I would like to set a query to MorphOS team and to all MorphOS users out there, and I want to hear your opinions about that.

Why MorphOS doesn't follow the steps of Linux, with the difference that if it will be free or commercial product it depends on MorphOS team. Why MorphOS cannot be available on all platforms like Linux (PPC, x86, even on 68k), and not only for Pegasos machines.

My opinion is that we will have far more support from software developers, and the most important is that we will have far more users than now, which means more money, better market, more developers and programmers and more power as OS.

Why all of you, asking if a new graphics/sound card, will be running under MorphOS, or why we must use only ATA100 and not SerialATA, why we must use only DDR RAM, why we can't use new graphics card from nVIDIA or Radeon, why you must wait for 3D support on Radeon, etc. etc.

Why we must wait for an update everytime, to see if there are new drivers inside for some graphics cards, that they are already 3 years on the market?

Just imagine MorphOS, running on a P4 @ 3.8Ghz Prescott, and that you can install every new hardware, without the fear that it will not work.

well i get what your saying and i 75% agree. WHile understand your wants and for most part agree, but i hate intell and it needs to go. =)

[ Edited by zephar123 on 2005/8/16 3:45 ]

[ Edited by zephar123 on 2005/8/16 3:47 ]

Why MorphOS doesn't follow the steps of Linux

If MorphOS follows Linux (becoming another Linux distro) there is no point in using it. One of reasons to use MorphOS to me is that it is not Linux. Also MorphOS is not portable to little endian machines, so from technical point of view this discussion is pointless.

and that you can install every new hardware, without the fear that it will not work.

And who write those drivers and where you can get documentation for those hardware from? Get real.

Come on Krashan, be serious. MorphOS is not Linux and you know that. Maybe you didn't understand what I wanted to say.

I also don't understand what you said about the drivers and the documentations.

@Krashan:
Quote:
Well, i think it's portable to LE platforms, as soon as you drop backward compatibility with existing 68k and MorphOS apps, or use a bigendian GCC like Amithlon does for "native" apps. The performance sucks like hell of course, but people are happy... :)

@All:
I think closed source systems are not portable. Note that even on Linux, you can't bring binaries between platforms. Since source is open, porting is _very_ easy, but you still need to port (ie. build for every CPU at least). And LinuxPPC also has troubles with closed source apps not being available for it. The most notable example is a recent Flash/Shockwave plugin for browsers... So Linux's way is _NOT_ the way to follow for MorphOS.

Quote:
well, you seem to define sucking diferent then of myself... Amithlon seems to be very fast already and in emulation mode alone, I don't know it could be slower for running half/native...
Then we have AROS, showing just how 'sucky' the *miga concept can be on x86... of course there are things that should have special care about it, but it doesn't means it would be a sucky solution...

BTW I don't see MorphOS being Linux just because it runs on all platforms, as I don't see it being Mac for the sake of running on PPC alone... wait, Mac is going x86, WTF, we got to urge on a new definition for mac!

Quote:

I guess you said it yourself! MorphOS is not Linux, so you can't use Linux drivers. Drivers don't just pop up from nowhere, they have to be coded, even on Linux. And if you are going to write new drivers, why not write MorphOS drivers instead of Linux ones?

@pixie:
Stuff compiled with that x86-bigendian GCC is practically not, or only marginally faster than stuff in 68k emulation. The performance is really sucky compared to what such a CPU can do when compared to "real native" code (ie. one on the Linux side, below Amithlon).

@takemehomegrandma

Why everybody stuck with the drivers for graphics/sound cards I don't understand.

My point is why MorphOS can't be possible to run like Linux, so we don't have to buy specific (Pegasos) machine to run it.

ah Phantom you're bringing an old taboo up here

MorphOS becoming Open Source and being ported to PPC Macs would be far enough for me, with the mac developpers coding drivers for mac hardware (including gfx and soundcards) would help us as well. But it can only be a dream, MorphOS if full of licenced software and i doubt any of the owners would give it away to open source. The Amiga market has always been ruled by money and this is not about to change.

@Phantom
Quote:
I think that if Pegasos is a good concept as it seems (low consumption, cool computing) and viable for itself, there's no reason why an optimized form of MorphOS can't coexist with one for x86 (arm...) which for its inerent nature wouldn't be as optimized but would, on the other hand, provide the same funcionality and a larger user base to it..

[ Edited by pixie on 2005/8/16 13:24 ]

Your are daydreaming.
Open sourcing MorphOS would not automagically bring in developpers that would develop drivers for the Mac (dead end anyway) or the PC or whatever.

AROS is there as the AmigaOS(like) opensource solution. Someday we may get AROS PPC / Morphos binary compability.
Why don't you guys help out to make it happens ? You just to have to click HERE and donate ?

Talk is cheap, it is time for action.

