I've been involved with the Amiga computer from the very start. My involvement continues, since I have an X5000. I've become increasingly disillusioned by the lack of progress. My visits to Amigaworld have become an annual event. I've decided to face some hard facts. The specialized hardware is holding Amiga progress back. Everything about it is bad (from a modern point of view). It is expensive, it is slow, it is decades-old technology. This idea isn't new. It's a refrain we have heard for almost as long as the Amiga has been around.

I think AROS is the future. AmigaOne and AROSOne should join forces. Like I said, it's a hard choice because an entire community has been built around Amiga and AmigaOne hardware. There are heavily invested interests, both intellectually and financially, in the PowerPC architecture that resists change.

It is my understanding that the Amiga exec core functionality has been converted to C. That's a portable language with a vast x86/Windows component. How hard would it be to recompile it and link it with modern hardware? Some work in that direction has been done already.

I'm comfortable with AmigaOS4. Its roots are plainly evident in the original Amiga 1000 and it's like a cherished grandparent that has been around forever. One thing that bothers me about AROS is its "differentness". I know it has to be different because of copyright issues, but maybe they could be resolved when Amiga's future is at stake.

_________________
X5000 / microA1(OS4.1 FE U2) / CodeBench / Imagine / Blender
Lightwave 2019 / Microsoft Visual C++

@thinkchip

Well, PowerPC as a desktop/laptop CPU architecture is dead at least since the Apple's switch to Intel. As I always say, enjoy what you have and be glad our beloved computer is not gone yet. X5000 route was always a dead end - really nice for those with money to buy it, but too expensive for all other people around here. A1222 was intended as an entry model, well, we all know, how it ended.

My personal pet hardware project right now is using a generic PC as a stand-alone Amiga computer (via QEMU). However, playing video games (including the recent Heretic 2 port) takes too much of my free time.

"Differentness" (as you call it) is more about a try to make more eye-candy-like Amiga experience, which often ends somewhat unbalanced. OS4 has nice balance between new and old desktop features for conservative people like me. Well, I would have been quite content even with a plain OS3 desktop, so AROS developers are already almost there (with a better window/desktop skin).

@thinkchip

I would say, the whole Amiga design ethos holds it back. The PPC was put in place to compensate for the loss of 68K. OS4 used it, even though only a small niche of software used WarpOS/PowerUP on PPC, and OS4 had no compatibilty with WarpOS or PowerUP at all. It only supported 68K in the OS. So PPC, which was really a bandaid CPU solution with little compatibility, was used as the CPU of choice for OS4. They could have chosen another CPU at that point. The Itanitum was still around and what was left of HP RISC.

I'll continue. Amiga is held back by the OS. OS4 is built to be compatible with 68K. So compatible to the point they practically cloned all the 68k structures and design. Now days we think of breaking into 64 bit, multicore and mulithreaded. Nope. AmigaOS is too rigid. Fine for its age but now days flaws in the design cripple it. Especialy the open nature. They designed it so the internals are exposed to the public and some operations require direct access to system objects. Not to mention the never ending "forbidden" problem where the whole system needs locking for one small object.

Then there are the users. There are a plehora of 68K users right now. While some 68K users were interested in the AmigaOne (and some still are) that was over 20 years ago now when it was something new in the scene. The rest went back to 68K. And, they don't like things like the AmigaOne, the most common reason being it's not a real Amiga, it doesn't play A500 games off floppy or something like that. Let me tell you, no Vampire or PiStorm user is busting at the seams to run Linux/68K on the Amiga, because it breaks the 64 bit barrier with real thread support. But why not? The answer to all the OS problems running on an Amiga chipset.

I'm not sure about AmigaOne being a force. Do you mean running AROS on an AmigaOne? There is a Sam port fo AROS. There's one thing that AROS does to break the 64 bit barrier. It breaks compatibility. Like a ColdFire 68K, it doesn't have 68K binary compatibility. Only source compatibility. Aside from on 68K which is a moot point. Now, would an AmigaOne AROS need 68K? Or could it survive wthout compatibilty and with modern features? It's not the PPC CPU that is holding OS4 back from being 64 bit, SMP and threadable.

