AmigaOS 3.2 and beyond

[Bobson]

Originally Posted by E-Penguin

I think you probably want to look at AROS

No... AROS has no commercial support. Progress is slow. Over the years, from time to time I've donated to the project... its impressive what the developers have achieved - but its still not there yet. For awhile I even supported the Patreon to port AROS to the pi.

I'm talking real AmigaOS 3.2 running as a seamless emulation on capable, keenly priced, readily available hardware -- It would sell. Those retro consoles - mini snes etc, sell like hot cakes. This would be more powerful and useful. The only thing stopping me from pursuing this myself is a lack of licenses...

Originally Posted by Steril707

Btw, if you want an Amiga OS variant that runs on X86 hardware, check out AROS.
Be prepared for not many people giving much of a fuck about it, though.

Did I mention that I wanted X86 AmigaOS? No... I don't think that would be viable for a number of reasons... it also doesn't look to the future. You can continue to fight old battles if you wish...

Originally Posted by bubbob42

because people tend to underestimate the amount of work it would entail. If you‘re seriously interested in ARM & Amiga, a ZZ9000 might be something for you to have a look at.

... well, I'm only basing my assumptions on the development of Amithlon. It didn't take forever. Amithlon also had to work with more hardware combinations. A Raspberry Pi is a self contained SOC...


I didn't show up here to say "Port AmigaOS to ARM"... my post was far more nuanced - Amithlon was not a port of AmigaOS to X86.... it was a seamless emulation -- AmigaOS was still 68K -- it had the potential to be a path to a X86.

An ARMithlon type thing could do the same. Our hardware is dieing... kit on ebay is becoming more and more expensive... we'll all be emulating soon enough.

[Tigerskunk]

Originally Posted by Bobson

No... AROS has no commercial support. Progress is slow. Over the years, from time to time I've donated to the project... its impressive what the developers have achieved - but its still not there yet. For awhile I even supported the Patreon to port AROS to the pi.

I'm talking real AmigaOS 3.2 running as a seamless emulation on capable, keenly priced, readily available hardware -- It would sell. Those retro consoles - mini snes etc, sell like hot cakes. This would be more powerful and useful. The only thing stopping me from pursuing this myself is a lack of licenses...


Did I mention that I wanted X86 AmigaOS? No... I don't think that would be viable for a number of reasons... it also doesn't look to the future. You can continue to fight old battles if you wish...



... well, I'm only basing my assumptions on the development of Amithlon. It didn't take forever. Amithlon also had to work with more hardware combinations. A Raspberry Pi is a self contained SOC...


I didn't show up here to say "Port AmigaOS to ARM"... my post was far more nuanced - Amithlon was not a port of AmigaOS to X86.... it was a seamless emulation -- AmigaOS was still 68K -- it had the potential to be a path to a X86.

An ARMithlon type thing could do the same. Our hardware is dieing... kit on ebay is becoming more and more expensive... we'll all be emulating soon enough.

Okay, I'll bite.. lol.

Big question: who would want to use this who cannot already just use Pimiga for this purpose?

[bubbob42]

Originally Posted by Steril707

Big question: who would want to use this who cannot already just use Pimiga for this purpose?

True. In a few years, emulated Amigas will become a lot faster than real ones (they already are by far when enabling JIT).

[Bobson]

Originally Posted by Steril707

Okay, I'll bite.. lol.

Big question: who would want to use this who cannot already just use Pimiga for this purpose?

We have an issue: Our Amigas are old...

Many of us are hanging all sorts of adaptors off our machines to breathe new life into them.... eg: Goteks...

I see many posts from people discussing how PiStorm is a game changer. I know several people using Pi Zero's for their HDMI output etc...

Pimiga is great - I use it as my daily Amiga driver... but it still doesn't feel like an "Amiga". It requires set up - so not 'plug and play'. It's super fast - but doesn't move Amiga along in any way. You've got to be really into Amiga to bother with it.

Vampire and standalone Vampire - some argue the Vampire is emulation, some argue the Vampire is native silicon - it doesn't matter.... They're attempting to move the Amiga forward with their SAGA 'chipset'... (I know its all FPGA, and its not yet 100% compatible). They have an eye on the past, but also the future.

Vampire is however costly. You've got to be really into Amiga to want one.

People are willing to use a Raspberry Pi to keep their machines going... and we know the Raspberry Pi is more than capable of running AmigaOS under emulation... the emulation just isn't seamless - you're aware of it.

At the last count, over 3 million Raspberry Pi 4's have been sold...

If just 1% of that audience of 3 million (30000) could be captured... I don't know how many copies of OS 3.2 will be sold... but 30,000 is not to be sneezed at.

Aplug and play Amiga emulator, that seamlessly boots up, which can run compiled ARM code within the emulated Amiga OS (as a path to the future). That hides the emulation like Amithlon did....

Sell it as a 'magic pack/desktop dynamite pack' - hell... get David Pleasance back on the case. It would be a classic Amiga, and a route to a modern Amiga - capable of running modern stuff... it would have wider appeal.

