A Love Letter To My Lost Amiga

My first love was a black wedge. It was 1982, and I had saved up to buy a Sinclair ZX81. That little computer remains the only one of the huge number that I have owned over the years about which I can truly say that I understood its workings completely; while I know how the i7 laptop on which this is being written works I can only say so in a loose way as it is an immensely complex device.

Computing allegiance is fickle, and while I never lost an affection for the little Sinclair I would meet my true electronic soulmate around eight years later as an electronic engineering student. It no longer graces my bench, but this was the computer against which all subsequent machines I have owned would be measured, the one which I wish had not been taken from me before its time, and with which I wish I could have grown old together. That machine was a Commodore Amiga, and this is part love letter, part wistful musing about what could have been, and part rant about what went wrong for the best desktop computer platform ever made.

Just Why Was The Amiga So Good?

If we had been so convinced by the promise of the Amiga platform to the extent of not seeing the shaky foundations upon which it had been built, just what was it that had seduced us? Perhaps at this point it’s worth taking a quick look at the competition in the world of 16-bit or above computers, to see what else we could have had in our dorm rooms. This is a view from the UK so your market may have had some other players, but for us there were probably five main contenders. If you didn’t have an Amiga you might have had an Atari ST, if your parents chose the computer for you then you had an Amstrad  PC-XT clone, if your school had used them you could have had an Acorn Archimedes, and if your parents were very rich you might have had an all-in-one Mac. The Apple IIgs wasn’t big for Brits, and though the Texas Instruments home computers were 16-bit they had been competitors for the 8-bit era.

A typical rendered Amiga scene, mirror ball, table, checkerboard floor, blue sky
In the early 90s being able to do this was a big deal. Morn, CC0.

Of the other platforms, the Atari had MIDI but the TOS/GEM operating system and desktop was clunky, the PC was what everyone told you was the thing to have but was slow with no sound and awful graphics, the Archimedes was cool and very quick indeed but a bit weird, and the Mac was awesome but unaffordable. The Amiga by comparison was affordable, had amazing graphics for the day, had good audio, had a slick GUI over a proper multitasking OS, and felt fast due to its custom coprocessor chips. By my final year as a student, people were buying 386sx-25 PCs with Windows 3.1, and even those supposed powerhouses felt sluggish and cumbersome compared to the Amiga.

There was of course nothing special about a 68000 running at 7.16 MHz, the key was in the Amiga’s ROM-based OS and those extra-special custom chips that meant that the 68k could compute rather than waste its time on graphics and waiting more than necessary for disk drives. Mine got me through an electronic engineering undergraduate course: it compiled C programs, edited audio for student radio, played marathon overnight Lemmings sessions, and even once spent more than a day rendering the obligatory ray traced mirror ball over a checkerboard surface.

We Made Amiga. They F****d It Up

The above phrase was a legend back in 1990 when I had my first A500, a message buried in the version 1.2 Kickstart ROM. It was a commentary from surviving members of the Amiga development team on Commodore’s dismal management and marketing, but we tech-obsessed youths drinking the Amiga Kool-Aid a a few years later were naive and ignorant of such matters. For us it was streets ahead of everything else and was sure to have a promising future. How wrong we were.

A still from the Amiga 1000 launch advert in 1985, a man walking down a pillared hall towards a distant computer on a pedestal.
The Amiga 1000 launch advert is a minute long. Incredibly, you only see a glimpse of the computer at the very end, and they don’t show you any of what it can do.

A very illuminating read as a former Amiga user came from our Hackaday colleague Bil Herd, in Back into the storm, his autobiographical account of working for Commodore in the mid-1980s. He was bringing us the TED series of 8-bit computers rather than the Amiga, but he illustrates well the chaos and ineptitude reigning in the upper levels.

The impression I gained from it and reading about the early history of the Amiga was that the engineering teams were building impressive hardware, but the company was hell bent on hammering everything to be another Commodore 64 rather than recognising that the computer market had potential beyond a mass-market home gaming platform. It’s a story repeated through the later years of the Amiga by [Dave Haynie], who gives us tantalising glimpses of the Amigas we might have had, when what we were given in the early 1990s was underwhelming and far too little too late.

The infamous phrase from the 1.2 Kickstart ROM referred to Commodore’s dismal marketing for the original Amiga 1000 in 1985 when it could have stood alongside the best of its competition or ahead of many. Commodore’s founder Jack Tramiel had alienated his dealership network before his departure, and a scarcity of supply alongside the most underwhelming of TV advert campaigns was not destined to make up any shortfall. The dismal marketing continued with the much more affordable Amiga 500 and its A2000 professional stablemate. The A500 became a hit despite, rather than because of, Commodore’s efforts.

From Infinite Promise To A Failed Game Console

An Amiga CD32 game console
Taking on Nintendo and Sega without some very special new tricks in your console made you either very brave or stupid in 1993. Evan-Amos, Public domain.

While their would-be competitor Apple upgraded their lines with new models containing faster processors and more expansion, Commodore rested on their laurels. Over the seven years following the launch of the A1000 they would repackage essentially the same 7.16 MHz 68000 multiple times in  a pretence of releasing new models , culminating in 1992 as the writing was on the wall, with the Amiga 600.

When they did upgrade to a 68030 with the A3000 it was in a top-end model out of reach of most customers, and when a 68020 made its way into a low-end Amiga in the A1200 it was the penny-pinching “EC” version without a memory management unit.

By the time Commodore finally went bust in 1994 after a failed attempt to take on Sega and Nintendo with what was essentially an A1200 in a console, the Amiga line was seen as primarily a gaming platform, and an increasingly outdated one at that. Ownership passed through Escom and then Gateway with no new models, and eventually ended up with the name licensed, now a minority platform for a few enthusiasts. The flame is still just about alight as the occasional OS update proves, but the dream is definitely over.

It’s easy to fall into wistful might-have-been stories, as no doubt happens whenever a few Amiga enthusiasts raise a glass. None of us were in the Commodore boss’s office, even though it’s hard to believe we could have done a worse job. It’s not difficult to see what they might have tried: meaningful hardware upgrades and a move to standard interfaces such as PCI among them. Who knows, if that had happened I might now be doing this in front of Workbench, instead of GNOME.

 

140 thoughts on “A Love Letter To My Lost Amiga

  1. It is always a pleasure to read what Jenny has to write.

    When I think of computers that I really enjoyed but will never see again, I think of the DEC-20. It was a big iron timesharing machine, not something that sat on my desk, but even today, I think back to things that that it could do. One very clear realization — it is all about the software. DEC really did offer amazing software. I could do things with TOPS-20 that I still have not seen developed with modern operating systems.

    In the early PC days, the 68000 really did offer things that the 8088 did not, in particular a linear address space. Assembly programming on an 8088 was a miserable experience whereas the 68000 was a joy.

    One wonders how different the world would be if the possibilities of the 386 had been more fully exploited rather than having been shackled to running DOS or Windows. It really is all about the software.

    1. One wonders how different the world would be if the possibilities of the 386 had been more fully exploited

      Actually, they were, although not commercially at first. That is how we got Linux. The rest is history.

      1. Well indeed, you are quite right.

        I was pondering linux and unix in general and in the light of computing history. The fact that unix had its beginnings as a timesharing system rather than a “personal” system has had a lot to do with why it serves us so well today on personal computers. A curious fact.

    2. OS/2 did make full use of the 80286 and 80386 processors. It also had supported virtual memory and memory-protection on the 80286 (OS/1 1.3). AFAIK, OS/2 was the only OS to fully utilize the ring scheme – to this day. Not even Linux did accomplish this. I heard many Amiga users valued OS/2 for being so sophisticated at the time, similar to how Amiga OS used to be.

      1. Yah I ran OS/2 warp demo back in mid 90s and coming from an A1200 it was a lot closer to Amiga multitasking smoothness and lack of OS friction than Windows 95 was at that time. 9x got there by the time you had a couple of hundred mhz, 32MB of RAM and the more refined 98SE. Not that 95 wasn’t an improvement over 3.1, it just didn’t get all the way there.

    3. “if the possibilities of the 386 had been more fully exploited… It really is all about the software.”

      This is really quite a complex question. The reason why the 386 wasn’t exploited until a decade after it came out even though it was the most popular processor, while 32-bit computing with the 68K series was well exploited from the get-go and still failed is all really the same thing.

      The 386 was dominant, because the 8080 was the dominant business architecture in the late 70s. The 8086 was designed the way it was as a 16-bit enhanced 8080 for upward compatibility reasons. That’s also the same reason why it was chosen by IBM and most of the business-oriented computer companies in the early 1980s; and it’s also the same reason why they chose the 8-bit bus, 8088 version for actual products (e.g. the Victor 9000 / ACT Sirius / other CP/M-86 / MSDOS computers): a quick, relatively compatible, upgrade.

      However, the choice of the 8088 meant that early x86 computers simply didn’t have the performance for software written in a high-level language or abstracted device I/O, which meant that all the early applications ended up relying on direct I/O addressing and segmented memory calculations.

      And because the 8086 and its successors the 80186 and 80286 were segmented machines based on the Multics virtual memory concept, almost no-one was able to exploit the better memory management of the 80286; still less the 80386 with its paged-baed MMU when it came out in 1985.

      Critically, it was the x86 software from 1980 all the way through to the mid-1990s which prevented progress in the PC world. From a non-PC users point of view it was a catastrophe and still is, because the all the legacy software (e.g. the built-in commands and batch language of CMD.EXE in Windows).

      So, conversely, the 68000 had to be better than a 16-bit 6800, because the 6800 had already lost in the first generation of microcomputers, to the 8080. Hence Motorola leap-frogged the 16-bit world and provided a 32-bit pdp-11 style architecture; one already designed for high level languages and compilers.

      But this action also cut it off from mainstream business computing, even though the vast data processing potential of 32-bit (or 24-bit) linear addressing would have helped enormously from a business viewpoint. Therefore it was companies that cared more about performance, true multi-tasking, multi-user and compiled languages who picked up on the 68K. Hence: Apple, Commodore and Atari jumped from the 6502 to the 68K (because at the time there was no 16-bit 6502); workstation manufacturers saw it as the quickest way to provide high performance GUI systems, because ‘C’ compilers could be quickly adapted from pdp-11 / VAX compilers. Unix was similarly easily ported.

      So, these companies benefitted from the technical superiority of the 68K’s architecture (which was *better* than everything Intel had up until the x64), but they couldn’t maintain the lead in the face of the inexorable advantage of incumbent systems. Literally: it didn’t matter how good the competition technically was, business interests only cared about Intel compatibility. And slowly, their technology caught up.

      1. I have heard that IBM had the 68000 under consideration, but it’s availability was delayed a few months and also Motorola wouldn’t second source it or license it to IBM.

