Fickle Fate and the Amiga Inferiority Complex

You know what, it's been a while since I ranted about things in the Amiga community that annoy me, and this has been eating at me for a while, but with the latest news concerning *ahem* a certain remarkably-priced custom board, this has been coming up a lot: the "can't beat 'em, join 'em" argument.

The version I've seen, specifically, goes like so: "PowerPC is a dead architecture and can never hope to compete with x86 for performance, therefore all PowerPC-based Amigoid projects are futile and must be opposed, as the only possible future is to move to x86." But while I wasn't a member of the community back in the mid-'90s, I would bet cash money that there was a parallel argument regarding the then-moribund 68k architecture versus the thrilling new future of RISC computing, PowerPC :/

Now, in the interest of full disclosure, I don't actually have much interest in any of the "Big Three" next-gen Amigoid projects, so I don't feel a need to rush to the defense of any of them, and while PPC as an architecture does interest me, it's in the potential of used Macs as nice moderately-powerful Linux boxes, rather than anything Amiga-like on any of the custom boards. And I have nothing at all against x86 for general-purpose computing, though I find it moderately gross under the hood. It's janky, but it's cheap and has the full weight of the industry devoted to making it currently the undisputed most powerful desktop architecture out there.

But what really bugs me about this line of argument is that A. it has underlying it some incredibly broad, sweeping assumptions about what the Amiga should be, and B. it is always, always uttered in a stark, imperious "you're an idiot if you don't agree with me" prescriptive tone. So let's take a look at this.

The presumptions.

The first presumption that underlies this argument is this: that the "new Amiga" (whatever you believe it to be) must be competitive, on a performance level, with modern PC hardware. This has as its source a deep, crippling inferiority complex that an alarming number of people in this community seem to suffer from. They feel, apparently, that since the Amiga totally could have destroyed its competition and overtaken the computer market (theoretically, possibly, in a scenario where Commodore corporate had any idea what it was doing,) then any "new Amiga" project must aim to be the "PC-killer" they've been dreaming of, and the only way for it to do that, apparently, is to be on the bleeding edge of performance. But since the only desktop architecture that currently is on the bleeding edge of performance is x86, they're faced with a conundrum: give up the dreams of PC-slaying for a more realistic goal, or become the enemy in order to destroy it? It seems, apparently, that they choose the latter.

Never mind, apparently, that for the past four years, the biggest growth sector in the computer industry has been the one with the least horsepower: smartphones and similar mobile devices. Now, I'm no Wired magazine prognosticator who thinks that everything is going to be a tablet in the future (far from it, but that's another rant,) but what I do gather from that is that for apparently a very large number of people, a piddly little 700MHz ARMv7 CPU is enough (or near enough) for their basic computing needs. Maybe not all their needs, and certainly there are people who really do need the raw horsepower of an i7, but it seems we've reached a saturation point where we actually do have enough computing power. And if a <1GHz ARM chip is it, then apparently we actually reached that point years ago and just never noticed.

Of course, that also implies that the key to being a widespread success is something other than having the biggest, baddest hardware around, but that point seems to have escaped the x86 fanatics entirely.

The second presumption is a bit less immediately apparent, but no less troubling: that x86 is and will always be the best desktop architecture. Now, I don't think they think this out of any bizarre loyalty to Intel; rather, I think this is another example of people assuming that the future will be exactly like the present because the present is all they really know.

But let's take a look back. In 1985, when the Amiga came out, x86 was a joke. The 8086 was already seven years old, a 16-bit chip with an ugly kludge (segment registers) to adapt it to a 20-bit address bus, and the 286 was only barely better, because it only added a proper MMU, doing nothing to address the fundamental ugliness, and even that feature was locked away in a protected mode that PC BIOSes were incompatible with, which meant that almost nobody even used it. The 386 was a more sane, not-so-terrible design, but it had only just come out and was in use as a high-end workstation chip, and wouldn't gain mainstream acceptance for at least another five years.

The 68000, on the other hand, was already an established name in Unix workstations and had made inroads into the personal-computer market with the Lisa and Macintosh, and with the 68020 out it was even properly, fully 32-bit! Not only that, but the architecture was much friendlier, in that it could natively handle addresses the full size of its address bus and thus didn't have any segment-register stuff to make you want to slit your wrists.

(Oh, and the folks behind the BBC Micro were working on something called an "Acorn RISC Machine," but ha ha, really, like a handful of systems guys were going to design a whole CPU? Get real.)

The point behind this historical ramble being, fortunes change. If a savvy nerd back then were asked to predict which of those three architectures would power the computer of the future? They'd probably pick the 68020. They might pick ARM, if they'd heard of it, and if the shenanigans it required of programmers in order to enforce single-word instructions weren't off-putting. But they sure as hell wouldn't pick x86. And while we can look back now and say "ha ha, whaddya know," it's worth considering, before you pin all your hopes and dreams on Intel's cash cow, that there may just be another strange turn of events somewhere down the road.

(In fact, I'm sure there is, and I'm not the only one. Microsoft is taking steps to make Windows 8 truly multi-architecture, the first time they've ever done this in the general consumer-computing market. And when a creaky corpo-bureacracy like Microsoft takes such a sudden turn, you better believe there's some seismic activity-caliber shifts about to take place.)

The attitude.

Honestly, there's less to say about this than about the assumptions being made, but just so it's said: you are not the ultimate arbiter of what "Amiga" means. Get over it. Of course we all have our opinions, and of course we all believe we're right, or at least in the general vicinity of being right, or what would be the point in having them? But if you've somehow got it into your head that you know How Things Must Go, and that this is The Only Way, you need to step back, take a chill pill, and come back when you're ready to argue less pompously and more zealously.

So what, then?

I don't know. Maybe x86 is the future. Maybe not. Maybe what is or isn't the future isn't important, and next-gen Amiga projects should adopt whatever architecture they feel best reflects their own philosophy. (I'm all for this.) Maybe they should just abandon architecture-dependence and write fully portable code. (I can see the logic there, though I find the prospect less intriguing.)

Whatever the case, I do not for a minute believe that the true future of the Amiga lies in resigning yourself to doing what the rest of the industry is doing just because the rest of the industry is doing it. If you want to see something on x86 because you like the cost-effectiveness or the raw horsepower, that's A-okay. If you want to see something on x86 because you've decided nothing else can ever hope to compete numbers-wise, please, go watch a heartwarming Christmas special to breathe a little spark of life back into the withered husk that is your soul.