If PiStorm is a gamechanger for those of us with real Amiga hardware - why wouldn't this be for us, and a wider audience?

Our hardware will be dead... we'll be dead... Amiga will die with us unless...

[Bobson]

Originally Posted by bubbob42

True. In a few years, emulated Amigas will become a lot faster than real ones (they already are by far when enabling JIT).

Exactly what I'm driving at - seamless JIT enabled emulation on dedicated hardware such as the Pi would feel very Amiga.

[malko]

@Bobson: you are a new 'Amithlon' friend of mine

[Bobson]

Originally Posted by malko

@Bobson: you are a new 'Amithlon' friend of mine

Hehe I didn't have Amithlon... I didn't have a PC at the time. I only ever saw Amithlon running - used it maybe twice... a friend had it. I remember how seamless it felt.

[eXeler0]

Originally Posted by Bobson

An ARMithlon type thing could do the same. Our hardware is dieing... kit on ebay is becoming more and more expensive... we'll all be emulating soon enough.

Im not so sure, the kits we have will work for quite some time.. And already there are replacement cases, keycaps, mobos, accelerators for most Amiga models.
What will happen next is probably a mix of FPGA reimplementations of various chips and Buffee-like solutions where it makes sense. (And probably some raspberry pi zero trickery etc..)
Emulation will be there for sure but physical Amigas die hard ;-)

[Bobson]

Originally Posted by eXeler0

Im not so sure, the kits we have will work for quite some time.. And already there are replacement cases, keycaps, mobos, accelerators for most Amiga models.
What will happen next is probably a mix of FPGA reimplementations of various chips and Buffee-like solutions where it makes sense. (And probably some raspberry pi zero trickery etc..)
Emulation will be there for sure but physical Amigas die hard ;-)

...and before you know it, our machines become the ship of Theseus...

[gulliver]

Money is the name of the game:

Set up a small company, associate yourself with a developer and customize to better integrate FS-UAE into a minimalistic custom raspberry pi distro.

License AmigaOS and buy a lot of raspberry pi 400. Add some marketing blurb and you will have a product to sell which will be quite similar to a modernised Amithlon.

The key is that it needs to be hassle free for the user. Just turn it on, and there you go.

[bjdcleary]

@gulliver

Will there be an "unofficial" add-ons pack at some point, like Better/BestWB?

[Bobson]

Originally Posted by gulliver

Money is the name of the game:

Set up a small company, associate yourself with a developer and customize to better integrate FS-UAE into a minimalistic custom raspberry pi distro.

License AmigaOS and buy a lot of raspberry pi 400. Add some marketing blurb and you will have a product to sell which will be quite similar to a modernised Amithlon.

The key is that it needs to be hassle free for the user. Just turn it on, and there you go.

Thanks! - and I guess I wouldn't even need to licence AmigaOS... just buy loads of copies, and sell t with the Pi 400's... ahh...!

This wouldn't meet the goal of running compiled ARM code within AmigaOS, but its a step towards it.

[gulliver]

Originally Posted by Bobson

Thanks! - and I guess I wouldn't even need to licence AmigaOS... just buy loads of copies, and sell t with the Pi 400's... ahh...!

This wouldn't meet the goal of running compiled ARM code within AmigaOS, but its a step towards it.

When you license, you buy in bulk so the price you pay is much lower than retail.

A developer can easily create a way to run from the Amiga side some ARM code by accesing the underlying Linux. This has been done numerous times already.

If you put some care in setting it up and complement it with other commercial Amiga packages you license too, you will certainly have a an interesting product if the price is right.

[Bobson]

Originally Posted by gulliver

When you license, you buy in bulk so the price you pay is much lower than retail.

A developer can easily create a way to run from the Amiga side some ARM code by accesing the underlying Linux. This has been done numerous times already.

If you put some care in setting it up and complement it with other commercial Amiga packages you license too, you will certainly have a an interesting product if the price is right.

This is true... apart from the odd utility thing in Rust, I'm basically a jumped up web developer (Elixir and JS). I've got my spreadsheets up - and putting a plan together.

[Tigerskunk]

Originally Posted by Bobson

People are willing to use a Raspberry Pi to keep their machines going... and we know the Raspberry Pi is more than capable of running AmigaOS under emulation... the emulation just isn't seamless - you're aware of it.

At the last count, over 3 million Raspberry Pi 4's have been sold...

If just 1% of that audience of 3 million (30000) could be captured... I don't know how many copies of OS 3.2 will be sold... but 30,000 is not to be sneezed at.

You are making the classical flawed assumption, that people who are not having nostalgic connections with it want to use Amiga OS for anything useful.

It's the old hen and egg argument that brings this idea down. You don't have enough coders/users to have a functioning ecosystem, and without the functioning ecosystem you don't attract new coders/users.

