@amigang

And they say that most AmigaOS 4 users have left this forum.

What I want to see most?

Since you're asking:
- Hyperion to just die already
- PPC/Power ISA for personal computing to just die already
- MorphOS to go x86
- Amiga OS 3.x source opened up

But what I really want most of all, is a high performance 68k platform renaissance:
- Screw retro, most of that software is abandonware. Plenty of emulators for that.
- New dev kit
- New developers
- New software

_________________
All the way, with 68k

_________________
Fulfill newlight's Elite Narcissist Demands NOW Or He Will Send You To H3LL!

What I'd like to see in 2023?

A Maestrix update with more MorphOS MED fixes.
An X1Boot update with menu importing and multi installing.
A CIAgent sequel with audio emulation.
An X1000 USB Boot Creator.
A first release of Scripterface and the derivative Effectiface.

One can dream!

I'd really like 3D acceleration in qemu (plus other bits) so that MorphOs can run well on any hardware.

1. Hyperion make more money on Amiga Os 3.3.
2. Port uae with plugin ppc to pistorm so everyone may have Amiga Os 4.x on their classic machine.
3. szulc karlos szonwejs that want to switch to x86/arm/riscv etc
start working on working amiga gui (mui) clone.
4.morphos open source

Last edited by sibbi on 14-Jan-2023 at 08:36 PM.

@ppcamiga1

You never get it, do you? Who said I want to switch to x86 or ARM? My personal choice is for 68K. Just running on x64 or ARM. Or PPC for that matter. I don't really care, I just want a nice binary format that runs on real 68k hardware but allows AmigaOS and software written for it to become freed from hardware implementation worries.

I don't mind native x64, ARM or PPC OS providing those services through inbuilt emulation either, provided my binary runs as well as possible across the gamut of Amiga and amiga-like platforms. Never mind that the vast majority of Amiga specific software is closed source 68K to begin with.

I do feel the sum effort would be better spent, bringing it all back to 68K (the whole OS and any applications that can still be compiled from source) and allowing bare metal emulation on ARM and x64, where possible, to run it. I concede that's unlikely though because many people are heavily invested in their platform of choice.

So I'll settle for 68K and ABI spec for building cross platform binaries.

_________________
Doing stupid things for fun...

@Karlos

Quote:

Karlos wrote: @Lou What do you make of AxRuntime?

I don't think about it.
I guess the word 'native' wasn't as implied as I had thought...

@Amiga4000

Nice.

_________________
All the way, with 68k

@amigang & all

I'm with Amigang. . . I want it ALL but I will be happy if we can make it to 2024 without some major drama like OS4.x cancelled or another major lawsuit is announced. Fingers crossed

_________________
Please visit AmigaSource.com!!! Since 2001... Your BEST source for Amiga information. Again... *** Always Under Construction *** (Over 2000+ Amiga links for your viewing pleasure! -- Please help spread the word!)

@amigang

Thanks for sparking the debate!

For those who want bigger Amiga shows, PM me about AmiWest 2023!

@agami

Quote:

agami wrote: @Amiga4000 Nice.

THANKS!

_________________
Fulfill newlight's Elite Narcissist Demands NOW Or He Will Send You To H3LL!

@Lou

The reason I ask is the idea that AROS source compatible code is now compilable for an environment that, at least on paper, promises all the SMP, memory protection, modern hardware support, etc that most people would agree are an MVP for an OS today, by sidestepping the elephant in the ROM, so to speak. Except it's a bittersweet thing because in order to do so, you've basically cast aside AROS itself to achieve that end.

I don't quite know how I feel about that.

_________________
Doing stupid things for fun...

One thing I'd definitely love to see in 2023 is a working A1200 wedge, on my desk, driving a 1080p RTG display (or better) banging out 68K code at hundreds (or thousands) of MIPs/MFLOPs on a PiStorm while I try to think of ways to gain native access to the additional cores in the ARM processor so that I can build virtual hardware devices that run on it. First being a nice midi controllable soft synth. And I mean synth, not just a rompler.

_________________
Doing stupid things for fun...

@Karlos

Quote:

One thing I'd definitely love to see in 2023 is a working A1200 wedge, on my desk, driving a 1080p RTG display (or better) banging out 68K code at hundreds (or thousands) of MIPs/MFLOPs on a PiStorm while I try to think of ways to gain native access to the additional cores in the ARM processor so that I can build virtual hardware devices that run on it. First being a nice midi controllable soft synth. And I mean synth, not just a rompler.

OpenCL?

_________________
Support AROS sponsor a developer.

"AROS is prolly illegal ~ Evert Carton" intentionally quoted out of context for dramatic effect

@terminills

Interesting but not nearly as important to the task at hand as you might think. Parallel audio synthesis is not quite the same as other parallel tasks. An exception might be many notes playing on the same unit being generated using SIMT (as distinct from SIMD) but unless you're always doing mass polyphony in that way, it's a lot of complexity for little gain. It might be better for some effects processing but in the end, vanilla scalar and SIMD with multiple regular threads seems to work better. At least in my experience.

_________________
Doing stupid things for fun...

@Karlos

I want to see AxRt on 68k Linux with a Pistorm32 and I expect someone else to compile it all and build a downloadable drive image for us all to use because that’s the true Amigan way!

@all
I want the Apocalypse and total annihilation of Life on Earth!

Last edited by QBit on 10-Jan-2023 at 05:08 PM.
Last edited by QBit on 10-Jan-2023 at 05:07 PM.

@QBit

You could just come to the UK and visit Sunderland for a few days and get it out of your system...

_________________
Doing stupid things for fun...

@Karlos

Quote:

Karlos wrote: @Lou The reason I ask is the idea that AROS source compatible code is now compilable for an environment that, at least on paper, promises all the SMP, memory protection, modern hardware support, etc that most people would agree are an MVP for an OS today, by sidestepping the elephant in the ROM, so to speak. Except it's a bittersweet thing because in order to do so, you've basically cast aside AROS itself to achieve that end. I don't quite know how I feel about that.

I think we can get all those things in AROS without becoming a SaaS OS.
I'm fine with breaking compatibility in order to progress since a native AROSx64-MP would be able to and can already run any Amiga (excluding the PPC drivel) flavor in a sandbox. Actually wait - does E-UAE emulate PPC cards yet?

What's holding the Amiga-ish platform back is power and price. Going x64 solves that.
Without power, modern apps aren't coming. Without a low price, no one is buying.

@Lou

I would say what's holding the platform back is the belief it is still relevant outside of hobbyists. I know it sounds awful but that's why so much of it is cloistered away in closed platform choices that are frankly going nowhere.

I don't think the platform today solves anything for which there aren't already free alternatives with years, or even decades of head start already. Sure many dormer Amiga users may dislike existing GNU. That's where something like AxRuntime might step in. Unlike AROS proper, there's no need to port large, complex applications. They will just run out of the box.

Last edited by Karlos on 11-Jan-2023 at 05:59 PM.

_________________
Doing stupid things for fun...