The Commodore 64 before it was way ahead in graphics and sound for its era, especially for the price. PC compatibles didn’t even catch up to that 8-bit machine until EGA and Adlib and Sound Blaster cards came around.
By contrast the Apple Macintosh hardware architecture was much more limited. But it was also easier to extend without redesigning everything from scratch.
That's interesting. If you have a moment could you expand on that? Thanks.
It's just that Commodore leadership was fundamentally extractive, and saw personal computers more as consumer widgets than platforms. The financial side of things was a bit shambolic, and they trusted the wrong money.
It still lacked protected memory and preemptive multitasking.
The Amiga OS still had far better multi tasking abilities than Classic Mac OS
"Commodore was one of the Atlantic subsidiaries directly implicated in this scheme, but the commission was unable to find any evidence of wrongdoing on the part of Tramiel or Kapp despite heavy suspicion."
Commodore founder Tramiel didnt care one bit if he sold typewriters, calculators or cow hides as long as he could make money by screwing people over https://www.filfre.net/2015/04/the-68000-wars-part-3-we-made... The moment C-64 was released whole* (except one person) design team was "let go" and subsequently founded Ensoniq. Tramiel replacements were even worse because on top of not knowing a first thing about product they also didnt know how to make money.
By the late 1980s, colour monitors were nearly as good as a B&W one for showing crisp text and lines, and so grey/black on white.
When I visited one of my dads business associates who ran a software company, for example, the vast majority of PC's in their offices had monochrome text screens. A handful had EGA graphics cards and colour screens. I remember playing Zaxxon on one of them and finding it hysterical that the graphics and sound on their most expensive machines was worse even than on my C64.
The Amiga looked like sci-fi technology in comparison with what most people actually had in their offices or homes until several years in. Ironically if anything the significantly better graphics contributed to it being pigeonholed as a game machine despite the original Amiga looking like an office machine and being way too expensive for games.
The problem was not so much the original or enhanced graphics chips, which were groundbreaking in 85 and maybe even until ~'88 or so, but that AGA in 92 was way too little, several years too late as an upgrade to still justify the price premium on the Amiga. 7 years at that time was an eternity for the graphics capabilities to stand still.
Had AGA started a couple of years earlier, maybe Commodore would have lived long enough to complete their next two upgrade projects (AAA[1] and Hombre[2]), but with the chaos of Commodore at that time it's no surprise they didn't pull that off. I feel like Commodore was still in the 8-bit mindset of thinking they had time to milk their existing designs for much longer than they actually could.
[1] https://en.wikipedia.org/wiki/Advanced_Amiga_Architecture_ch...
Hombre, their last attempt, was one one hand meant to provide a clean break by moving to a PA-RISC core (chosen to allow for Windows NT compatibility but in retrospect might just have deferred their death) in a dual-chip pair that'd combine the CPU, DMA, an upgraded blitter that was closer to a basic GPU (transforms, texture mapping, gouraud shading, z-buffering), a 64-bit RISC-like copper replacement, 16-bit sound, with a display controller on the other chip.
But to avoid breaking backwards compatibility, they also planned putting an entire AGA Amiga on a chip, and recompile AmigaOS, so they could end up with sort-of reverse of what we got with the PowerPC expansion boards or the older Janus Bridgeboards (putting an x86 CPU straddling both a Zorro and ISA buses, letting you run PC software at full speed in a window on the Amiga), of having the new CPU run the OS, and using the embedded 68k Amiga-on-a-chip to run legacy software.
A valiant attempt, but of course way too late - Commodore ran out of money before completing even the chipset before Hombre in their pipeline.
Hombre would've been a big break, though, with what looks like they had something resembling an actual plan for a very broad range of uses from the start instead of their usual haphazard Darwinist "survival of whatever nearly ready design still happens to please the right executives" Commodore usually practices.
It ranged from putting it on a high end graphics card, to set-top boxes, CD32 like entertainment systems, all the way up to high-end machines running AmigaOS and WindowsNT and based around a machine concept from David Haynie (Acutiator, going back to '91) that planned for exchangable modules for everything including the CPU, with the ability to even adapt the bus control account for different CPU requirements.
It's an agonisingly fascinating "what if" of alternative hardware, while at the same time in retrospect having some serious flaws (PA-RISC...) and probably being way too ambitious for Commodore even at its peak given the amount of discipline shepherding a project that ambitious over the line without some executive ego trip derailing it would've taken...
By 1992, the 486 had been out for a few years. 386's with SVGA and soundblaster cards were getting pretty cheap. The writing was on the wall...
Even then we hoped for a resurrection because it still seemed possible - I remember we tried pitching a software dev. project to Escom in '95, shortly after they'd bought the Amiga rights. I'd co-founded an ISP, and another Amiga fan among my co-founders and I were considering developing an alternative Amiga web browser back when it was still viable for tiny teams to do so. But of course Escom went bankrupt a year later, before we'd gotten anywhere with them.
> If they had put AGA in the A3000 (1990)
Yeah, I agree. The A3000+, in other words[1], only 2 years before AGA. Such a missed opportunity even in '91/'92 - it was faster than than the A4000 (better memory bus, SCSI instead of IDE), and had a DSP for 16-bit audio recording and playback or a modem.... And of course in being designed for the A3000 case it would've been oh-so-much prettier than the A4000 bulky PC style case. In classic Commodore style axed without a rational reason.
That two year difference would've at a minimum given them the reprieve to possibly get AAA out (plagued by problems paying for test runs through '93)...
What could have been... I still wish for the feeling of a modern "Amiga". I contributed code to AROS for a while, but it's "too Amiga" in the sense that it accurately reflects AmigaOS 4.x for the most part, but what I really want is something that matched a modernised version of my rose-tinted nostalgic view of Amiga, and I don't know if that's possible, but I occasionally wonder if a Linux-based AmigaOS-inspired desktop environment might get me there... (of course, finding time to do anything about that is another matter...)
[1] http://www.bambi-amiga.co.uk/amigahistory/prototypes/a3000pl...
Given that reputation, when I finally got WinUAE setup, I couldn't believe how awful Workbench 1.0 looked. "Well", I told myself, "I guess it was from the mid-80s. You can only expect so much. It's not bad compared to Windows 1.0..."
So...you can imagine my dismay when Workbench 2.0 (1990) and 3.0 (1992) looked...it wasn't garish anymore; it was incredibly drab, but still awful. Compared to their contemporaries - Windows 3.0 and 3.1, respectively, to say nothing of MacOS at the time - there was no contest. If you were doing anything besides playing games or video work with the Toaster I couldn't imagine buying an Amiga at that point, given what else was on the market.