Which OS Would Be The Best Amiga Way Forward...

[commodorejohn]

WB3.1. What can I say, I'm a contrarian.

[commodorejohn]

I personally have accepted there will never be a new Amiga, so that leaves Amiga compatibles and that's that. And because of that I don't really care to be honest as the only answer is a complete X86-64 rewrite of everything.....so in other words forget overpriced PPC MOS and OS4 options so a supercharged AROS I guess.

I don't think that's even remotely the only answer. It's still possible that someone could come up with a more reasonable PPC-based system - it's not like the architecture is dead, it's just in a rough patch for consumer computing - but it's still holding its own in the supercomputer market, and more relevantly, in the console market. Who's to say it doesn't make a bit of a resurgence as far as cost-effectiveness goes?

There's also ARM, which is kicking more ass than ever before now that it's become the architecture of choice for mobile computing of any variety - it's already obliterated the competition in the mobile-phone and tablet markets, there's a few netbooks out there sporting it, and Apple's apparently moving its laptop line over (and quite possibly its desktops, who knows?) Even Microsoft is getting in on the act - and not just for mobile platforms. You could do a lot worse for a new AROS machine than a Tegra board, that's for sure.

And there's also the Amiga clone-upgrade projects - sure, Natami isn't going to be as inexpensive as an ARM system, but for those of us who really like the 68k, it's quite promising - and FPGA Arcade looks pretty encouraging, too. So no, x86 is hardly the only choice. And frankly, if most of the platforms still using it are the PCs that need it for compatibility (and even the one PC OS where that's a factor is exploring other avenues,) then it looks like x86 might very well be on the way to becoming legacy hardware itself, barring another change in the winds.

(That said, I do agree that OS4 is a dead end - Hyperion obviously have very little clue about even the market they're targeting, if their definition of "classic Amigas" is "classic Amigas with a PPC board." How many of us even have such a setup? At least MorphOS can be assed to make their OS run on more than that very particular subset of system configurations.)

But for me Amiga ONLY equals the machines Commodore made 1985-1994. And therefore the best OS for Amiga is actually the one supplied by Commodore.

I'm with you there, although I'd count NatAmi and FPGA Arcade if they turn out to be what they're looking to be. But I don't begrudge people who want a more powerful system running an Amiga-like OS.

[commodorejohn]

I love the "but mommy, he started it!" posts...

Blargh, yes. Remember when this thread was actually about the merits of different operating systems?

[commodorejohn]

PPC is far too expensive to achieve i7 cheap as chips speeds or value for money, and ARM is barely suitable for use as a netbook IMO

It's true that PPC is currently not cost-effective for desktop computing - but if things continue such that all three major console platforms are based on it, that's a fair bit of money flowing and an incentive to continue to improve it. And ARM is indeed still lagging behind x86 for desktop use, but it's rapidly catching up to the Atom line. And both of these have the advantage of being native RISC architectures, not CISC architectures run in emulation on a RISC microarchitecture. They might not be optimal solutions now, but that doesn't mean they might not be in the near future.

X86 is the only mass produced CPU we have

Uhh...what?

The software alone can not fight back against the sheer number crunching power available to people running code on super computer levels of performance on the latest i7 or AMD Thunderbird chips. FACT.

So who says it needs to? This stupid power-computing dick-measuring contest with platforms that have far more money and research effort pouring into them is the single biggest reason NG Amiga hardware has never gotten anywhere. This community needs to get over its massive inferiority complex once and for all.

[commodorejohn]

Might want to try renaming it to .tar.bz2 there, Franko - some archivers (including at least the implementation of tar on my system) won't understand that it's a zipped tarball otherwise.

[commodorejohn]

Hmm...oddly, it looks like there aren't very many tools for classic Amiga that handle both tarball and bzip - you might have to decompress it with bunzip first and then use untar to unpack it. That's a bit frustrating...

Oh, beaten. Eh, what the hell, I'll leave it for posterity.

[commodorejohn]

AmigaOS is pretty neat, but a better OS could be built. However, what would the market be for it and how would you attract developers?

I don't think those are really the point, at least not in the initial phases of development. Any new OS is going to start off as a hobby project, and remember the golden rules of small software development: "Every good work of software starts by scratching a developer's personal itch" and "Large programs that work start off as small programs that work."

I don't think anything good is going to come out of trying to create a new OS around marketing considerations. Rather, you'd do better to find an OS paradigm that interests you, do your best to do a good implementation of it, and see if what you have interests anybody else. Remember, Linux started off as a college student's hobby project to learn 386 assembler!

[commodorejohn]

No, the GPL doesn't require you to redistribute your modified version of someone else's source - it just sets the terms under which you're allowed to if you do decide to.