It's a small wonder that the remaining community was strong enough to create something like 3.2 and all the other stuff that is happening right now like new games, demos, little software projects, etc. But these people are not newcomers to the Amiga, but old fans who have been with the system for 35 years now usually.

[Bobson]

Originally Posted by Steril707

You are making the classical flawed assumption, that people who are not having nostalgic connections with it want to use Amiga OS for anything useful.

It's the old hen and egg argument that brings this idea down. You don't have enough coders/users to have a functioning ecosystem, and without the functioning ecosystem you don't attract new coders/users.

It's a small wonder that the remaining community was strong enough to create something like 3.2 and all the other stuff that is happening right now like new games, demos, little software projects, etc. But these people are not newcomers to the Amiga, but old fans who have been with the system for 35 years now usually.

This chicken and egg situation is true of any new platform or technology...

But I'm not assuming 'build it and they will come'. That never works. The primary purpose of this project would be to provide a future for AmigaOS on mass produced, cheaply available hardware.

If the experience is right (that is, the whole thing feels seamless) - and people have a reason to try it (eg - the Pi was primarily designed as an educational too - so combine it with learning resources designed to meet the curriculum), we may find some adoption. Even if that idea is flawed, I'm sure it will attract the same crowd who bought into things like the mini snes, mini mega drive etc...

- I'm not so deluded as to think that Amiga will take on Apple, or Windows (although, I think the way things are heading with Windows, it'll be linux in frock soon enough) and return to the glorious days... but I've also never wanted to give up on the idea of a modern Amiga, or expanding the user base.

I'm reading this... its a bit dry... https://lwn.net/Kernel/LDD3/

[acidbottle]

Interesting debate going on.

I concur with the logic that aside from reeling in some semi nostalgic former users, I fail to see the point in having a super future amiga with all the bells and whistles of a modern pc.

PC and MAC (and Rpi, FPGA etc etc) exist and have pros and cons, you pay your money and takes your pick.

The Amiga once was the leader of the pack in terms of music, gfx and development. The world moved on an Amiga shuffled about sideways and let the others pass.

Now we have numerous developers creating wonderful hardware and software, usually and mainly as hobby and not a full flowing business. Money is always a good motivator but you are not going to be a millionare Amiga developer these days! I believe many want to keep the classic feel going but with a mind of pushing boundries forward if possible. The Amiga will evolve but not as a direct competitor to anyone, but within its it own community.

The community market is there, pure hobbyist. I want hardware I couldnt obtain or afford back in the 90s, able to run anything I throw at it.

Im waffling on a bit actually, just want to say the idea of future Amiga is great but some of us just want a nice classic experience, products like 3.2 really helps with this (to keep this entire post on topic). Yep, think thats all I really wanted to say!

[Bobson]

Originally Posted by acidbottle

Interesting debate going on.
Im waffling on a bit actually, just want to say the idea of future Amiga is great but some of us just want a nice classic experience, products like 3.2 really helps with this (to keep this entire post on topic). Yep, think thats all I really wanted to say!

Who said anything about being a millionaire... I nearly came close once, until it all went spectacularly up in smoke...

So now I just work on stuff I'm interested in. Stuff that pays what I need to live. My only criteria is: Never ever run out of money... try not to make a loss.

It would be a classic - another machine to run 3.2 on... but with an opportunity to do more with it. Anyhow, its a moot point what you all think I'm investigating it... seeing if I can do it.

I spoke with a pal of mine who lead design of the SOC on the Raspberry Pi... just shooting ideas really... he said:

"To be honest I don't understand some people's interest in old OS's and games. It's just nostalgia. But lots of people are interested so I'm not knocking it. I suppose there is some educational benefit from them. They are more open and easier to use than Linux."

"Device drivers at kernel level are too low-level for you and too high-level for me ?. However the people round here love Amiga and like to get their old Amiga games running on Pis"

... he then sent me links to stuff and told me to get reading...

[Tigerskunk]

Originally Posted by acidbottle

Interesting debate going on.
Im waffling on a bit actually, just want to say the idea of future Amiga is great but some of us just want a nice classic experience, products like 3.2 really helps with this (to keep this entire post on topic). Yep, think thats all I really wanted to say!

That's basically what it has come down to. And the Amiga scene is the best it has ever been since the 90ies because of going back to its hardware roots.

Modernizing the hardware setup by changing the platform has been tried with AROS for x86 and OS 4.0 for PPC and failed. And not even it failed, it created huge rifts in the community by trying. Funny enough AROS has its biggest success since it found a home on good old 68K with the Vampire people pushing that angle.

You are better off using any of the other options (Windows, Linux and MacOS) to drive your daily business than with Amiga OS in its current state.

And for those who want to use an Amiga on their PI Systems, there is Pimiga. And if that isn't integrated into the hardware well enough for you, write software for the OS so it does. Amiga OS is quite flexible as you can see by the number of crazy Frankenstein systems around there and allows for such things.

[hth313]

Will the new 3.2 run as a superkickstart (if I fix the bonus code) on my 1.4 based A3000?