        It was probably a wise decision for them as 680×0 availability did become problematic with increasing demand from Apple/Atari/Amiga and all the industrial applications. i.e. The amount of 680×0 machines we saw was pretty much the limit of possible given the narrow CPU pipeline.

        1. It was all down to the price at the time. 68000 was available, but only in a 64-pin ceramic package with gold pins. Since it was so big, they couldn’t reliably wave solder at the time without cracking the ceramic, so it needed to be socketed. Because you can’t socket components with pins and sockets of dissimilar metals (at least that was the IBM guidance at the time) the soicket also had to be gold plated. This meant the price was not competitive with the 8088.

          Motorola were rushing to get the 40-pin muxxed bus 68008 through qualification in time for IBM Boca – in a plastic packate – but couldn’t convince enough people to delay. And that single decision, ladies and gentlement cost our industry very dear.

          1. I had to replace the sockets in my A500 as the original became increasingly unreliable. I had to leave the top unscrewed so I could use a lifter to pop the ICs out and then reseat them
            After going to gold sockets , no more problems

      2. x86 didn’t really become a ‘finished’ platform until the 80386 because the 80286’s protected mode was only half baked. It couldn’t switch back to real mode without a system reset. So any operating system using 286 protected mode would have to be able to run everything 100% in that mode after switching to it. IIRC that also would have required a BIOS or other system ROM written to work that way.

        Windows For Workgroups 3.11 got partially there. With all DOS drivers protected mode compatible, 32 bit disk and file access could be enabled, drastically reducing how much Windows would need to access things via BIOS calls. The OEM only Windows 3.11 release still had Standard mode to run on an 80286, and for running on 80386 it had support for only one of 32bit disk or file access. I don’t recall which one but only having one was useless because it still had to drop to slower methods to access storage.

        Windows 95 finally was the all protected mode OS on the PC platform. Properly setup, it lingered in Ye Olde DOS Realm just long enough to pass the baton to 32bitville. But it was still possible to screw it up with an old DOS driver. It still could, if needed, dip a toe into DOS to run old software and could even lobotomize itself to reboot into a DOS only mode.

        It was intended to be a bridge between old and new and the later OEM releases were quite good. IMHO Microsoft nailed the UI with the addition of the enhancements that came with IE4 and later, except for the scrolling Start Menu that (IIRC) IE4 tried to foist upon the users. Kill the scrolling menu and anything on the system could be launched in 2 single clicks. That’s why Start Menu replacements like OpenShell are so popular for newer versions of Windows. They bring back that 2 clicks to anywhere ease of use.

        1. The 286’s inability to switch back to 8086 mode wasn’t an accident. It’s clear from the design of the 8086 that Intel were pushing towards fully-fledged segmented addressing for their processors, to run under Segmented Virtual Memory. And, in fact the 286 is upward compatible with the 8086 if 8086 programmers stick to indirect far calls and relied on the OS to allocate far memory (i.e. by treating the segment registers as abstract segments).

          In that case there’s no need to switch back to 8086 mode once we’re in protected mode.

          Unfortunately, programmers didn’t do that. Certainly the early versions of MS DOS were written without any serious memory management, forcing programs to do arithmetic with segments, typically by normalising them: Seg’:Offset’=(Seg+(Offset>>4)):(Offset&0xf). Also the 8088 was so barely faster than the 8-bit CPUs it replaced, competition forced programmers to take as many short-cuts as possible.

          So, the 286 could have worked, but even so, a segmented virtual memory OS would have been fairly terrible. Internal fragmentation of segmented memory allocation makes swap space a mess. Intel didn’t seem to understand that even in the late 70s, Multics was already failing in the academic world and that the most practical implementations also used paged memory.

          In that sense, yeah, the 386 saved x86 from itself. CS=DS=SS=ES=FS=GS=0 ! Segment problem solved ;-) . Nevertheless, Windows ’95’s GDI and USER were still 16-bit internally, where 32-bit APIs were Thunked down to their 16-bit equivalents.

          https://news.ycombinator.com/item?id=21451800

      3. The 68EC020 was a good choice for a low end Amiga. Nobody needed more than 8MB of ram on a stock A1200 at that time. You could expand it later by adding a faster CPU without the ram limitation.

    4. Ah memories!
      National semi introkit (SCMP)
      Nascom 1
      Nascom 2
      MK14
      Kim 1
      Jupiter ACE
      ZX81
      Commodore PET
      Osborne 1 (luggable if you worked out :-) )
      C64
      C128
      Spectrum
      BBC computer
      AMIGA 500, 1000, 2000

      Had all of them at some point

      All relatively affordable for average families… made a family of enthusiasts and programmers & hardware people.

      All before I even touched a PC

      1. Yes, they were fun toys, indeed. Programmable calculators with a hex pad like the SCMP, MK14, KIM1 were part of several construction sets. Here in Germany, we had the Sharp MZ-40K or the Busch Microtronic 2090.

        https://www.homecomputermuseum.de/sammlung/detailansicht/comp/Computer/show/microtronic-computer-system/

        However, calling those microcontrollers “computers” is a bit of a far stretch, maybe. They’re more like programmable logic or control systems used in elevators. Or they could be described as teaching tools, as experimental “computers”. Nowadays, we would call them Arduinos, maybe. ;)

        But I suppose that’s part of the problem here.
        Many people writing in the comments have wonderful memories of their toy “computers” and compare them with real computers.

        But what’s a “real computer”, anyway ?

        To people like me, it’s a personal computer that has a fair amount of working memory, proper i/o, say a serial terminal or a screen+keyboard.

        – Ideally, it can handle 80×24 (line 25 is a status line), this was the standard resolution of terminals at the time (VT100, VT52 etc). Many applications simply expected this as a minimum. Some applications used even higher resolutions (132×24, VT220).

        A real computer can process information and hold it on a storage medium, like a HDD, floppy, RAM Disk.

        In essence, anything that can run a microcomputer OSes like CP/M.
        Because, CP/M from the 70s was defacto the original microcomputer OS, so it was the reference. It was cloned a dozen time, as well, making it a whole platform.

        Oddly, some popular home computers like the C64 could qualify as personal computers under certain circumstances, too, since they could simulate 80 chars in software or run operating systems like GEOS.

        Or the other way round, they could qualify as terminals (if CP/M cartridge was used, for example).

        A cassette may also qualify as a storage medium, but it doesn’t allow random access, to individual files, it’s rather sequential only. Programs/data must be loaded in memory as a whole first. However, there’s s chance that they won’t fit. Media with random access don’t have that limitation.

        Behemoths like a mainframe or a minicomputer also had these qualities. It had a storage drum, a FAST RAND fixed-disk or core-memory. Tapes and punch cards were used to import/export programs and data, but not per se used as mass storage. They also had electro-mrchanical typewriters or glass terminals for i/o. With that aforementioned 80 chars per line..

        Likewise, many higher-end home computers like the Apple II, Commodore 128, TO 9, the Atari ST or Amiga, the MSX 2 (has floppy drives, unlike MSX1) may qualify as personal computers or real computers, even though they aren’t Personal Computers™ (IBM compatible PCs).

        ..

        I mean, I don’t blame them. I loved my Sharp MZ-700 “Personal Computer”, too. But it merely had datasette and a 40×25 display. So it hardly qualified as a personal conputer, it really was more of a toy or home computer with a Z80 CPU. Sure, once if upgraded with a floppy drive and once it ran CP/M, then *maybe* it became sort of a personal computer.

        Even though most programs had a hard time running in 40×25 resolution. So it was necessary to implement some sort of soft scrolling to fit 80 chars on screen.

        80×24 also was the resolution that mailboxes aka BBSes of the time had expected.
        Unless they were operated by home computer enthusiasts, I guess.

        Or, weather forecasts via shortwave (RTTY, Navtex stc).
        Again, 80 chars per line were the standard (minimum)…

        It’s what “real” systems used.

        The 40×25 resolution was a home computer thing, because they were meant to be used with blurry TV sets at home.

        1. Not to pick too many nits but your definition of “real computer” there is a bit 80s-centric. Teletypes were limited to 72 character width lines and many early glass TTYs used to access big iron were limited to 60 or 66 characters. I don’t think anyone would say those weren’t real computers. As for why many 8-bit micros were limited to 40 column text that’s more down to display memory considerations and the desire for colour graphics rather than TV limitations. Colour TVs from the 80s can display 60 column text quite well if they’re fed a luma-only signal over RF or even 80 column text if they have a composite input.

          1. well, he is German. lol. From an entirely different perspective, I might say that CP/M wasn’t a “real” operating system and go from there… but yeah, doing a kind of survey of 80s home computers, and then defining “real computer” after-the-fact… it’s an elaborate “True Scotsman” fallacy.

  2. It’s easy to fall into wistful might-have-been

    Indeed. I see little chance for anything else than Wintel to happen back then. Even IBM didn’t manage to make its OS/2 popular and they didn’t have to compete with a different hardware platform. Apple managed by a tiny margin.

    1. Not quite. OS/2 2.11 and Warp 3 were successful for about two years, prior the release of Windows 95. At least here, in Germany.

      In this short time frame, quite a few popular programs were ported to OS/2. I remember WinCIM (OS/2 CIM) and the Opalis decoder software (Opalis/2; for BTX service, part of a Videotex family).

      Visual Rexx was sort of a neat Visual Basic alternative (a Rappid Application Development tool, RAD), also.

      Here’s sort of a gallery with commercial software packages for OS/2.

      https://www.os2.org/viewtopic.php?t=1448

      1. IBM spent far too much effort on marketing the ability of OS/2 Warp’s ability to format a floppy disk without bringing everything else to a halt. Those were some cringeworthy TV ads. Their major blunder was in not establishing a Games division and not creating an enhanced multimedia API.

        Microsoft from the start with Windows 95 promoted its capability with games and had their games division plus the DirectX API to accelerate games and multimedia while making it easier for game and multimedia software authors by having a common API for access to sound, graphics, and Human Interface Devices. Every program could talk to the system in the same way rather than every game and media program having to do that on its own.

        That didn’t stop some companies, (notably Adobe, Apple, and Microsoft themselves) from ignoring DirectX and other Windows APIs and common interface elements to reinvent all the wheels they didn’t need to. Would be interesting to know why the MS Office team was so stubborn. Users could learn to use Windows, then would already know all or most all they needed to use any other software, except for Office, Quicktime, Photoshop, Illustrator…

        1. Hm. This might be true, yes.

          From what I remember, there was a DIVE API in Warp 4 (’96).. It was similar to DirectDraw.
          The new 32-Bit display driver, GRADD, was also available. The first 32-Bit drivers were available years before in v2.11 I believe.

          Windows 95 (’95). The original Windows 95, -the official shelve version-, didn’t include DirectX, however. Not even WinG.

          Until Windows 98 was out, I didn’t even heard of DirectX in real life (I wasn’t much of a gamer).