About C and x86. Well to cut a long story short I've been discussing this recently. In theory I think it can. But not as generic portable code. AmigsOS itself is not portable, or specificially, is not endian portable. It is big endian coded. So I think it would need a custom compiler to only emit big endian compatible code. It's the 2020s now, the endian wars are over, x86 supports big endian directly. (Arguments online against networking protocols using big endian are lame and archaic now.) This would limit the x86 instructions and address modes, with memory to register and vice versa, so it would "act" more RISC. But could be possible. There is an Amithlon i686be compiler that does just that as well as the Intel BEC. What's more, GCC now supports storage atrributes, to mark byte order of data.

Last edited by Hypex on 16-Sep-2024 at 09:11 AM.
Last edited by Hypex on 15-Sep-2024 at 06:12 PM.

@thinkchip

If Aros was the future it would have already succeeded .
The so called AmigaNG is dead because it is barely an upgrade to AmigaOS. Compatibility should have been broken years ago. If you don’t have full support for SMP, Memory protection, Ressource tracking etc it is dead. AmigaOS should have all others others mainstream OS features or it’s useless. Don’t forget security.
Haiku is more interesting as an alternative OS IMHO.
The Exec developer team is not even able to have multicore support and they are at it since many years. IIRC Jason Mc Mullan developed a prototype for (SMP) Aros in a short period of time by comparison.
Quite frankly if it is not done by Amiwest 2024 you can switch to an other OS.

Last edited by kamelito on 15-Sep-2024 at 07:37 PM.
Last edited by kamelito on 15-Sep-2024 at 07:37 PM.
Last edited by kamelito on 15-Sep-2024 at 07:36 PM.
Last edited by kamelito on 15-Sep-2024 at 07:35 PM.

@kamelito

For most Amiga users, compatibility certainly an important feature (if not most important). We could use other OSs with some skin otherwise (remember Commodore OS Vision?).

Last edited by pavlor on 15-Sep-2024 at 07:40 PM.

@thinkchip

AmigaOne/Amiga OS4 is dead. That is not coming back.

Use AROS and/or Morphos. Morphos will likely have an x86-64 port in the future.

_________________

@WolfToTheMoon

Quote:

Morphos will likely have an x86-64 port in the future.

Well, I hope there will be still some Amiga users left then.

@pavlor

They could use a PPC sandbox like they do for 68k.

@kamelito

Quote:

They could use a PPC sandbox like they do for 68k.

There is no sandbox for 68k in OS4. Both PowerPC and 68k applications are treated the same (eg. can use both 68k and PowerPC libraries).

@pavlor

You got my point.

kamelito Quote:

If Aros was the future it would have already succeeded . The so called AmigaNG is dead because it is barely an upgrade to AmigaOS. Compatibility should have been broken years ago. If you don’t have full support for SMP, Memory protection, Ressource tracking etc it is dead. AmigaOS should have all others others mainstream OS features or it’s useless. Don’t forget security. Haiku is more interesting as an alternative OS IMHO. The Exec developer team is not even able to have multicore support and they are at it since many years. IIRC Jason Mc Mullan developed a prototype for (SMP) Aros in a short period of time by comparison. Quite frankly if it is not done by Amiwest 2024 you can switch to an other OS.

Too much reality for AmigaOS 4 desktop fans. SMP support for AmigaOS 4 was announced over a decade ago. The following article is from 2013.

https://www.osnews.com/story/27421/amigaone-x1000-successor-coming-amigaos-to-get-smp-support/ Quote:

AmigaONE X1000 successor coming, AmigaOS to get SMP support

https://blog.hyperion-entertainment.com/multicore-and-amiga-present-and-future/ Quote:

Multicore and Amiga: Present and Future Posted by Thomas Frieden on February 25, 2015 Symmetric Multi-Processing (SMP) is on the wishlist of AmigaOS users for quite some time now, and while progress has been made, we’re still not there yet. To explain the progress, let us first look at the concept, and then point to where we are in the whole. ... Where are we now ? The development of SMP support has been separated into several distinctive steps. The first step was to rewrite the scheduler in C for easier accessibility. In the very end, this step might be reversed again, rewriting the then SMP capable scheduler back into assembly language. The second, more fundamental step was to decouple the scheduler from its current data structures. As you might know, ExecBase contains a lot of list for task that are ready, or waiting for a signal. This has now been achieved. The current development build uses a scheduler that no longer uses the original AmigaOS data structures, but a structure that is replicated for each core. The next step is to have each core in the development system (currently, the X1000) to run the scheduler. Test code will then start tasks on the different cores and see how they behave. We have already experimented with this and the results look promising. The tests basically showed that the lockout mechanism for Forbid works as planned. As a final step, the balancing will be introduced, which then finalizes the first implementation of SMP support in AmigaOS.

