Kronos Quote:

Not really: https://www.broadcom.com/products/embedded-and-networking-processors/communications/bcm58712 If they used a smartphone chip (even a low end one) they would have a much much better GPU.

The RPi Foundation started with an old and popular smart phone/cell phone SoC to minimize cost. It locked them into 32 bit and the old VideoCore GPU which they have retained to maximize compatibility (32 bit Thumb2 code density is valuable on low footprint hardware too). The hardware choices were perhaps not the best choices but the RPi gained popularity because they aimed low in cost. The original goal was a low cost SBC for education and as a RISC OS successor. It was more successful than they expected. There were higher performance SoC choices then and certainly higher performance choices today. The communication SoC you pointed out may not have been available at the time the original RPi was released and it likely would have been too expensive for their needs including a larger and more expensive board for the PCIe Gen 2 x4. The Broadcom SoC is similar to the NXP PPC communication SoCs which ARM has replaced. Even A-Eon did not use the highest end PPC communication SoCs from NXP. There is a generation after the SoC used in the X5000. Cost is a major factor and all the SerDes add cost.

@Kronos

Quote:

I'd say if you were to try to build a team around a project "modernized and fully useable AROS based 68k system on rPI"

That's not actually my wish to be honest. What I want is a way to have a binary executable written for "Amiga" that's guaranteed to run on anything implementing "the standard". The standard being a compatible subset of 68020/68882 instructions (e.g. those that run well on real hardware and are still amenable to emulation, JIT in particular) and API libraries that are implemented by whatever the host is. That could be AROS (on any supported BE/32-bit), it could be AmigaOS3.x (on real, fpga or virtual), AmigaOS 4.x, MorphOS.

I have no wish to dictate what OS people run or the hardware they run it on, I would just like a unified standard that doesn't necessitate compiling for the whole menagerie of targets when all the most popular ones have robust 68K support.

I rail at PPC primarily because I see it as the absolute, inevitable kiss of death to two of the platforms I think have (or had) the most promise of advancing whilst retaining "seamless" 68K binary support.

_________________
Doing stupid things for fun...

@Kronos

Quote:

Kronos wrote: a team around a project "modernized and fully useable AROS based 68k system on rPI" ....

There was a developer who worked on AROS ARM big endian, using the possibility of this endian mode he also started working on an 68k emulator using JIT, so AROS ARM BE should get a transparent 68k emulation ... but then he invested more and more time in his 68k JIT.

And not they are bolting their RasPis in old 68k computer for mouse and keyboard connection ....

@OneTimer1

??? You can no longer play games on those amigas? Can't you benefit from whdload? The chip set ceased to work?

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

@Karlos

Quote:

Karlos wrote: @kolla Quote: OS4 is outdated and buggy, Roadshow for OS4 is outdated and buggy. Why bother. The only way to fix it is to have more users that you can't simply expect to be satisfied with the state of it. The only way to have more users is to make it more accessible and not hugely expensive. Nobody but a fool pays top dollar for a broken system running on hardware that will likely die and be irreplaceable. The only way to fix that it so target affordable hardware people won't feel completely ripped off over when they get a WIP system that's still rough around the edges but has mor chance of being supported. This is all blindingly obvious in a mirror universe where people fully appreciate the folly of PPC and aren't firecly loyal to a particular hardware architecture.

Good morning, class. Welcome to "Logic 101".

_________________
All the way, with 68k

@Kronos

Don't get me wrong, I would prefer the AMD-based x64 route. I love the AM4 platform. I have 6 machines in active use right now (3x B550 and 3x B450).
As I said, with good centralised management and funding, this is the better option for the long run.

Quote:

Kronos wrote: @Karlos Lots of things end up non trivial and if you wanted to catch them all you would not only create bloatware for real (which then won't run good enough on real Amigas)...

I think we are at the point where we can stop thinking about legacy Amiga hardware without at least an Apollo 080 in them. Or at least we should be.

There would have to be at least 5,000 Amiga 500s, 600s and 1200s with a V2/V4 Apollo card or PiStorm card. That is more than there are PPC machines running AmigaOS 4. If the OS 4 niche constitutes a development target, then the high performance 68k is that much more a legitmate target.


Quote:

Which is the reality, far less "heavy" SW has been released for über-fast 68k in the past 20+ years compared to PPC NG and I just don't see that changing no matter how "good" your arguments get.

That's not a good way to have a conversation/debate: Basically closing yourself off to any potentially reasonable arguments.

Nevertheless, we can't look at the last 20+ years for this. The kinds of high-performance 68k we're talking about has only really existed (uninterrupted) for the past 4-5 years.
It's where we've seen the rise in 080 systems, AROS 68k, and Hyperion releasing updates to Amiga OS 3 with back-ports from AmigaOS 4.
And the really fast stuff (emu68 on Raspberry Pi 4) only over the past 2 years.

Amithlon doesn't count because the market actively rejected x86, and actively pursued exotic hardware. Plus, H&P also didn't want it existing.
Today, emu68 is welcomed far more than the A1222. The market conditions favoring the development/porting of "heavy" SW for über-fast 68k is relatively new.

Unless there's a big swing and either AmigaOS 4 or MorphOS are ported to über-fast 68k, it will be a much slower process, but I still expect that we will soon see some "heavy" SW which only runs on 080+, or only runs "well" on emu68 RPi 3+.

_________________
All the way, with 68k

@agami

If you think Amithlon was rejected by the market then you couldn‘t be more wrong.

On the contrary it was the only time were über fast 68k had any relevance both with developers and users.