          Sure, some games have used it before and silently shipped with the DirectX runtime on CD-ROM. Other’s used GDI, DIB or WinG or its own library (there’s a Sonic CD version that uses that DINO API).

          It wasn’t totally uncommon for games to do that at the time. MicroMan for Windows 3.1 had its own graphics library, WAP, I vaguely remember.

          But in the mid-late 90s, DirectX wasn’t in the people’s mind, really. Unless you were a gam.. computer game player.
          Up until ~2000, many software products still mentioned Windows 3.1 in the minimum requirements.

          The most prominent+international example that comes to mind is that amateur radio callbook CD..
          But there were others, too. Encyclopedia, communications software that shipped with modems, astronomy software etc.

          Microsoft released a Year 2000 CD at the turn of the millennium, which included Y2K capable versions of File Manager for Windows 3.x, even.

          https://archive.org/details/Microsoft_Year_2000_Resource_CD_April_1_1999_X04-75964_Microsoft

          I remember this, I was there. :D

          Now and then I casually visited the local computer shops and bought shareware CDs, did read the description on the game boxes (back then, when games still had big card board boxes). No word about DirectX. QuickTime was mentioned a few times, however.

          OS/2 Warp had its short time, I think, but the hype, not to say propaganda, about Windows 95 was huge. It dominated about everything at the time. It was nearly impossible to withstand it.

          It even affected lifes of people which had no interest in IT.
          Windows 95 was sort of a phenomenon in pop culture.

          For example, it gave birth to all those PC/Windows jokes (“I’m running Windows on my PC, I have a problem!” – “You already said that.” ; “My PC doesn’t work anymore!” – “Do you run Windows?”, “How do I know? -“Is the writing on the reset button still readable?”)

          Windows 95 overshadowed other OSes, including Windows NT, *nix, young Linux or Apple’s System. It caused DOS to die as a platform, too.

          It’s kind of surprising that OS/2 did survive this, at all.
          It still lived on as eComstation for a couple of years, now it lives on as ArcaOS.

          Here in Germany, OS/2 was being advertised as a “Migrationsplattform” (migration platform) for a while. Before Windows 95 was released.
          It was essentially a way out of the sub-par world of DOS/Windows 3.1x.

          OS/2 could run the classic DOS and Windows applications well, if not better sometimes (more memory, use of HPFS, quick file access through OS/2 HDD driver etc), but also could run native applications, both 16 and 32-Bit. The planned Power PC version had lost 16-Bit OS/2 compatibility.

          In addition, OS/2 allowed communication with mainframes.
          IBM had various terminal emulator products in its portfolio.
          I assume that’s why commercial users, like banks, had a faith in OS/2.

          It was stable and didn’t change as much over night as Windows.
          OS/2 not only ran in ATMs, but also on PCs in the bank buildings. And in other professional fields, I suppose.

          Of course, this market was little in comparison to the many markets served by Windows 95. But it was an important one, nevertheless.

    2. I think the possibility was there for the Amiga to have done an Apple. Strangely also the Archimedes, I used one of the first ones at school and it really was streets ahead.

      Not the Amiga with that management though.

      1. One of the 1st jobs for my time machine is to take enough gold bullion back to 1990-91ish for Donald Pleasance to buy Amiga out of Commodore’s filthy paws. :-)

        I think Commodore’s german division could also have done a better job.

          1. Didn’t it fall out that it would have been plenty of money considering later changes of hands. It’s like that due who wants not a penny less than $3000 for his “I know what I got” car, refuses $2000 and when he’s pissed off all potential buyers, gets $400 from the wrecker.

            Glad I was corrected though, would have been embarrassing to just hand like $5 million to SMERSH.

      2. Not wrong about the Archimedes. A mind blowing machine for the time. Streets ahead and amazingly DOS PC compatible with software emulation. It ran that about the same speed as an original 4.77 MHz 8088 PC of the time.

        I wrote tons of code in ARM2 assembler. That thing was lightning fast but Acorn fell apart financially, especially with the Olivetti takeover.

        On the other hand it was awesome to see the processor architecture spun out into ARM – the world finally caught up with that beautiful piece of RISC design.

  3. I feel you. My story is much the same, though i bought into the Amiga 1200 at the time, which to me was the best machine I’ve ever owned and loved it dearly. That was the last computer that had a personality, that you could love and covet like the One Ring.

    There was a mysteriousness about the Amiga, anything seemed possible, you would look forward to the next game, demo, disk magazine, graphics application etc and whatever popped up at the local Amiga club. It was a time of excitement, discovery and experimentation that I never since have experienced on any other emerging platform.
    PC’s and macs are unlovable by comparison, though some apples could be regarded as having at least a heart, but PCs were soulless hateful things in my eyes.

    I’ve backed up most of my data onto Amiga formatted floppy disks with the intention to recover them later on the PC but due to no way of reading the disks (I had to sell the Amiga to fund a PC). I sorely missed the music, images etc I had on it on the PC.
    I did transfer the bare minimum stuff on PC disks, the latest scene and objects I worked on the Amiga (Lightwav 3D data) that I transferred to PC. The main draw was its superior processing power. That softened the blow a little.

    I held off upgrading to PC until around 1994/5 when I got a P75 system with 16MB of ram, win311 and later the buggy W95 and W98. I explored the Gravis Ultrasound, soundblaster 16, midi, 24 bit true color modes etc. but resolving driver conflicts, patches for games to run the hardware you have, fixing damaged MBRs etc. It was an awful experience until Windows 2000 and later XP arrived. It took a good 10 years before I felt I was at the same point when I left the Amiga.

    Luckily through my own DIY floppy reading hardware and software I created around 2015 I was able to recover the 300+ fungus infested floppies with a success rate of 99.3% and recovered most of the data I once held in high regard. I’m now able to recreate (and have done!) the old Amiga 1200 hard disk I once had and run it in an emulator at 200x the speed. A long time wish has been granted, and felt like coming full circle.

    I too often wonder what computing would be like had the Amiga taken off like the PC did and if we’d have an even better platform and experience anno 2022. I’m convinced things would’ve turned out better somehow. If only companies would be inspiration driven rather than money and power grabbing people who should have no business dealing with society. Alas.

    1. ” If only companies would be inspiration driven rather than money and power grabbing people who should have no business dealing with society. ”

      Like NeXTstep and Steve Jobs? Apparently “inspiration” and “altruism” sound good on paper, much like “utopia”. The pursuit of money, by ourselves and others makes this world go round even if we don’t want to admit it like it’s a character flaw.

      1. No like the original Amiga design team, most startups that then sell their great companies with inspired ideas to investment companies that then make a right mess of things etc. I mean, look at the monstrosity of Rings of Power by Amazon. The schizophrenic mess that is Windows.
        Yes, economy has brought great prosperity, but also large inequalities and problems. It’s by no means perfect, that’s something that must be admitted as well. I’m no commi, but the current capitalism is the other end of the spectrum.

  4. I remember showing my Amiga off at a local computer show. The Radio Shack/PC/Apple crowd chucked and said, “It’s a game machine.” Yet when the PCs got around to doing what the Amiga could by about 1995 with the debut of Windows 95, they called it multi-media. Brand loyal hypocrites is all they were with blinders on of their choice manufacturers. It’s why I’m not an Apple fan to this day and I reluctantly use PCs.

    1. To be fair, the Amiga lacked proper data storage at the time.
      – Hard disk drives
      – CD-ROM drives

      While technically possible to use on the Amiga, too, they weren’t a thing in the Amiga prime days (A1000, A500, A2000, A2500).

      Normal PCs, on the other hand, had HDDs since the mid-80s, when the A1000 came out.

      By normal, I mean PCs used at work. With Hercules or EGA graphics and 10 to 40 MB HDDs.

      Not a p*ss poor student’s PC with CGA graphics and grandma’s B/W TV set in Chippendale optics. ;)

      By 1990-1992, when CD-ROM drives were slowly getting mainstream, PC owners slowly got soundcard with on-board CD-ROM interfaces.

      It was just a matter of time until users got the matching single or double speed drive as a retro-fit for their PCs.

      This was the time when multimedia software was increasing in popularity.
      PCs could read dozen of Megabytes from CD-ROM and store it on their 40 to 250 MB AT-Bus drives or venerable filecards/hardcards.

      VGA (’87) also was around in practice since 1988. Many highly integrated 286 desktop PCs with chipsets (NEAT, Headland etc) had VGA on-board. My had an ATI card on-board, I remember.

      With 256 Kb of baseline video memory, VGA clone hardware was good enough to draw 640×480 in 16c (mode 12h) or 640×400 @256c (mode 100h) or 800×800 in 16c (mode 6Ah)..

      All these things made multimedia titles possible.
      CD-ROM technology had several hundred Megabytes of storage, but transfer speeds were slow (150KB/s).
      Almost as slow as the Amiga floppy drives. 😂

      Without a cache or the ability to store parts of the software on another, quicker medium, it wasn’t feasible to do proper multimedia.

      Personally, I recommend reading the Trivia of Beneath a Steel Sky.
      It explains how depressing slow abd incompatible the Amiga platform was from time to time.

      https://www.mobygames.com/game/amiga/beneath-a-steel-sky/trivia

      PS: It’s kind of ironic that many of the shortcomings could be fixed by installing PC hardware in an Amiga.

      The PC emulator boards of the late 80s/early 90s gave access to all those peripherals the Amiga lacked so badly. HDD, CD-ROM, a Real-Time Clock..

      1. By the A1200/600 days the HDD was an easy fit and the CD-ROM was very much a possibility. But even then they did a Commodore and went for laptop HDDs rather than the cheaper 3.5″ ones.

      2. Your statements sound odd to me. There were several hard disk interfaces available for the Amiga in 1987 (and a couple in 1986). Most were SCSI interfaces, though a couple supported MFM drives. In 1983, when hard disks just started being available for PCs, they were very expensive luxury items. The first Mac to support SCSI hard drives was the Mac Plus, which came out in 1986. CD-ROMs were very rare prior to about 1988. The first 40MB hard drives didn’t come out until 1990. And as far as floppy drive speeds, Amigas were about as fast as they got, limited only by the disk spin speed. They held 880KB when PC floppies only held 720KB (when PC floppies increased to 1.4MB, hard drives were more common). The Commodore 64 floppy was indeed hamstrung by its very slow serial interface, but that’s another story.

        1. Hi! I think mnem explained the situation nicely.
          Better than I ever could, I admit.

          So maybe I should just keep quiet.
          On the other hand, I would really like to explain my
          point-of-view when I was writing those lines earlier.

          Okay. As far as I understand it was like this:

          In a very simplfied way, many Amiga users were happy with the base-line setup.
          Kids or teens couldn’t afford more than this, also.
          PC users, however, had to invest more from the very start and
          so they did things “right” from the beginning and bought some “extras”, too.
          They did “put wood behind the arrow”, so to say.

          ..