[ Edited by cdfr on 2005/8/16 9:51 ]

Quote:
For the final time, this whole point is not about MorphOS. The point is about apps for it. How do you want to run existing MorphOS apps on an x86 version of MorphOS? You CAN'T maintain 100% ABI (a.k.a. binary) compatibility with current MorphOS, if it runs natively on an x86 CPU. (And same goes for AmigaOS4 anyway.) It can't be solved by simply emulating a PPC CPU, for endianness reasons. (And now don't come with Mac OS and Apple please, they had to face complettely different problems while doing their transition, and of course they have the manpower to do it.)

Same goes for a theoretical MOS/x86 app. You can't run it on PPC, w/o recompile. While 90% of MOS apps are closed source, recompiling it is quite hard. Also for many apps, simple recompiling is not enough. I speak for myself, but for example a port of Free Pascal would need quite a lot of work for MOS/x86...

I can't understand it's so hard to get this. MorphOS won't go for x86 in the foreseeable future, and because of a reason. Just live with this. End of the story.

@cdfr

gosh, take it easy, you sound like give some cash and STFU. We can babble for a while, nothing bad with daydreaming.


AROS doesn't gather a lot of interest as it runs on a machine (x86) that is already saturated with loads of Operating systems. The Mac doesn't have more than MACOS and Linux, there's some space for MorphOS, especially after the Apple and Intel's wedding.

Quote:

Because there aren't any drivers?

No seriously, I think MorphOS can be ported to run on pretty much any PPC based computer. It could possibly be ported to other architectures as well ("possibly" as in "not *totally* impossible from a technological point of view allthough it won't ever happen for lots of reasons, of which some (but not all) are technological"). But I think it comes down to this: MorphOS is, unlike Linux, a closed source, proprietary operating system with commercial ambitions. It also uses third party components that are proprietary and licensed. Theoretically, perhaps it would be possible for all stakeholders to agree on spending time, effort and resources to port the OS to alternative platforms and releasing it for free. I recall BBRV suggesting the same thing as you do - "why don't you port it to some Mac hardware" (presumably between the lines: "for free, on your spare time")?

However, in reality, I think most of the involved people rather want their efforts to pay off in one way or another since that's why they are here in the first place, and they might feel that there is a limit somewhere to how much work they are willing to spend without at least *reasonable possibilities* of some kind of financial reward related to it? If there are no reasonable opportunities for a reward, then they might choose to concentrate on easier things or things they like more, things more fun. Nothing wrong with this, I'm pretty sure that's how we all feel when we go to our respective daytime jobs every day. I mean, sure, from time to time I put in a little extra effort and extra work, maybe doing a little overtime without charging for it, but *there is a limit* to how much extra work I am prepared to add in for free when there is no obvious opportunity of financial (or other kind of) return connected to it.

(Please not that the following is merely speculations from my side, which can very well be totally clueless! You have been warned! )

If *everyone* involved in MorphOS would have been so keen on releasing all their work completely for free all the time, then I think we would have seen a "MorphOS 1.5" release for the **Pegasos** ages ago. I mean, the Pegasos is a little like the "homeland" after all, it's what everyone is using already, unlike all kinds of alien hardware that you may have in mind. And while we have seen updates of the OS for Pegasos recently (which I think was *great*, and *not* minor in any way), and also MUI for that matter, fact is that *not everything* was updated/released and/or not from all developers. At least that is how things seems to me.

Or am I wrong?

I mean, MorphOS has had an Altivec enabled kernel since ages, 3D, TCP/IP stack, gigabit network driver, and there are lots of other things that still only is available to the closed circle of developers, despite it must have been more or less painless and effortless to release *more* than the 1.4.4/1.4.5 from everything that *must* have been developed/finished since september 2003, when MorphOS 1.4 was released. Things *could* have been released, I'm pretty certain of that, but they haven't been yet. Not even for the Pegasos (and here you are talking about Mac ports ). I think the reason is the same; currently there is an unbalanced situation regarding work performed and financial rewards (or *realistic hopes* thereof). MorphOS is developed by many different individuals with their own respective view on this, ranging from those who released most of the latest updates on the one hand, all the way over to David Gerber (and some other?) on the other hand, and probably a few people in between.

Bottom line is that there is nothing wrong with an OS being proprietary and commercial. I think this is the only way of making real, solid progress. But at the same time, being commercial makes *commerce* quite fundamental, doesn't it? *THIS* is what's currently lacking for MorphOS; commerce or at least realistic commercial opportunities that people can buy in to. If/when there will be solid commercial opportunities for MorphOS on *any* current or future b-plan firmware ("CHRP2") based hardware (ncluding but not exclusively the Pegasos, since the firmware apparently will work like a HAL and make the same OS pretty much run everywhere), then I'm pretty certain that we will see a quick and sudden boost. The same goes for other hardware I guess, which is what making comparing with Linux so wrong; unlike for Linux, there has to be a real commercial opportunity for it, otherwise nothing will happen. This may be what we are waiting for now; the coders may go at various length with MorphOS without it (depending on their amount of free time and "fun factor"), but in order to secure a solid future for this commercial OS there has to be a solid underlying commerce to build it upon, and we are waiting for the entrepreneurs (BBRV?) to provide this situation ...

