• Yokemate of Keyboards
    Yokemate of Keyboards
    Posts: 2706 from 2003/2/24

    boot_wb wrote:


    I think much points in the direction that MorphOS 3.0 will be x86.


    Interesting! Any links/quotes on which you are basing this? (it truly is the first I've heard of it!)

    No links or quotes, just some observations and deduction, all mixed together. It's all speculative of course! :-)

    The first question I ask myself is whether MorphOS development will continue. I don't see anything suggesting that the development of MorphOS is about to end or even slow down from its current pace. It has kind of come too far for being terminated permanently, and I think the developers like to do what they are doing with the OS. They want it to continue, they want it to evolve, and they want it to reach some kind of success. So no, I don't think it will stop.

    So what about hardware? Back in the Commodore/Escom/Amiga International/Phase5 days, the PPC looked like a good choice. The Amiga walked in the footsteps of Apple in its 68k to PPC transition. PowerPC was backed by three large companies, back then there were also third party Mac HW companies, and the PPC platform had a great future laid out for it. When Amiga fell, MorphOS came in its place, picked up the banner, and continued the march ahead, still following this legacy. The PPC had few clouds in its sky back then. But some developer recently said something line of: "if we would have known back then what we know today, we would have chosen differently".

    The 8610 was the last interesting PPC chip, and *nothing new* (of interest) will follow that one. In essence, the PPC as a platform *we* would be interested in, has died. What's left is a big pile of aging Mac hardware. They are still kind of "modern" and usable, they perform better than anything MorphOS has come in contact with before, but they are not evolving anymore, and like dead fish it flows *with* the stream in the river on its way out to the great vintage ocean. It can only go in that direction, and it's only a matter of time until it gets there.

    I think it's a good thing to support the Mac PPC HW. After all, the current MorphOS is PPC, so why not make the most out of it? Unlike the Pegasos/"A1"/Sam440/etc, the Mac HW was mainstream, so it's cheap, easily obtainable, and the most powerful there is when it comes to 32-bit PPC. Cheap, powerful, and easy to get - isn't that what everyone wants? Yes of course! With the Mac Mini, PowerMac and PowerBook, the MorphOS team has covered everything from desktop to laptop, and the best and last PPC computers of their respective kinds. This should satisfy users (like Dreamcast270mhz for example) that wants PPC and nothing but PPC, and they will have plenty to choose from. But MorphOS has then gone as far as it can on PPC. PPC can't take it any further than this. It's the end of line.

    So if the MorphOS team still wants to continue MorphOS's development (which I think they will), and to do so on non-vintage HW (who wants to develop a *new* OS for HW that people will consider "vintage"), then they will have to jump to a new CPU architecture. Everyone is aware of that; in a response to the comment "I only regret that again we have an annoucement about old hardware", a developer said "Fair enough, but don't whine if it ain't a PowerPC based box ;)". No new PPC HW coming, with a possible exception of a desktop MB based on the 8610, and under any circumstances, this will only delay the inevitable.

    The next question is "what platform then"? Well, there are basically two options; the ARM, and the x86.

    ARM is massively used in various consumer electronic devices, like mobile phones, various kinds of set top boxes, TV's, PVR's, handhelds, and soon also netbooks, tablets and such devices. Extremely energy efficient, highly integrated SoC's with specialized controllers and support/accelerator cores, at really low prices is characteristic for ARM. Personally, I think this would be a perfect match for MorphOS; a lean OS that has the small footprint and low overhead of OS's like Symbian or Palm/Garnet but at the same time offers many of the higher end features one would expect from the much bulkier desktop OS's like Vista, OSX or Linux. It's an OS that sits in between those two categories, an OS that offers the best from two worlds. IMHO, this would make it suitable for the new kind of ultra portable computer products that are evolving from the traditional mobile HW, like smartbooks, tablets, and god knows what else, products where Symbian or Palm isn't enough, and Windows and Linux is too much.

    But the MorphOS team has never showed any kind of interest for this kind of hardware, neither by their words, nor by their action. Everything they have said, everything they have done, and *not* done (in regards to OS development, UI development, application development, etc) proclaims that MorphOS is a desktop OS for a limited and certain kind of computer enthusiasts in Amiga's legacy, nothing more, nothing less. And why not? It's a "hobby OS", both for the developers and the users. Who needs to think about finding a commercial purpose for the OS then? The developers likes it as a desktop, and this is also what the current users expects. And despite ARM's recent stretches upwards, way into the Intel Atom's territory, it will never become a full fledged desktop CPU. So it doesn't quite match.

    What remains is simply the x86. This is the perfect desktop CPU. It comes in many flavors, ranging from very low end and cheap, to extremely high end and expensive. Nowhere are there more HW manufacturers, nowhere is the competition more cut-throat. The x86 HW is mainstream, so it's cheap, easily obtainable, and the most powerful there is. Cheap, powerful, and easy to get - isn't that what everyone wants? And it will be here forever. No need to jump architecture again, ever. So I think the answer is given. And as a more or less pointless sidenote - I recently saw Ralph Schmidt make a few comments on the latest Intel Atom chip. I know that this doesn't mean much, but IMHO taking the time to read up on a chip, form some kind of analysis/comparison of it enough to make an ever so brief comment it online, at least signals some interest and knowledge of it. I only mentioned it because I have *never* seen him make any comments whatsoever on *any* ARM chip.

    I think the architecture jump will come, I think it's inevitable, and I am quite certain it will be to x86. This will secure a future HW base. They may still continue supporting PPC though, and they probably will...
    MorphOS is Amiga done right! :-)
    MorphOS NG will be AROS done right! :-)
  • »02.01.10 - 13:58