          That’s why writers of PC application software could rely on certain things, I guess.
          Like assuming the existence of a fixed-disk disk drive or a second floppy drive.

          Or they could recommend the use of a RAM drive, at least.
          RAMDRIVE.SYS was part of DOS and the memory above 1 MB, if available,
          wasn’t fully used for other purposes yet (a printer spooler, maybe).
          So users could use that as a replacement for a HDD or floppy.

          Windows 2.03, released ’87, for example, was the last Windows
          that offered the option to install on a floppy disk.

          “Your statements sound odd to me. There were several hard disk interfaces
          available for the Amiga in 1987 (and a couple in 1986).
          Most were SCSI interfaces, though a couple supported MFM drives.”

          Maybe. That’s quite possible. SCSI was very expensive, though.
          My father who teached me about IT always had been a professional PC user,
          who started using computers in the 1970s at university, then developed at home, too.
          He also witnessed the homebrew era of the time, the rise of CP/M and MP/M, 8″ floppies,
          the era of mainframes/terminals etc. So it’s no wonder that I’m slightly biased, maybe.

          To me, a real computer can display proper text (80×24; terminal standard),
          has a real keyboard, an optional pointing device (mouse/trackball, lightpen, drawing tablet) and enough RAM.
          The stock ZX81/Spectrum doesn’t fall into that category, -which is more of a toy computer
          that’s suitable as a controller for a model train or for a running light-, but the Amiga could.

          It wasn’t a bad platform, after all. Just a bit underpowered, even by 1985 standards.
          That’s the time when speadsheets in Lotus 1-2-3 started to run out of memory;
          when 512KB or 640KB of DOS memory became too little and when EMS with bank-switching was born (’85).

          The Amiga 2000, the least beloved Amiga, did most things right, I think.
          It was expandable, but still low-priced compared to a Macintosh II.
          In other words, it was the PC/AT pendant of the CBM world.

          That it was made/designed in Braunschweig, Germany, does touch my heart a bit, too.
          Unfortunately, the Amiga community didn’t like the A2000. It was “too ugly”, “too PC like”.
          Though the artists among them used them (and they had an HDD), at least.

          And so it happend that the Amiga 2000 was vanishing silently in ’91.
          Instead, those low-end A1200/600 models came out afterwards in ’92.
          They consisted of laptop-hardware, essentially.

          But by 1990, commercial Amiga box titles were already rare.
          So it perhaps didn’t matter anymore. Professional users were nolonger being considered.

          Whenever I check those old computer magazines, -as an unbiassed user (I think; no Amiga diehard)-,
          it becomes clear that the Amiga was treated as a toy computer (again!) by 1990 onwards.

          The Amiga was next to the Super Nintendo, the Sega Mega Drive, the Gameboy..
          By contrats, just a few years earlier, Amiga hardware and software was covered
          in generic computer magazines, still. As an equal, side by side with Atari ST, PC, Macintosh..

          That’s why I consider drawing the line for the “real” Amiga platform at 1990-1992. Roughly.
          All the later stuff, A600/1200, A3k, A4k, Amiga OS 2.x or 3.x was already in the hands of die-hard fans.
          It was a self-centered eco system. It has become akin to a car-tuning scene.
          – Sure, there was Deluxe Paint (also on PC) and the Video Toaster (NTSC only), some may say..

          Software of other platforms was nolonger being ported over, however.
          That’s at least the impression which the old magazines provide to me.

          Sure, there was still new Amiga software being made.
          But it focused on demoscene, MOD music, raytracing..
          Things the Amiga was strong in, anyway.

          “Normal” every day software, like a telepone book software, software to create stereoscopic images,
          a login-software for CompuServe or AOL, a software to fill out bank formulars,
          a software to create business cards (self-employed people were common back then,
          but even a flower shop also has a need for a business card), an astronomy software,
          educational software for kids, a card reader program for telephone cards,
          a digital stamp catalogue, a control software for your model train..

          In short, all the kind of software that could be found on “shovelware” CDs for PC/Mac.
          – The software normal users may had an interest in, besides playing games or tech-demos.

          Or let’s take MathLab, AutoSketch/AutoCAD, Photoshop.
          A feature-reduced port of Mosaic or Netscape.. Nothing.

          All this was nolonger being considered to be on the Amiga platform.
          But on the Macintosh and OS/2, and certain Unix systems (Solaris)..

          By the early 90s, it was nolonger the time of the Amiga as a full-fledged general purpose computer.
          Rather, it was all about gaming, accelerator/power pc boards etc.
          Again, it was as if you were reading a magazine about car-tuning or model making.

          The Atari ST had become similar niche at the time, but it vanished more silently.
          Remaining users were still using it behind the scene, as a music computer for MIDI/Cubase,
          to run a lab software, to maintain databases..

          All the PC-like stuff that ran on GEM desktop in monochrome in 640×400.
          Or in higher resolution with the appropriate upgrade, if applications relied on VDI/GEM.
          There was an ET4000 ISA card adapter for Mega STE, I vavguely remember.

          “CD-ROMs were very rare prior to about 1988.”
          Yes, I won’t dissagree here.

          What I meant to say was that slow CD technology required some
          sort of cache or temporary medium to be useful.

          This was provided in the form of HDDs or disk caches
          (MS-DOS 6.x came with Smart Drive).

          Without a cache, many PC games in the early 90s didn’t correctly work, either.
          150 KB/s (speed of original audio CD) were very slow.
          If an application had to load both its program code and game data off CD,
          it would end up stuttering during game play.

          This happens if you play those classic Lucas Arts games,
          like DOTT or Sam&Max on a single-speed drive without Smart Drive or another cache.

          But it also happened with CD-based encyclopedia or video playback (QuickTime).
          If parts of the game weren’t installed on HDD or an RAM Drive, things would be even worse.

          This leads to another problem – RAM. The most important bit in a computer.
          Both PC and Amiga had little memory for the tasks they were expected to perform.
          On PC side, it was increasing over the years, at least.

          An Amiga 500, however, left the factory with the same configuration for years.
          This means that application writers had to cope with this somehow.
          Sure, enthusiastic Amiga owners could solder and add more RAM, but not the kids.

          If a later produced A500 only had SIMM slots in another trap door or something..
          Because, 30pin/72pin SIMM was standardized, not a proprietary thing.
          It was found on soundcards and IDE controllers etc, too.

          All these things are addressed in that rhyme at Mobygames, by the way.
          And I honestly swear that I had exact the same impression before I was reading it first time.

          “The first 40MB hard drives didn’t come out until 1990.”
          Not quite, I’ve seen 1987/88 BIOS charts with entries for HDDs as large as 120MB.
          MS-/PC-DOS 3.30 (’87) had a 32MB parition limit, which caused trouble to owners of 40MB models in the late 80s.
          That’s why ads in late 80s magazine do talk about “special software” being required to fully use a HDD in DOS.

          The usual filecards in the mid-80s varried from 20 to 30MB, often.
          Early 5MB HDDs were already obsolete and out of production by 1988/1989.
          I read about this in a book about Aldus Page Maker (for Windows 2x).

          Speaking of books. The Amiga books about PC emulators were hopelessly oudated, it seems.
          A 1991 book I had just read does talk about MS-DOS 3.30 (’87) and “MS-Windows” and has no clue about XMS.
          Again, in 1991, when DOS 5 and Windows 3.0 were common (it has pictures of v2). *sigh*
          It was written by an Amiga enthusiast who owned a Bridgeboard (3-8990-314-2).
          I come to the conclusion that the Amiga users always lived a bit in the past
          or were at least somehow disconnected with the rest of the computer world.

          “They held 880KB when PC floppies only held 720KB
          (when PC floppies increased to 1.4MB, hard drives were more common).”

          I see were you’re coming from. Yes, it’s true if we talk about 3,5″ DD drives.
          PC-DOS 3.30 (’87) added support for 1,44MB HD drives, though.
          Many ATs had merely 3,5″ DD 720KB drives initially, but also 1,2MB 5,25″ drives.

          XTs could handle them if their floppy controllers were more recent and capable of 500 KB/s.
          The older models could do 250 KB/s only, which limited them to 360KB/720KB.
          Source: https://minuszerodegrees.net/diskette/some_floppy_maths.htm

          By using DRIVER.SYS or DRIVPARM.SYS from DOS 3.x, BIOS support could be added for HD drives.
          There’s also an “2M-XBIOS” program that does a similar job these days.
          https://minuszerodegrees.net/2M-XBIOS/2M-XBIOS%20-%201.44M%20as%20B.htm

          Last but not least, it was also possible to over-format DOS floppies (in the 90s).
          The DOS program “MAXI Disk” turned a 720KB disk into an 800KB model, for example.
          Others used non-standard formats, like VGA-Copy. Of course, this may caused compatibility issues.

          1. I’m not going to address most of that because we’ve already talked about some of it and other parts of it are more opinion / perspective based. There is one thing I would like to address though. The idea that software development for the Amiga wasn’t really a thing after about 1990 (except in niche areas) is largely a USA-centric viewpoint. That is the region where the Amiga had the least success and if that’s where you were located it’s probably skewing your viewpoint a bit. Here in Canada where the Amiga was far more successful there were retail computer dealers, software outlets, and even department stores still carrying both boxed applications and boxed games for the Amiga right through the early 90s up until ’94-ish when Commodore folded and that sort of stuff finally started to disappear from the shelves. My understanding is the situation in Europe (and especially the UK) was much the same, if not even extending a little longer in time there than here.

      3. I think this is more of a perception and price point thing. During the time where Amiga was relevant low end PCs didn’t come with hard drives either (floppy-only “turbo PC” clones were available all the way through the end of the 80s) and high end Amigas certainly did come with hard drives (included as standard in big box Amigas from the A2000/HD in 1988 and onwards). The real difference is low end Amigas sold *a lot* and on the PC clone side people tended to spend more money and not go for the bottom of the barrel floppy-only models. Remember that the Amiga 500 debuted at $700USD in early 1987. It’s not like you were buying a PC with a hard drive for the same price as a base spec Amiga without one, an XT clone in ’87 with a 20MB hard drive was roughly double the price. (Cheaper XTs were the Tandon model for $1400 and Leading Edge for $1500 both without the monitor, for example.)

          1. No problem! Context is always important and sometimes it’s easy to forget what things were like back then. Definitely for me as a tween looking at buying a computer with my own money from a paper route and mowing lawns the price point of the Amiga compared to other options was a big factor. PCs and Macs were just too expensive for me to even consider and the Atari ST wasn’t very popular in Canada at all (I don’t think there was even a retailer for them in my city) so that left the Amiga as really the best option for someone on a tight budget.