At least, this is how things seem to me. So personally, I'm not holding my breath for a MorphOS1.5 release on the Pegasos (and *certainly not* desktop Mac releases), but for the commercial opportunity that has to be there for any of this to happen ...

(wow, this became a long post ... )

Thanks for the insults. Everyone can daydream but I can not bring a possible solution for what has been asked in the thread ?

AROS is possibly a solution for what people are asking here, and AROS does run already on MAC (hosted on linuxPPC).
Now if this bounty would be released you could boot on a liveCD starting AROS right away and run your morphos apps on any Mac and maybe next on new gen consoles.




[ Edited by cdfr on 2005/8/16 12:12 ]

@cdfr

i can't see any insults, not even subliminal, in my previous post.
maybe you misread it

STFU is not exactly friendly. It sounds like in your previous post you put these words in my mouth.

My intent is not to sound rude but to get real.
What are the chances to have MorphOS opensourced and ported to different machines vs the chance AROS could become really usefull if we are willing to have interest in it and help them out.

The MorphOS / x86 / Mac subject has already been discussed to death without any result.
If results is what people want I just try to help them out thinking out of the box and finding a possible solution.


[ Edited by cdfr on 2005/8/16 13:05 ]

"xxxxxx or STFU" became an expression that isn't rude at all but nevermind. There was no offense, sorry bout that..

yeah you're right the subject has been discussed a lot but we don't have much to discuss anyway. oh yeah ! does anyone know when MOS 1.5 and radeon 3D drivers will be out ?

[ Edited by SoundSquare on 2005/8/16 19:29 ]

no pb, I guess it must happen when 2 french guys write to each other in English

As far as "1.5" is concerned - it seems like it will be a while yet.. but this is NOT a question for this website but for the MOS ML. Please post question there

magnetic

i live in France but i'm not french :)

@magnetic

lmao
i was kidding mate, i know about it

[ Edited by SoundSquare on 2005/8/16 22:09 ]

That may explain why you grasp english better than I do.
And maybe French too as after a few years in the USA I definitely speak Franglish

And I think magnetic asking to post 1.5 questions to the (temporarely down ML) is funny


[ Edited by cdfr on 2005/8/16 16:27 ]

@takemehomegrandma
And thats not even mentioning things like USB2.0, Firewire or Bluetooth, things that is pretty much expected these day.
BTW What is this about the firmware acting like a HAL? Sound interesting but I found nothing doing a quick google.

[ Edited by dan on 2005/8/16 22:50 ]

@magnetic

Just thinking out loud, and of course I may be totally wrong (like I probably am most of the time), but I hardly thinks my thoughts are very unique in our community ...


@"frogs"

Let's see some French kissing instead of fighting!


@dan

Quote:

Indeed, but those are new things. The things I mentioned is known to have been *finished* for more than a year (or even two) ...


Quote:

Well I may not understand all the details correctly, but AFAIK it roughly works like this (please correct or complement me): OS developers "ports" their OS to the Open Firmware, and once that is done, it will run on all hardware that has that firmware installed (drivers in the OS for unique motherboard features if any (like the marvell gigabit ethernet), or standard non-motherboard specific components like gfx cards, PCI cards (or on board stuff) of various kinds, etc, will of course be needed, but things like a driver for a Radeon 7000 card is not motherboard specific but more of a general matter for the OS regardless of the motherboard, right?). Cool, huh?

That is how MorphOS could run so easily on the Freescale HPC1&2 motherboards with Marvell Discovery 3 and Tundra TSI108 controllers, 7447A and 7448 CPU's etc - bplan "ported" the firmware to the boards and then the OS would simply boot up on them! The only thing the OS needs to care for is to be properly adapted for the OF, which will then cater for everything else.

If you go to http://www.pegasosppc.com/products.php you will see 1) The Pegasos motherboard, 2) the ODW, and 3) the CHRP2 - which is exactly what this is about, the *new* common hardware reference platform industry standard for system developers (both OS and hardware builders). You probably know that Genesi has been quite determined to bring pretty much every single PPC OS on board to the Pegasos (well except MacOS of course, as well as a few others perhaps, like OS4)? Why? Well, this is the reason! A motherboard developer/manufacturer that chooses to use the Genesi OF will automatically (pretty much at least) have all those OS's running on their system! And all OS developers that brings their OS to the Genesi OF will have their OS's running on all those motherboards!

This move seems to be embraced by Freescale, who seems to look at this as an important strategic move to strengthen the entire PPC industry (at least from the looks of it IMHO). I guess b-plan is putting serious effort into their firmware; this is something that will benefit every single one in the industry!