed Z80 boards, banked RAM boards and 80 column video boards to run CP/M applications or Pascal P-system applications for controlling lab equipment. Pascal supported a serial console and I had an inexpensice VT-100 compatible green screen terminal on my workbench.

    The other cool thing we got from Childen’s Palace were lots of Vectrex video consoles which, if I recall, were going for $39. The Vectrex had a 6809 and a pretty neat vector graphics Asteroids type game built in. We really didn’t care about the 6809 board. We bypassed everything brought out the magnetic deflection board X and Y, and video blanking connections to BNC jacks. We drove the displays combining two 8-bit D/A channels of Mountain Computer A/D-D/A board for each axis. We were able to display up to 8 channels of EEG waveforms being sampled on the input side of the board.

  4. My 6502 Apple ][ clone (called a Wombat) also had an add-in card with its own Z80 based CP/M computer that gave me a full 80 column or higher display, which I used for WordStar and VisiCalc etc.

  5. I remember having an atari linked to a second unit running as a Microsoft pc, I believe it was possible to get a recased system that was an apple /pc/atari hot switchable system. It worked for me

  6. This really is an amazing project from a technical standpoint, basically a homebrew discrete TTL/CMOS CPU with a reconfigurable instruction-set… and without using an FPGA?! Looks like a lot went into making this, and I’m sure a lot could be taught/learned from a design like this.

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.