    2. Why you were at the supposed top of the world with your Amiga the PC was giving us chunky pixel graphics mode, Wolf 3d, Doom, Tie Fighter, System Shock, Photoshop, Wordperfect, 3D Studio version 1.0 and an upgrade path I could afford with tons of different options from many companies and varied prices. You thinks it was snobbery. I don’t think so! Amiga fanboy ism and delusion rewrite history but millions had their choice and choose the PC not out of some blatant agenda but out of genuine desire and user happiness. It’s crazy for people to think I and others didn’t enjoy PC’s.. just crazy.

      1. That’s not the average PC people had in the 80s when most PCs people had at home were equipped with CGA or EGA graphics and PC speaker sound. The Amiga was popular from it’s release in ’85 until the early 90s. Most of those things you mention weren’t a common thing on PCs until later. Wolf3D was ’92, Tie Fighter and System Shock ’94, 3DStudio ’96. Meanwhile the Amiga had Lightwave 3D in 1990. Heck, the first PC clone I bought in ’94 (a 486DX-40 with 8MB of RAM and a 1MB VLB SXGA video card) *still* didn’t come standard with a sound card. Yes, the PC won the race eventually (the release of Doom in ’93 is generally considered the final nail in the Amiga’s coffin) but saying the Amiga was never in the technological lead is exactly the kind of historical revisionism you’re accusing other people of.

        1. Yah it was the mid 90s “multimedia PC” really that gave gaming capability out of the box, you had to piece it together before that.

          As said elsewhere, 8 bit owners were looking down on CGA graphics, EGA wasn’t a huge improvement, and although VGA became available, the affordable lower end systems stuck with prior standards going into the 1990s. Monochrome was still selling for “serious” uses because of high res text and low eyestrain monitors.

          1. That’s pretty much the size of it, yes. Most of my other friends who were from a lower middle-class background like myself that had a PC at home, it was a turbo PC or XT class machine with a Hercules or mono-CGA setup and a PC speaker for sound. The Amiga was affordable for the masses, not the classes, as a certain YouTuber is fond of saying. That’s why you didn’t really see PC games drop support for CGA and PC speaker until well into the 90s, because up until then that’s what the bulk of PC users were still running with.

            During the time when I had my Amiga 500 I had exactly *one* friend who had what could be considered a proper gaming PC, a fast 286-16 with 2MB of RAM, VGA graphics, and an Ad-lib card. He was the only one because it was an expensive as hell computer and his dad could afford it because he was a senior level engineer. Even that “gaming” PC was only better for certain types of games than my Amiga 500. It definitely played flight combat games like A-10 Warthog smoother than my machine, but my Amiga still had the leg up for fast action games and realistic sound effects for a tiny fraction of the price. That’s not even getting into the fact that you couldn’t even get a proper joystick for the PC. They were all analog non self-centring style ones useless for anything but flight sims. Most PC games at the time were still played with a keyboard. The Gravis PC GamePad (the first for the PC, iirc) didn’t even come out until 1991.

          2. “As said elsewhere, 8 bit owners were looking down on CGA graphics, EGA wasn’t a huge improvement, and although VGA became available, the affordable lower end systems stuck with prior standards going into the 1990s.”

            To be fair, the graphics of the ZX Spectrum had a similar, uhm, “charme” as CGA had.

            Cyan/White/Magenta/Black were overly present in both, making you wish your monitor was monochrome only.

            Personally, I’d favor Hercules over that any time. It’s merely monochrome, but it doesn’t hurt your eyes.

          3. “As said elsewhere, 8 bit owners were looking down on CGA graphics, EGA wasn’t a huge improvement, and although VGA became available, the affordable lower end systems stuck with prior standards going into the 1990s. Monochrome was still selling for “serious” uses because of high res text and low eyestrain monitors. ”

            Another thing that comes to mind is software development. Those “boring” PCs had a huge library of compilers and development tools. And the necessary storage needed for all the project files! I wouldn’t be surprised if many of those fany 8-Bit games were being developed on PCs, or PC-like systems and through the use of cross compilers.

            Something like this happened with the Japanese Sega and Nintendo consoles, for example. The games were being developed on 1980s PC systems such as Sharp X6800, PC-98 or Windows 3.x PCs (early 90s).

            That’s akin to todays mobile app world. All those smartphone and mobile device enthusiasts talk about the post-PC era, but forget were their applications do come from – they’re being developed on PCs running programming IDE’s and Android/iOS in emulation.

        2. “but saying the Amiga was never in the technological lead is exactly the kind of historical revisionism you’re accusing other people of.”

          Sure it was, but so was Windows 95.
          Which many people have strong feelings for. In both ways.. 😥

          It maybe really depends on the point of view, I guess. 🤷‍♂️
          Sure, IBM compatible PCs were less capable, usually. And the Amiga had a more sophisticated OS than, say , the Atari ST.. It had auto-config, even, some sort of early ISA Plug&Play.

          But there also were MS-DOS compatibles (not PC clones), which the history books seem to forget.

          Systems like the Mindset, the Sirius 1/ Victor 9000, DEC Rainbow, BBC Master 512, etc.

          Then there were other platforms, like MSX/MSX2, the Acorn Archimedes, or the Japanese Sharp X68000 (Amiga rival), the PC-98 (the Japanese IBM PC).. In France, there were TO7/70, TO8, TO9..

          The American Tandy 1000 was released a whole year before the Amiga 1000 and had good graphics and sound – for a PC compatible.

          In old Europe, the Amstrad/Schneider PC1512/PC1640 from ’85/’86 were similarly popular as the Tandy 1000. The 1640 had EGA on-board, came with a monitor (were the power supply was integrated), despite being affordable.

          Many users replaced the 8086 by an V30 processor and installed a filecard or used the second 5,25″ drive bay.

          Sure, neither of them could do the arcade sounds and graphics of the Amiga or Atari ST, but these PCs were far from being niche systems.
          So yeah, it’s really hard not to be biased somehow.

      2. Did you read the article? I went to great lengths to explain how Commodore’s bad management caused the Amiga family to fall behind by the early 1990s when PCs reached that level.

        One thing the PC took several years to catch up on though was in its user experience, multitasking and a decent GUI. It would be Windows 2000-era before I had anything close.

        1. OS/2 was my solution to that when I eventually had to get a PC. Win3.x was a joke and Win 9x, while a big step forward still wasn’t what I would have called an good OS after coming from the Amiga. Before I got a WinXP PC in 2003 I was dual booting OS/2 Warp 4 and Win98SE for gaming only.

        1. I had a ZX81, too. How on earth were they doing homework or word processing on those machines? 🤷‍♂️ Did they buy those upgrade kits with chassis+mechanical keyboards? How did several pages fit in memory and how long were the loading times on cassette?

          1. word processing… why… most kids in my class didn’t have a computer and when they eventually did get a computer a few years alter, they may not even had a printer.

            Homework and such was done with ordinary pen and paper. Strangely, in 2022 lot’s of homework is still done this way. The promised future of doing everything digital is still far away. Schoolbags are still full of heavy books.

            So… the ZX81 people never really suffered from the horrible keyboards, since there wasn’t a reason to type anything else then listings (which could be a problem though…).

          2. “word processing… why… most kids in my class didn’t have a computer and when they eventually did get a computer a few years alter, they may not even had a printer.”

            Why did parents buy computers for their kids back then or supported them ?
            Wasn’t it because they thought/hoped that computers would help them learning? I don’t think their focus was on Pacman, Asteroids or Head over Heels. ;)

    3. I have owned most amiga models. Excludinh A4000T but inc A3000T.
      IMHO, the death of amiga was the 15khz
      interlaced display. To use any hires mode you had to source a hard to find high-persistence of vision monitor or an A3000 with 31khz scandouble activated.
      A3000 were very expensive but were a beast with 6830 & Math co-cpu.
      They were still a work around.
      Giving me many migraines.
      Games were in lowres to get more colors.
      Often hooked up to a TV.
      3rd party GPU cards came too late.
      But the AAA chip set would have fixed this.
      Commerwhore could of released this in the A2500 they had the design back then.
      Ali killed the amiga when he dropped development on the AAA chips.

      (C) was in cash flow crisis due to bad management. They were still selling in UK, Europe & Australia A500 bundles in truck loads. But over paid upper management & over populated middlemanagement chewed through the profits.

      A little known but very important legal battle hit (C) at the worst time.
      They were sued for unpaid licencing fees on use of the XOR Cursor. This is when the character you move to switches it’s colours.
      ie White text on Black -> Blk text on White.
      Yes this was patented & they lost the case with costs in the Millions.

      So what made an Amiga so amazing!
      Ask any amiga head what they did on it that they stopped doing on their next PC.
      Creating original content on their own.
      Or with 2 or 3 mates. The 4k demo sene was a great example.
      But it was the lone single code, graphic artist or musician who produced rather than simply consuming 3rd party offerings.

      Prior to WinXP very few kids or young adults get the thrill of showing their friends what they created from scratch.

      Amiga OS & programes was the MineCraft of the late 80’s & 90’s.
      I believe that is why MineCraft was such a fomominal success.
      It was the 1st mmo sand box.
      That enabled anyone to be creative.
      Then show the world what they made.

      So pissed I did not see it during the Doom Quake mods years.
      Andre’

  5. I used to sell computers back then. IBM compatibility was No 1 on the list of priorities for most people buying a computer. It was hard selling the Amiga or Atari ST to parents, they all had this imaginary idea that the computer would be used for serious tasks. With that in mind they wanted it to be compatible with the PC at their work.

    The only people that bought Amigas were kids that were spending their own allowance money or had managed to convince their parents that they knew best. Atari STs we’re mostly popular with aspiring musicians. For every 1 Amiga, we sold a dozen PCs.

    The 68000 processor was much better than the intel offerings at the time. The problem was simply IBM was the standard and seen as a serious computer, that alone had people following the crowd. The other platforms were all incompatible with each other an competed for a small chunk of the market, mostly home gamers that were upgrading from an 8 bit system.

    Interesting times in computing history that were great to be a part of, but there was little that could have been done to counter the move to PC compatibility.

    1. Here in the US there was another group of Amiga buyers. The video producers. I sold Amigas in those days, and many of our customers were videographers using Amigas with genlocks to make wedding or commercial videos.

      1. Loved your article. Made me dreamy of my yesteryears. Even now as I write this reply via an on screen keyboard on an Android smartphone that has as much power as all the computers I’ve ever owned or used in my 50 years of life. I still love my ancient Amiga. My old A500 ( which hasn’t worked since the late 90’s) sits on a shelf where I often tell my kids about this wonderful tool I had when I was their age, before the world wide web! I ended up buying Cloanto’s Amiga Forever Plus 10 DVD sets., Mostly for the ROMs. But use various Amiga Emulators on my old Mac Mini & a Raspberry Pi 3B to relive my glory days of Amiga heaven … I truly miss the days when I could use DMCS to rearrange sheet music, then output & play true 4 channel Stereo thru my A500 to a preamp out to a dual deck 4 track recorder. I was a budding music composer by God. Nothing I had access to in those days could do anything close to that setup. It was glorious! And using the Lightwave 3D and VideoToaster at work, running off a souped up A2000, I never saw any computer with those capabilities, in my arena, at least till the early Y2K era. I never was a hardcore geek, I never cared which Pentium was needed or fiddling with endless drivers to get some ATI Rage card or SoundBlastr card to actually work. Oh well. Peace to All, Merry Christmas. Thank you Ms. Jenny List for the wonderful article.

