MorphOS on x86???

[Piru]

@cicero790

LLVE

How exactly does that solve the driver issue?

[dammy]

Finally: This is an Amiga community. If AROS can run the Amiga software then the programs that are liked can be used. Final writer and so on. And it is possible to make better versions because the added power of AROS and the hardware under it.

AROS has EUAE for years.  There is a bounty for integrating EUAE with AROS that is open to donations.  There are indeed several options for the Amiga Community to deal with the above:

1. Fire up EUAE when you need to run an Amiga68K app.
2. Routinely donate to the EUAE bounty on P2P until a developer full fills the bounty.  If you want AROS to run on your existing hardware, donate half of the money you are saving by not buying new hardware.
3. Port whatever software you are missing to AROS or rewrite from scratch.

I'm simply amazed people on here expect AROS developers to do their bidding for FREE.  I guess they still haven't gotten the hang of OSS vs paying for hardware/OS in a package.  Sure, if your buying hardware, I can clearly understand you demand to run your preferred software with minimum of effort.  This is because your spending $, probably a significant amount, that will go to pay the hardware and software engineers plus everyone else in the loop that brings you their product.  This goes for OS4 or MOS.

AROS, OTOH, unless your coughing up cash for bounties it's costing you nothing.  If you paid nothing into AROS, how the hell can you even expect a developer, who your not paying, to do things you want if it's not of his/her interest?  AROS is not being funded by your taxes. Your buying any product will not yield directly to any AROS developer's paycheck because they don't get a paycheck.

To quote an old US saying (forgive the drug reference plz), "Grass, Gas, or Ass, No One Rides For Free."

Dammy

[cicero790]

@Piru
You are the one with knowledge here not me. I return the question by asking:  Is there a way to use the native pc drivers. I don’t know if this is possible but if AROS could somehow utilize the drivers so they didn’t have to be rewritten??

About LLVE: I overheard a discussion about apple going this way for cpu independence that all.

[cicero790]

"Grass, Gas, or Ass, No One Rides For Free."  :lol:

I will contribute.

[warpdesign]

 don't think they're stupid, nor I think they could be removed. We wouldn't have any software then.

Yes. And we could start writting software...

Do we have that much software right now ?
If we exclude 68k software (that tend to crash, hit hardware,.. anyway), do we have that much native software ?
And aren't already all native software available in x86 ? Provided the OS and some third-party libs (SDL,...) are available, wouldn't it be possible to easily port these apps on an hypothetic x86 MorphOS ?
And most software is still in the works anyway... So what would it change ?

   most software could be recompiled


You live in a dreamworld.

Maybe, maybe not. But seems like people writting the original Quark/QBox also lived in such a dreamworld... I doubt Quark/QBox was only meant to run ABox... Or what a waste!

I'm really wondring what's the fear of moving forward...

This could not work at all, so what ? It's not for the money, right ? so...

And no I couldn't write my own... Designing an OS is a huge work... And you need to be experienced in both OS design and programming. I'm in neither...

[cicero790]

downix you said:The Amiga was about being the next big thing. Frankly, there is no NBT in desktops. There is, however, in smaller, lightweight products. And the Amiga, and MorphOS, are uniquely situated to such markets. Instant-on, multitasking, and scalable. And the weakness of the design, not so relevent anymore once you get to that scale.

I say: I can see your point. And agree. But I have a different view about Amiga's possibility to snatch back market shares and finding its place next to PC and MAC. And why? because being in the Amiga OS is an experience, that is special.  

[Piru]

Yes. And we could start writting software...

Who exactly? I don't see that many devs around anymore. And it would be really hard to attract anyone to platform with zero existing apps.

Do we have that much software right now ?

More than would be with x86, that is none.

If we exclude 68k software

But I don't. Granted I only use couple of the m68k apps anymore, but it is invaluable SW library to have in case you need something. Having aminet to look into is way better than having no software at all. And no, UAE just doesn't cut it IMO.

And aren't already all native software available in x86 ?

No, for some hypothetical OS we have 0. Just being x86 doesn't automagically give you software.

Provided the OS and some third-party libs (SDL,...) are available, wouldn't it be possible to easily port these apps on an hypothetic x86 MorphOS ?

Only slightly easier when dealing with bad little endian dependant code. Other than that it wouldn't make porting  easier.

But seems like people writting the original Quark/QBox also lived in such a dreamworld... I doubt Quark/QBox was only meant to run ABox... Or what a waste!

Quark works just fine thank you. Just because all the possibilities weren't used doesn't make it any less valuable. Not a waste at all.