Some of the announcements about SMP have been taken down but articles remain.

PPC likely has the weakest memory consistency model of any architecture ever used for the desktop which makes SMP more difficult to implement. I wonder how many memory barriers and lock acquire (sync/isync/lwarx/stwcx) type of instructions were added to try to get SMP working. I bet AmigaOS 4 is littered with them reducing performance and still no working SMP. ARM also uses one of the weakest memory models on the desktop.

https://en.wikipedia.org/wiki/Memory_ordering#In_symmetric_multiprocessing_(SMP)_microprocessor_systems

The x86-64 architecture uses a strong memory consistency model like most CISC architectures. RISC architectures were searching for a performance advantage to offset the RISC performance disadvantage and once again did not find much performance.

Last edited by matthey on 15-Sep-2024 at 09:48 PM.

Guys, really. There's almost no software written for PPC Amiga OSes, including the OS itself, that isn't in C or another high-level language. People just need to get over the cult fetishism for it and move to an alternative architecture. All the hard work of becoming portable was done moving from 68K to PPC in the first place. It's cringingly ironic to be "stuck" on PPC under the circumstances.

_________________
Doing stupid things for fun...

@Hypex

Quote:

What's more, GCC now supports storage atrributes, to mark byte order of data.

Thats one is the KEY to migrating to another CPU,
if this works then the system structures can stay the same.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@pavlor

I agree compatibility is more important than SMP or full 64bit support.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@Hypex

Quote:

Hypex wrote: @thinkchip I'm not sure about AmigaOne being a force.

I think in this case AmigaOne is synonymous A-EON. Despite the fact that A-EON have stopped using the AmigaOne mark, many people still associate AmigaOne with A-EON and its X series of AmigaOS 4 computers (X1000, X5000) and the A1222 stillborn attempt at an entry-level AmigaOS 4 computer.

As forces go in our Amiga niche, A-EON is considered one of them. I have on many occasions advocated that A-EON should've started the switch to AROS the second the friction increased with Hyperion. I know they hitched their wagon to AmigaOS 4 to legitimize their products in the eyes of Amiga OS lineage chauvinists, but that was more than a decade ago. And an OS means diddly if there are bugger-all apps to run on it.

They need to meet the users where they are.

Whether they choose ARM64 or x64, A-EONs best option is to legitimize AROS. AmigaKit is already doing so with System v46 on AROS68k, with native ARM datatypes.

Last edited by agami on 17-Sep-2024 at 03:37 AM.
Last edited by agami on 17-Sep-2024 at 03:27 AM.

_________________
All the way, with 68k

@thinkchip

I think the AROS variant and its offspring is confusing. I can’t keep them straight. Amiga should return to its roots in 68k and thumb its nose to anything else.

_________________

@thinkchip

Quote:

thinkchip wrote: I've been involved with the Amiga computer from the very start. My involvement continues, since I have an X5000. I've become increasingly disillusioned by the lack of progress. My visits to Amigaworld have become an annual event. I've decided to face some hard facts. The specialized hardware is holding Amiga progress back. Everything about it is bad (from a modern point of view). It is expensive, it is slow, it is decades-old technology. This idea isn't new. It's a refrain we have heard for almost as long as the Amiga has been around. I think AROS is the future.

It depends on what do you mean with "future".

Every OS which is a port or reimplementation of the Amiga OS has no future, because it's missing too many modern, mandatory, features.

Besides that, AROS is on a better position compared to others because it's open source and people can make experiments (that's why it has 64 bits and some SMP implemented).
Quote:

AmigaOne and AROSOne should join forces. Like I said, it's a hard choice because an entire community has been built around Amiga and AmigaOne hardware. There are heavily invested interests, both intellectually and financially, in the PowerPC architecture that resists change.

It happens when you make wrong choices.
Quote:

It is my understanding that the Amiga exec core functionality has been converted to C. That's a portable language with a vast x86/Windows component. How hard would it be to recompile it and link it with modern hardware? Some work in that direction has been done already.