    1. That was no pun – at all. It could possibly be miscontrued as a dig at Gnome, but it is most reasonable to take it as the simple statement that it is – if the Amiga had succeeded, then we might all be living in a world with the Amiga Workbench on our computers instead of using Linux.

  6. I kinda missed the Amiga era by a few years. The Amiga 1000 was way out of my budget and instead I bought the much-maligned 68008 based Sinclair QL for University. At Uni most of my colleagues had 8-bit computers (typically BBC micros), but the Atari ST-520 was becoming popular in the latter half of the 80s and by the end of my Uni degree I knew at least one person with an Acorn Archimedes 320(?); another with an Amiga 1000 and another with a Mac Plus. But these were definitely in the minority.

    On the other hand, I knew very few people (basically none) who had a PC. At the University of East Anglia’s computer systems schools (SYS) we had two Mac labs (one with 512kB Macs and another with Mac IIs), which were constantly fought over; as well as the Digi-lab where a suite of Mac Plusses were used for embedded computing (hooked up to embedded 68000 boards). We did have a PC/AT lab on the same floor and a PC compatibles lab on the floor below, but hardly anyone apart from business students used them.

    My QL and its scarily dodgy tape loop micro drives got me all the way through Uni. I used the Computer One Pascal compiler to write programming assignments; added a proper 9-pin printer and monitor; and finally wrote my dissertation on the supplied Quill word-processor. By the mid-90s I’d switched to Macs, which I’m using this minute to write this comment.

    But I did rather drool over the Amigas during the late 80s and early 90s and still rank as one of the all-time awesome computers.

  7. Hmmm. In 1987/88/89, my Jr/Sr years of college, my deskside machine was a Sun workstation (2/120) purchased used for less than my roommate paid for his Amiga. And it was 5 or 6 years old at the time. Was augmented by a 68000 based desktop Xenix system in the next room that was the uucp/email server for the house. Strangely enough, built by Radio Shack, at one time the largest seller of UNIX like systems in the world. Kinda wish I had kept both of those systems.

  8. One of the really cool might-have-beens that never escaped development was a card with 4 (iirc) 68881/68882 floating point co-processors specifically intended to do serious number crunching for graphics and other numerically heavy computation. Aforementioned roommate worked on it when working @ Commodore. I’ve seen one, and it was a nice piece of engineering.

    1. I designed a terminal server demux board for the Data General MV8000 which used seven 68k processors to connect 128 terminals over Ethernet to the system. IAC-128, code name Callisto.

  9. I went from parent supplied Spectrum 48K to Amiga 1200HD when I had my own money. I almost went for an Archimedes, but real lack of anything beyond kiddie educational titles, and the reluctance of dealers to demonstrate the “PC compatible” mode stayed my wallet. I am still convinced that in late 92 that it was the most superior machine available for the money, unless you had a crystal ball and the ability to drop $3000+ on a well enough specced 486 DX2 66 to get you through to about 1996/97 .. and that extra ~$2500+ only really bought about a year extra in usefulness for daily driver computing. Still to “scrape by” with the lowest cost 486 spec available then to have lasted as long as an A1200 was useful would have been about double the cost… and Amiga or PC ain’t nobody singing the praises of 486SX25s in the retro world today.

    Amigas really came alive as a serious machine when you got a hard drive in them, the booter floppy gamer thing was difficult to overcome however. On the one hand it provide large potential user base, on the other hand, they were happy doing what they were doing.

  10. Another 68k fan here. It’s very PDP-11-like. I did a couple of designs with it, and it was easy to use. Always thought the Intel stuff looked like they slapped it together. The 68k architecture was cleaner. I have a 68030 eval board somewhere up in my attic…

  11. Although started with a C64, i have the most fond memories of my Amiga 500, which I still have around. Lately I aquired a couple of Amiga 2000, which were out of my league back them.

    I really like the simplicity of the computers back them, together with all the manuals that had the schematics in them. Nowadays everything is just a plastic blob with no info about it :D

  12. In 1990, I was working with Pixar to bring Renderman to the Amiga. Historical note: This was the moment in time Steve Job was not at Apple but at his new computer company NEXT. Jobs hated the Amiga. Saw the Amiga as competition for his plans. btw, Pixar was bankrupt at the time and Jobs bailed them out…all pre Disney. Our Amiga port was cancelled by Jobs himself….Ed Catmull was onboard but the man with the cash made the rules. Jobs went to great lengths to stop the Amiga.

  13. I started one a SuperPet at school, then got a C64 in 1982, I then got an Amiga1000 at launch and then an Amiga 2000HD.
    The problem with the Amiga comes down to a couple of things.
    1. If Commodore owned KFC they would have marketed it as warm dead chicken in a cardboard bucket.
    2. Lack of third party support. Borland I am looking at you. You announced TurboPascal for the Amiga at launch and never delivered it. Wordperfect did support the Amiga but that leads us to the next issue.
    3. Piracy. On the Amiga it was a huge problem. The big reason was the over dependence on the home market. If you can not make money you will not write software for it.

    1. 3) I a good point. Looking back, it seems to me that the Amiga users were in parts responsible for the downfall of their beloved platform.

      While piracy was a thing on all platforms, the PC users did it not as openly. They usually were more mature, also. Due to their profession or inadvertent closeness to bureaucracy/business life.

      Commodore softwars was, by contrast, literally traded at the school yard. And those scene people openly encourage piracy and cracking (crackto scene). I mean, they were young and had no interest in money. That’s understandable. There was no feel of guilt, also. So it’s understandable that things went the way they did. They didn’t know how much they harmed the reputation of the platform.

      It’s to no surprise that this must have scared of commercial software makers. Merely the good hearted or technically educated developers/publishers saw the potential of the Amiga and gave it a chance, still. That’s why many of those, which were also book writers, covered the Amiga in their bools and bundled floppies with their books.

      That being said, this is just a possible explanation. Who knows what the truth really was?

        1. Right, but it was from a completely different era.

          And personal computer didn’t mean Personal Conputer™ yet.

          That ubiquitous x86 PC from IBM wasn’t even created yet.

          This letter was written in the times of the MITS Altair 8800 and IMSAI 8080.

          This was the era of pre-commercial, private microcomputer users that were all hobbyists.
          Radio amateurs, university people, DIY fans, model makers, etc.

          These “computer enthusiasts” met in their user clubs, wrote magazines/newsletters in their tree house HQ on a typewriter. And provided all members with software, if needed – a concept which Billy didn’t really like.

          People in those “user support groups” (?) literally built all their microcomputers by hand. They made the chassis, etched PCBs, wired the transformers, soldered, programmed their EPROMs, wrote their own floppy routines for CP/M, ported programs from one Basic dialect to another etc.

          Microchips were still thought of as toys by the industry, at the time.
          Merely a little group (by comparison) of people believed in microcomputers at the time.
          They were pioneers.

          Last but not least, it also was the hippie era.
          The idea of sharing all kinds of things was still strong.

          This was a totally different world compared to, say, 1985.

        2. “It might have started in one era. That doesn’t mean it ended in that era either.”

          You completely miss my point, I’m afraid.

          a) In the 1970s, there was almost no commercial software for private computers. A market didn’t exist.
          Many software writers worked at university, in which sharing information was normal. Or wrote software themselves at home. There was defacto no copyright issue, as software was treated as “Public Domain” by default. The mainframe world was different maybe, not sure.

          Microsoft, by contrast, started writing BASIC interpreters for microcomputers as a living. This was new at the time! They were pioneers, too. The hobbyists didn’t realize this new business field and continued doing their things as usual. Hence the letter!

          b) Those early “PCs” in the 1970s were the forerunners to both IBM PCs and home computers.

          It thus makes no sense to instrumentalize that letter against users of x86 PCs. 😂

      1. It was more about home use vs work use. A lot of people pirated PC software to use at home after it was bought for the office. With the Amiga there was no buying for the office. Commodore also didn’t push the Amiga for business. After the PC came out Commodore just gave up on that market. Apple had a good presence in the Business market before the PC and kept it for a while until the Mac came out. The C64 had a huge fault for business software, the disk drive system. Not only was it super slow but it lacked the Unix/CP/M/MSDOS seek command. The Pet had faster drives but still lacked the seek. That being said a lot of some businesses used PETs and C64s
        What Commodore should have done IMHO to make the Amiga a much bigger player would to have been.
        1. Include a hard driver interface on the 1000.
        2. Put enough ram sockets to take it to 1MB or more.
        3. Put a 422 interface on the machine so you could do an apple talk like network with it.
        4. Market it to businesses and developers.
        That being said the Amiga was not really a failure. It was around for about a decade and is well remembered to this day. A pretty good record when you look at all the other systems that PC killed very quickly.

        1. Yes. I knew someone with a Mac Classic, and I assumed the software was copies, because he hadacarefully photocopied manual.

          And he made films, I’m sure he wouldn’t want illegal copies.

      2. I always treat publishers whingeing about piracy with a grain of salt. A so-called “lost sale” is only that if the person would have spent the money otherwise in the first place. I suspect there was a similar overstatement of “damages” as we see with the record industry during the whole mp3 debacle. Media publishers of all stripes like to pretend the axiom “you can’t get blood from a stone” isn’t applicable and every pirated copy somehow represents a real loss, which is of course nonsense.

        An anecdote isn’t data but for what it’s worth I feel like I was a pretty typical lower middle-class teen with a part time job in the Amiga days and about 2/3rds of my games were purchased copies and the other 1/3rd were pirated copies. Now applications were different, nearly all pirated but honestly what teen has hundreds of dollars for a legit copy of Lotus 1-2-3 or Deluxe Paint IV? They were never going to be getting that money from me piracy or not. If I couldn’t find a pirated copy I would have certainly just gone without. There’s no way any amount of marketing could have convinced 14 year old me to forgo buying 3 or 4 games to save up for a word processing program.

        1. “I always treat publishers whingeing about piracy with a grain of salt. A so-called “lost sale” is only that if the person would have spent the money otherwise in the first place.”

          I feel the same. It’s a bit arrogant to assume that less profit equals loss.

          This logic would make every mathematician want to tear his/her hair out. 😂

          The products aren’t gone, after all. They’re still in the warehouse and might be sold later on.

          Or the other way round, the targeted user group would never pay for it, anyway. So it’s no loss, either. Maybe the users would never have spent for a miserable piece of software, anyway. So it’s no loss, either.