I'm really wondring what's the fear of moving forward. This could not work at all, so what ? It's not for the money, right ? so...

There is no fear, there is the reality of the situation that I've explained couple of times already. Making 10 (20 in some cases) years of code suddenly work in an endian safe manner would be a massive workload, too much to handle. Coding drivers for every imaginable HW would be ever more work. Plus the quickly moving target the x86 is wouldn't make it any easier.

And no I couldn't write my own... Designing an OS is a huge work... And you need to be experienced in both OS design and programming. I'm in neither...

So I guess I can just ignore your comments in the future?

[Hammer]

Not the most scalable either, as you can find chips like PPC, ARM and MIPS running in a wider variety of devices.

X86 scales from embedded micro-controllers to super computers. Majority of X86s falls between netbooks to servers.

From 2005 market share(from memory)
~700 million ARM
~200 million X86
~140 million MIPS
~60  million PPC

PS; I need to find an updated market share stats.

[Hammer]

Painkiller wrote:
I would rather see MOS on PS3 than any other platform. Here are a few reason:

-Cheap
-Very powerfull processor
-Will be available for many years to come
-Many people already owns it

Only thing bothering is the fact that accelerated GFX are disabled when using other OS with PS3. This could be over come by writing an acceleration driver for the SPEs, but that kind of think would eat up a lot of development resources.

Transgamming made Swiftshader product that renders DX3D 9b on X86 CPUs.

[warpdesign]

Who exactly? I don't see that many devs around anymore. And it would be really hard to attract anyone to platform with zero existing

There aren't, but quite difficult seeing there's no hardware to run MorphOS, don't you think ? With massively available hardware anyone could potentialy try MorphOS and/or develop for MorphOS. Right now, no one but people having spent a lot of money can...

So I guess I can just ignore your comments in the future?

Not being experienced in OS design doesn't mean I cannot make any interesting remarks...

But you're free to do so of course... If you feel so superior to non experienced people...

[Piru]

@warpdesign

Not being experienced in OS design doesn't mean I cannot make any interesting remarks...

Interesting perhaps, but those remarks are somewhat misguided due to lack of insight and practical experience. Thus, if you don't mind I won't continue this particular discussion further.

Obviously you're entitled to have your opinion, but I don't agree with it. If you feel I'm wrong, you can always prove me wrong by starting your own x86 OS and making it successful. Good luck.

[Painkiller]

There aren't, but quite difficult seeing there's no hardware to run MorphOS, don't you think ? With massively available hardware anyone could potentialy try MorphOS and/or develop for MorphOS. Right now, no one but people having spent a lot of money can...

What do you mean? Efika is still available from numerous of stores and the price ain't bad.

For the future it is nice to see them doing a port for Mac Mini as there are loads of them on sale at ebay and other places. With that platform I don't think there is any reason to whine about the missing hardware as it is cheap and can be bought pretty much from any country. Yes it is used but who cares.

[ferrellsl]

@TheMagicM

Please refer back to the first post in this thread....

[ferrellsl]

@mshulz

I think we've both been misunderstanding one another.  I've been saying that there is nothing to prevent MorphOS from being ported to x86 from a technical perspective.  Then I was jumped by Piru because I said that the port would never happen because porting MorphOS wasn't a money-making venture.  I know that the expertise exists to conduct such a port.  I started out in this thread by rejecting the claim that a port would never occur because of endian-ness issues. Someone stated that endian issues made it impossible to port MorphOS to x86.  Those issues CAN be overcome as evidenced by several operating systems, including AROS.  BTW, thanks for porting AROS to PPC.

[downix]

Hammer wrote:

Not the most scalable either, as you can find chips like PPC, ARM and MIPS running in a wider variety of devices.

X86 scales from embedded micro-controllers to super computers. Majority of X86s falls between netbooks to servers.

From 2005 market share(from memory)
~700 million ARM
~200 million X86
~140 million MIPS
~60  million PPC

PS; I need to find an updated market share stats.

Not that much different, save MIPS now has a Chinese firm pushing it hard in that rapidly growing economy.  PPC has also grown.  

x86 does scale, but not as completely as other options for a reasonable price.  New MIPS netbook announced last week, $100.  New x86 netbook announced last week, same features, $350.  

Not saying it can't go x86, just that x86 is not this magic bullet gee-whiz-fix-it-all.

[dammy]

When enough people want MorphOS on x86, the MorphOS team will do a return on investment analysis. If they can make money by porting it to x86, then they'll do it. Why else would they do it? Out of love for the OS??? Get real. They need to eat and pay their bills too, just like everyone else.

Question is, are they making a killing being bundled with EFIKA?

Dammy