Yes, and it's called AROS: why do you want to reinvent the wheel?
Quote:

I'm comfortable with AmigaOS4. Its roots are plainly evident in the original Amiga 1000 and it's like a cherished grandparent that has been around forever. One thing that bothers me about AROS is its "differentness". I know it has to be different because of copyright issues, but maybe they could be resolved when Amiga's future is at stake.

For the OS4 roots it's better to take a look at its history: La rivoluzione mancata di AmigaOS

As you can see, it was just a make-shift, because the plans were very very different.


@pavlor

Quote:

pavlor wrote: @thinkchip Well, PowerPC as a desktop/laptop CPU architecture is dead at least since the Apple's switch to Intel.

No, it was already well before it. Here's another article on this specific topic: Apple e i processori: non è amore (eterno)
Quote:

pavlor wrote: @kamelito For most Amiga users, compatibility certainly an important feature (if not most important). We could use other OSs with some skin otherwise (remember Commodore OS Vision?).

You can use (Win)UAE for full compatibility, even with games.


@NutsAboutAmiga

Quote:

NutsAboutAmiga wrote: @Hypex Quote: What's more, GCC now supports storage atrributes, to mark byte order of data. Thats one is the KEY to migrating to another CPU, if this works then the system structures can stay the same.

It's not an easy task, as you can read from this article: AmigaOS & co.: da PowerPC a x86 o ARM

However, the main problem for an OS should NOT be the endianess. To be more precise, if an OS relies on the endianess, then it's very badly written.

OSes should be endianess agnostic!


Now, regarding the "Goodbye PowerPC?", I've already said it 13: PowerPC: un’architettura al crepuscolo per i “nuovi� Amiga?
and 11: Freescale (ex Motorola) “ufficializza� la fine dei PowerPC
years ago.

Welcome on board!

P.S. All articles are in Italian, but you can easily translate them with DeepL (the most accurate) or Google Translate (if you like to keep the formatting).

@thinkchip

some things

don't wrote posts titled "Godbye PowerPC?"
accept that switch to PowerPC was announced by Amiga Technology 29 years ago in 1995
first PowerPC cards were deliverd 27 years ago in 1997
some people may not want to switch from PowerPC simply because it passed so many years with PowerPC on Amiga may have nice memories connected with PowerPC from they youth

so leave PowerPC as it is and if You want start working on new Amiga like solutions on pc

aros is win98 level os more than 25 years behind win with no worth of use software

If You want people to use Amiga like solution on x86
You have to provide something decent at least on win7 level
something that will be Amiga Os X which means Amiga gui and graphics on top of Unix
with some nice exclusive software for it

So

first take trolls like szulc, szonwejs, karlos, di mauro, wolftothemon, v8 etc
force them to hard work on mui clone
when it will be done made nice decent gui builder for mui
when it will be done port amiga gui and graphics to unix
when it will be done port decent webbrowser, mplayer, pdf reader
when it will be done made some new nice exclusive office software for it
ofcourse with support for docx xlsx
when it will be done made some nice exclusive software for 2d and 3d graphics


when it will be worth of use
people will use Amiga like solution on x86

@thinkchip
The problem is not only the hardware, but the lack of software, there is little point in having the most modern and fastest car if we do not have a circuit to use it.
And even if you do something for modern systems, people still stick to their old hacking practices, so why want the most modern Amiga when we can play hundreds of classic Amiga games for free?

@kamelito

Quote:

Quite frankly if it is not done by Amiwest 2024 you can switch to an other OS.

Which OS do you have in mind here? Most certainly not the ones supplied with (NameAFruit) Pi n (suffix), is it?
I do have a whole slew of those contraptions around here, from Raspberry Pi 2, 3, 4/4, 4/8 and 5, Orange Pi 5plus and a couple of more on their way just for fun and I'll tell you straight away: they don't actually interest me ONE BIT (although they have plenty of them...). I'm not a Linux enthousiast and probably never will be. And Linux is the what drives them.
The only thing I use them for, and come to think of it: 'them' means RPi3, is some simple programming stuff, some basic developments and 'getting-to-grips'. Not much more.

Currently I don't see anything remotely interesting in the way Amiga OS is, but maybe you do.

OldFart

N.b.: for daily use I have a Windows 10 box, which is even less appealing

_________________
Life is a waste of time.
Time is a waste of life.
Get wasted all the time
and you'll have the time of your life!