          “Now applications were different, nearly all pirated but honestly what teen has hundreds of dollars for a legit copy of Lotus 1-2-3 or Deluxe Paint IV? They were never going to be getting that money from me piracy or not.”

          Exactly. And business users had to pay the software, anyway. At least for workplace use. Because, a police raid could happen any time.

          My father once told me about it, by the way.
          There was a police raid at his university, because the PC users had used pirated copies of Norton Commander. It’s a true story, it happened in the 80s.

          That’s why he was so relieved that he had owned an original copy of Norton Commander all along.

          Things like this can happen at the workplace, but rarely do at home. Home users aren’t commercial users, they’re private users. Unless there’s a connection to the commercial world.

          Also, in some countries, say Germany, it used to be legal to make and own a few private copies.
          For yourself (working copy, say boot up disk for DOS), for a family member and a close friend (which are like family members).

          That’s something the history books do forget, it seems. If you had a brother or sister, it was legal to share the games or applications you bought with him/her. He/she could use all the duplicates of yours. Again, in some countries..

          VHS was a similar story, by the way. Blank VHS had taxes on them, so it was legal to lend a movie, make a copy and give the original back. Because, you paid. Twice. The taxes which were part of the blank medium and the fee you paid for lending the VHS. Unfortunately, the movie industry didn’t approve and called you a thief/pirate. Even though you paid, more than once. Silly world. 😂

          1. Ah yes, we had a similar solution here in Canada. The government answer here to the mp3 issue was to establish a percentage “media levy” charged on all blank digital media (CD-R/RWs, flash storage, hard drives, etc) that funnelled a small amount of money from every sale to the recording industry. Much more sensible than the US system where grandmas were sued for hundreds of thousands in “damages” because their grandkids downloaded some songs while they were over for a visit.

  14. Back in the eighties I programmed an Amiga for a nationwide, live radio show where the DJ had “conversations” with the computer. I used the built-in speech synthesis, which seemed on par with the late Stephen Hawking’s — only it had to be programmed phonetically sometimes for better “performance”.

    The only problem was that my program had an edit mode and a play mode and in my ignorance — assuming that all conversations would be pre-scripted — I had it say “Hello there” as an indication it had changed to play mode, which, of course, ended up on air when the DJ wanted to change the computer’s response (passed on pieces of scrap paper) while he was talking.

    Ah, those were “the days”!

  15. I missed out on the Amiga; I recall admiring animations that’d been made with it, and seriously lusting after the Video Toaster. My family’s first machine was the classic TRS-80 Model 1; the only drawback was that we got it in 1987, when it was 10 years old. (We got an Apple IIgs a year later, and I switched to x86 machines in ’94 or so.)
    But the computer I remember with the most rose-colored glasses nostalgia wasn’t mine, it was a school computer my 8th grade math teacher had — a NeXT Cube. Physically a thing of beauty, and I recall doing something with cellular automatons and fractals, and how those just flew; looking back, they must have been in Mathematica.

    1. I worked as a higher education marketing consultant for Commodore for a while, during the A3000/CDTV era (the article skips CDTV, which was groundbreaking, and goes straight to CD32, which faced a market with other CD-ROM consoles). NewTek releasing the Video Toaster created a freelance job market for me setting up TV studios. I have an enduring soft spot for all things Amiga. Still, the NeXT cube remains the sexiest computer of that age.

      1. I skipped the CDTV because this isn’t a history of Commodore but a “What went wrong” analysis. It was a cool product, but not relevant to what went wrong. I have covered it before, as it was hugely relevant to my first career in the CD-ROM multimedia business.

    2. In the mid 1980s, my parents had a couple refurbished TRS80 Micro 3 computers. As kids we’d learn Basic, so we could play Oregon Trail or write our school reports, I remember it had 48k Rom, a green screen, 2x 5.25″ floppy drives. That’s what I thought computers were – big clunky, AIO units to write word processing, do cataloging. Nothing for fun & games. To play games you’d play on an Atari, or a Colecovision… Then in summer of 1989, I was a Junior in HS, my best friend had this whiz bang computer called an Amiga 500! I was blown away – it could do graphics, video, audio, music, games like nothing I was used to. I wanted one. I saved every dollar, my allowance, lawn mowing cash, etc. Bought my own Amiga 500 with the 13″ NTSC monitor, it came with AmigaBasic which was a breeze, I loved even the clunky CLI and Workbench 1.3 gui. Soon I was buying every extra I could scrimp for – the 512kb expansion with real time clock, 2x external 3.5″ dsdd 880k disk drives ( I loathed the endless disk swap), a general midi box so I could hook it up to my Casio synthesizer. My 2 top programs I loved were Deluxe Music Construction Set, & Deluxe Paint III. By the time I graduated HS, I was working thru Trade school for Cable TV production. At work they had several A500’s and A2000’s, one with a Video Toaster. These were serious machines. As I owned my own A500, I had a leg up as a user vs all my older & wiser coworkers with their crap 286 setups. I used all the gear and became a video engineer before I was 20. What I hated too as others pointed out – lack of real storage & lack of CD support. I personally could afford the acceleration & HD, or extra Ram offerings that one could get from the local computer store, like a HD/RAM sidecar expander for my A500. It was sad for me as 1994, Commodore was defunct, and only places I could get Amiga parts was from the UK or Europe. I was still a young techie with a couple college degrees in tow, working in cable tv. I ended buying my 1st Mac in 1995, a IIci. No where as cool as my Amiga. Thanks for the great blog, some fun responses, a trip down memory lane…

  16. The 68000 is a big-endian CPU, and having programmed in assembler for everything from microcontrollers to IBM z/Series computers (my very first, hands-on programming experience was System/360 assembler on S/360 Model 30 with a whopping 16kB core memory, two 5MB DASDs (Direct Attached Storage Devices, IBM-speak for harddrives) and seven — or was it five? Argh time interferes with memory — tape drives, for a major mortgage institute … sorry for the digression), I find big-endian slightly less intuitive when programming than little-endian.

    1. I thought big-endian was weird when programming in Basic and having to process the upper byte first. However, when looking at a stream (or memory dump) of hex digits, big-endian is nice because multi-byte numbers and addresses appear correctly. In order for the little-endian Vax assembly listing to show addresses correctly, it wrote the bytes in right-to-left order in the print-out.

  17. Hey Jenny. Don’t forget that I have -your- A1200 here. All cleaned up, refurbed, upgraded OS, Goetek and Compact Flash. I’m only keeping her warm for you.

      1. Yeah you are right. I’ll flog it on eBay! Seriously though, you get her back when passing next.

        You opened my eyes to the Amiga and it’s one of my favourite cores on my MisterFPGA. (Which I even bought a tank mouse for).

  18. True, but for me the thing was that I had no problem reversing the long numbers (as in longer than could be specified directly in assembly language, which often had BYTE, WORD, DWORD and QWORD, handling 8, 16, 32 and 64 bits respectively) and addresses were specified with labels in the assembler, so no problem here either. I rarely needed memory dumps and debuggers, so that did not present a major problem either.

    The big-endian problem is addition and subtraction with numbers longer than the word length of the CPU where you would have to offset the start position of the operation from the address of the number. On the other hand, sorting by direct comparison was easier as the little-endian CPUs are the ones needing offsetting for this, but I must have done more addition/subtraction than sorting.

    My preference for little-endian is probably due to working mostly with 8-bit systems in the beginning. With 8-bit systems it was natural to use little-endian for long calculations and comparisons were often done by subtracting the numbers to compare and see if the result was negative, zero or positive.

    That said, I miss the 68000 family as it had a consistent and orthogonal architecture with few surprises.

    1. The predecessors to the 68K, the 6800, and 6809 were both big-endian, as was the TMS 9900 so they were used in real computers. There are some architecture advantages in supporting big-endian processing. For example, if you’re reading a 32-bit value through an 8-bit data bus, then on each step you perform:

      val =(val <<8)|newByte;

      On a little-endian machine you need a separate counter:

      val = (newByte<<(byteCount++))|val;

      Big-endian is also easier for accumulating variable-length data, where some part of the data itself indicates there are more bytes to come (e.g. an 8/16 bit instruction set or MIDI files).

      1. @Julian Skidmore (Using this as the Reply thingie so far has not worked well for me),

        Your comment about reading 32-bit values over an 8-bit data bus is spot on, and is probably the main reason that network order normally is big-endian to save precious time when loading addresses.

        For sure, big-endian is in general the best way for anything but mathematical operations on numbers longer than the word length, but my first experiences with 8-bit systems were mostly math-related (IEEE 754 floating-point libraries, FFT routines, and such), which meant little-endian was intuitively the way to go, even on 680X systems that used big-endian addresses.

        When at university I programmed most of the first commercial microprocessors like 4004, 8008, 8080, 8088/8086, Z80, 6800, 6809, 680XX, 6502, 1802, NS160XX and odd ones like the AMD2900 bit slice family as well as the one-bit (yup!) Motorola MC14500 plus others I have forgotten now, so I have been exposed to a mix of big- and little-endian systems, but still have a slight preference for little-endian, not that it matters any more as I haven’t done any real assembly programming this millennium.

        1. Cool. Looks like you have some great experience. I have no experience of the 4004 nor 8008, 8080, 6800, 1802, NS160xx, AMD2900 (not to be confused with the AMD29K ;-) ). So, apologies if my previous comment was patronising.

          My first 8-bitters were also Little-endian as I started with Sinclair’s 8-bit Z80 machines, but then I moved to the 68008 based QL followed by Macintosh (though they’re little-endian since the Intel transition!).

          I’m not sure if I prefer one or the other. I find little-endian listings harder to read – a bit trickier for debugging (e.g. on Cortex M0+s which I often do now), but there’s a real computational elegance about little-endian too, e.g:

          uint32_t x=49152;
          uin16_t y=*(uint16_t*)&x;
          y=49152.

          Another aspect of course is that big-endian maths is an artefact of how we adopted Arabic numerals. In Arabic of course they’re written right-to-left, in little-endian order, so for Western scripts, I find that when I read a large number I have to skip to the end to find out how many digits there are and then skip back to start getting a sense of the magnitude of the number. But for Arabic readers, they start off with the least significant digit and so only need to scan forwards to gain a sense of the magnitude. So, after nearly 800 years of having simply transcribed Arabic numerals, we tend to thing that big-endian is a natural way to express numbers, but in fact we would have been better off if we’d switched the order.

          e.g:
          سرعة الضوء 299792458 متر في الثانية.

          (the speed of light is 299792458 metres per second via Google translate).

  19. REBOOT. What we really need now is a reboot of Amiga. How? Are you serious? What kind of drugs are you taking? Yes, I am serious! No drugs involved! And the how is easier than you imagine! FPGA PCI development board for PC. 1st we implement a 68060 Amiga on the FPGA. We create bridge using a dev board to access modern graphics, sound and open source code. We embrace the concept that 32/64 bit architecture isn’t needed for everything. Remember RISC? 68K ode can accomplish a lot! Sometimes we can accomplish more with 32/64 bit code. Complex graphics can truly benefit from the more complex architecture. Bridges to those architectures can augment. Algorithms can help interpolate, scale back and extrapolate as needed. The flexible, scaleable, simple architecture technology, FSSAT or FAST will be the way of the future assigning functions based on reducing complexity intelligently and assigning them smartly to achieving the ultimate efficiency and design.

    1. I stumbled on Jenny List’s thoughtful, “Love Letter” and reader-comments while searching for the pinout of my Amiga 2000HD’s serial port.

      A few of the less cash-strapped members of the Baltimore Amiga user group (run by Ed Hopper in the late ’80s / early ’90s) purchased subsets of the full RKM and hardware library. These were treasured items shared among the user group like sacred scrolls. It is good to see those books are now freely available. Much thanks, ytrewq, for notice the RKMs are on the Internet Archive!

      Also available on the Internet are Aminet’s CDs ( https://archive.org/details/aminetcd ), and Aminet’s archive of Amiga software ( http://aminet.net/ ).

      imqqmi’s comment, “There was a mysteriousness about the Amiga, anything seemed possible, you would look forward to the next game, demo, disk magazine, graphics application etc and whatever popped up at the local Amiga club. It was a time of excitement, discovery and experimentation that I never since have experienced on any other emerging platform.” sums it up nicely.

      While Atari was doing all possible to hide the hardware and software interfaces of its products, and Apple required a formal application (read, “pretty please, allow me to kiss Job’s ring”) with a painful “application fee” to possibly be accepted as an Apple software developer, Commodore welcomed anyone and everyone to develop hardware and software for the Amiga. Users could easily register as “Developers” with Commodore for a nominal (10 USD) fee, and afterward receive a monthly package of technical documents with one or two 3.5 inch “floppy” diskettes. The technical documents were focused examples and tips for interfacing with the libraries (e.g. text-to-speech), and the diskettes included reference code and fun software (e.g. a new game). This was a day vs. night difference between Commodore and the other retail computer manufacturers, and no doubt greatly contributed to the Amiga’s longevity – extending now decades after Commodore’s collapse.

      For persons beginning electrical / computer engineering and computer science studies, the Amiga provided an in-home lab for augmenting and reinforcing material presented in the classrooms – and it was also a hell of a lot of fun just to play with.

  20. I still code for the old tub and my A1200(HD) is connected up to test code which is actually written on FS[-]UAE, on both APPLE OSX 10.5.7, Linux Mint 20.3, with a clone of my A1200 setup…
    AMINET is the second oldest _repository_ predated by APPLE’s Info-Mac…
    To this day, I am still in love with the AMIGA OS…

    1. But the PCJr. and Tandy 1000 were already available in 1984, prior to the release of both Amiga/Atari ST. 🤷‍♂️

      And in Europe, the Amstrad/Schneider PC 1512/PC 1640 were quite successful, too. The later 1640 model was released in 1986 and had EGA on-board, too.

      To later PC clone owners, VGA cards and AdLib/Sound Blaster clones were not unheard of in 1990.

  21. I had an Acorn RiscPC. Awesome performance but doomed again by poor management concentrating only on education market and forgetting the rest and the awesome community that supported it. Sigh, at least we got ARM out of it…

    1. I didn’t even “get” that strategy at the time. I mean yah, they’d done well in that sector with the BBC B and Master, but then they made this much more powerful machine and a lot of stuff seemed aimed at primary level still with hangman type spelling games and “math invaders” type stuff that had more bells and whistles, but was handled adequately in function by the 8 bits the schools already owned, while kids were still enchanted enough by using a computer at all. I mean now, kids have seen/had tablets before school even, so sophistication really needed to get them engaged, but back then it was wasted. Much of the software seemed like stuff you could hammer out in BASIC in a few hours. (For equivalent functionality on 8 bits) So they mostly seemed to be competing against their prior products, at greater expense while admin thinking those were assets depreciated over 10 years with maybe 6 years still on the clock.

      IMO they should have at least pitched more at secondary/teen level. Snuck in an inoffensive game with the software pack that might have the kids asking their parents for one for home.

      Also if you saw those going into use in schools at that time, it was the Masters and Bs with the CP/M and drives that got bumped up to older students for “serious” CP/M stuff, even though that ship had sailed and the younger ones getting bells and whistles on Archimedes, and it was that same feeling as having the programming lab stuck on first generation i3 machines and seeing new i9s going to the library for catalog lookups. (When at college I experienced it as 286es in the lab and P90s in the library)

      Anyway, yes, a very capable machine, that at least we got ARM architecture from, but in choosing for myself at the time, I could see it going like the QL, mainly user supported for home users.

      1. About 20 years ago I was in Steve Furber’s research group at Manchester University. Engineers and hardware designers at Acorn were super-frustrated by the lack of vision at Acorn and it’s inability to market the Acorn Risc Machine in its own right (it’d been bought out by Olivetti by then). It took Apple’s Newton project to get ARM spun-off and ultimately outlive its parent company.

  22. PC sold because many people would borrow the software installation disks from work to load at home. Employers didn’t mind if people did more work at home. Microsoft didn’t mind because every PC included a paid-for OS license. The main software makers were irritated, but people who pirate generally won’t pay no matter what. It wasn’t until Microsoft produced Office that anti-pirating measures were put in place.

  23. I have a historically significant amiga. It is an A1000, and it is the one they submitted to UL for approval. I think it is serial number 17. An interesting tidbit is it was missing the ram expansion cover when I got it, and I never thought much about it, but it turns out that at that point in the game they were not sure if they were going to go with the check logo or the Amiga logo on it, so there is a good chance it was not missing, as much as it the design was not committed. Kind of a neat piece.

  24. Well the Amiga and community is still kicking. MorphOS, CoffinOS, System One OS, and PiMiga has all taken original and new hardware to its evolution. For 2022 there were more than 30 new hardware options for purchase, which makes it the craziest year ever for the old systems. The prices for all retro gear is insane. I hope we can keep up the momentum as many of the original users are between 50 and 80 years of age. We are the cockroaches of the computer world, but with any luck 2023 we’ll get the pi storms for the 1200/3000/4000. finishing up the cheapest way to push original hardware to where the new ones are. The Zulu scsi fixed the issues with the sd2scsi. All my Amiga have IDE in them now. All my Amiga are HDMI out now. The secret file for the Plipbox is known now, so getting the Amiga on the net is easy. How many new games where there this year? So many. Its great time to be an Amiga User.

  25. |   “So, apologies if my previous comment was patronising.”

    Don’t worry, I’ve been around the block a few times or more so absolutely no offence taken.

    |   uint32_t x=49152;
    |   uin16_t y=*(uint16_t*)&x;
    |   y=49152.

    Isn’t it lovely how you can make super elegant C code that breaks for no good reason when transferred to another architecture. C is probably the first language that came with a virtual “gun” permanently aimed at your knees, not your feet. I cannot tell you of the horrors I’ve seen … probably only surpassed by modern day Java programs, not because Java is inherently bad but rather because it is often badly used and severely abused.

    I gather that C, with it’s lack of trainer wheels and guide rails, is probably the biggest cause of modern-day security problems — and I love C, my favourite programming language together with Perl. No sarcasm implied or intended.

    |   via Google translate

    And here I was very impressed with your Arabic then you write this. Tsk! 😇

    And thank you for the reminder about the origin of our number use. It is rather impressive that we are still going the wrong way about something invented several thousand years ago, but I’ve sort of got used to it now, so please don’t change it for my sake.

  26. A lot of these comments present it as a feud. But in the eighties, there was still room for diversity. Byte was still covering computers that weren’t IBM compatible. The Atari, the Amiga, the Radio Shack Color Computer, Unix boxes, 68000 systems, Macs. The dominance of IBM type computers took time. It wasn’t necessarily one was better than the other, but at some point you needed to be compatible.

  27. I had a black satin jacket with my name embroidered on the breast, a large “Boing” ball embroidered on the back and ACES stitched across the shoulders. I still have that jacket it’s a couple sizes too small for me, stashed in a closet. Next to a couple A500’s an A2000 two basic monitors and at least 2large suitcases full of every software imaginable donated by ACES computer club members who got out a year or so after commodore failed. Earlier a comment was made about creating content, that’s what held us together and I miss those days and what that jacket represented. We were a gang of nerds with the baddest machine on the planet.

  28. I think for me Commodore was more a cultural icon, a milepost in life, rather than (as witnessed in the reply thread) just a box of stat’s. I started behind the power curve with used pets and 64’s before I’d finally gotten a New Amiga 500.
    I had no great knowledge of programming, fixing 18 wheeler flats for a living, and having a quite limited semi-rural life. My 500 was a game machine, (thanks to Sid Meyer, mostly) prior to a quite accidental meeting with a local Amiga BBS sysop.
    That changed my life – totally – and set me on a very different path.
    Once my first fido-net message from the UK arrived, I was permanently addicted – with no turning back.

    We centered our social lives, our spare time, and our friends around Commodore. Electronics pros, Ham radio groups, students, retail and blue collar folks – all you needed was a passion for the box. Endless discussions, and whole new vistas appeared as we dreamed of what could be.

    Cracks in that vision did begin to appear soon however. We wound up going to specialty stores that had to order each unit, offered very little serous support. PC’s began to get mainstream – but hey, we emulated PC’s online – and became the elite hacker choice of machine as a result.
    Our lives – the culture of home then business computing had its roots in the (for me commie) machines, software and personality’s , for me, changed everything.

    If you would have told me I’d have made a great life and living for myself and my family with those experiences (OMG a 40 meg hard drive! I’ll never fill it up!!!) – no offence – I’d think you were “touched” and not pay for your pizza at the meeting if you were short.
    What the Amiga / Commie meant to me went WAY beyond the box. It was escape from the lifepath I so desperately wanted to avoid. A tool to accomplish the academic achievements, and learning I thrived on. A circle of lifelong friends, memory’s, and ex-wifes, and Life.
    Yes, I was in pain after the company folded. I moved on. But I really can’t forget how it changed all of us, PC/Commie/MAC alike. It was all I could afford (barely at that) and it payed me back a million plus times over in life. Heck – i’m a bit misty now. But not just the box, the whole experience.

  29. I remember the Amiga 500 with great fondness. It had sound and graphics that blew everything else away. I got into ray tracing with Imagine. I even added a Mega Midget Racer daughter board to speed it up to 30mhz which substantially lowered ray tracing times.

  30. Funny that nobody mentioned BeOS while traveling down the memory lane. It was a fresh breeze and highly productive OS with a nice GUI, just like the Amiga. I remember how snappy it felt the first time I tried it and compared to Windows 95, very responsive even during program startups and heavy calculations.

Leave a Reply to NQCancel 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.