The issue is that Karlos does talk the talk but won’t do the walking involved.
If you want a system that is attractive for current NG users and developers you need to add features and „rich“ APIs that won’t work on a A1200HD with a 4 MB RAM expansion. If you want to support that A1200 you won’t be attractive to NG users and if you try to do both you end up with a fractured an inconsistent system just like 68k is today.

Hence little to no heavy SW in that category for the past 20 years.

Last edited by Kronos on 19-Dec-2023 at 05:11 AM.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@Kronos

Quote:

The issue is that Karlos does talk the talk but won’t do the walking involved

You say that, but I'm an actual contributor to the OS. It was, admittedly a long time ago now. And you you know how I got started? Fixing driver bugs and performance issues for the Phase5 hardware I had. And that moved on to NG hardware. It's probably fair to say, I sacrificed my "pure" NG machine to that effort, all the hardware swapping.

Now. Being a father of 3 young kids in eve of 2024 in an extremely demanding job means that my ability to do such things is basically in the past now. It takes me months to do what I could do in days before such is the constraint in my free time.

However, I really don't need to. It's already happening, whether you like it or not. The "Amiga" is now, well, the Amiga. Whether that's a physical commodore era machine with any number of regular and/or exotic hardware accelerators, entire compatibles and emulations.

The "Standard" is less well defined, but OS3.x is - and this can only happen in the bizarre reality we inhabit - being actively worked on including improvements and fixes previously only known to OS4. Yes, improvements and fixes, they do actually exist

For now, if you stick to an intersection functionality set, you can actually write system friendly software that actually runs on all of them, just about.

Agami is more of a optimist than me, I actually think OS4 and MorphOS will die a quiet lonely death here because whether you think the 68k market is fractured or not, every splinter has more potential than the each NG camp. And don't pretend they aren't equally splintered, just because NG camps tend to adopt a no questions asked just be grateful for whatever we get while it lasts mentality. I spent enough time on purist NG forums to know that.

Last edited by Karlos on 19-Dec-2023 at 06:15 AM.

_________________
Doing stupid things for fun...

Quote:

Kronos wrote: If you want a system that is attractive for current NG users and developers you need to add features and „rich“ APIs that won’t work on a A1200HD with a 4 MB RAM expansion. If you want to support that A1200 you won’t be attractive to NG users and if you try to do both you end up with a fractured an inconsistent system just like 68k is today.

We A L L should keep that in mind, even AOS4 could be seen as an example how bad modernized APIs (from AOS4) work on older (PowerUp) systems. You might be able to start them but a lot of modern (ported) games won't work well on old under powered PPC systems with less than 512MB RAM.

Actual operating system for desktops should have a TCP/IP with HTTPS and maybe with WiFi supporting the standard encryption methods, something that will hardly work on a 68060/50 and MP3/MPEG decoding will be hard even for 300MHz PPCs ... it stopped being an useful alternative years ago.

Quote:

Kronos wrote: If you think Amithlon was rejected by the market then you couldn‘t be more wrong. On the contrary it was the only time were über fast 68k had any relevance both with developers and users.

The Amithlon was an 68k/AOS3 Emulator/Wrapper running on Linux, the idea could be easily recreated using AROS 68k + running on an UAE, sold as a complete system ( Hard- + Software )

Nobody does it today, because standard WinUAE had gotten so much better and cheaper.

And if someone ever tries a recreation, he need the acceptance of the 'community' who does their ritual dances around names, legacy and wedge cases.

@OneTimer1

Amithlon was somewhere between that an a proper NG system.

At one point H&P reported 3000 units sold actual was most likely higher.
Despite all the FUD by self declared community leaders an entities trying to push their alternatives that wouldn’t be ready for years.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@OneTimer1

Quote:

The Amithlon was an 68k/AOS3 Emulator/Wrapper running on Linux, the idea could be easily recreated using AROS 68k + running on an UAE, sold as a complete system ( Hard- + Software )

Drivers

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

Quote:

Kronos wrote: Amithlon was somewhere between that an a proper NG system. At one point H&P reported 3000 units sold actual was most likely higher.

I never thought they sold this much ...

@OneTimer1

Back then there were 1000(0?)s of real "power" users. People who did more with their Amiga than w#nking of on the longest SysInfo bar or adding yet another eyecandy patch to their system.

People who had mostly left the chipset behind years before.

Amithlon was a dream come true for them, something that also was true on a smaller scale with NG later one.

Few of these real users remain and I fail to see how any of them would go along with Karlos vision (or lack thereof), hence anything 68k will die just like anything PPC or nonPPC NG. Maybe a few years later, but once biology has run it's course the only "Amiga" still relevant will be 7MHz, 1MB and Kick1.3 running some old games in a museum. Or in UAE for old people to relive the games grandpa was so fond about 50 years before....

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@Kronos

in my view there is a big distance between amiga based platforms like aros or morphos or amigaos and todays mainstream platforms. It is completely impossible to bridge that gap. This includes OS features, drivers and software. You would need a billionaire to have a chance there. Amiga today is retro and retro is 68k. 68k is quiet good covered today with both new hardware and software. The only project I see that has chances in NG is AxRuntime from Deadwood, merging Linux and Aros. Anything else has no chance to me because of lack of resources.

@Kronos

Quote:

Kronos wrote: Back then there were 1000(0?)s of real "power" users. People who did more with their Amiga than w#nking of on the longest SysInfo bar or adding yet another eyecandy patch to their system. People who had mostly left the chipset behind years before.

Yes I know, but I had the feeling an Amithlon would only cement the status quo and would not add much useful to AOS3.x. I was more of a MOS guy, I still have my Pegasos1 (!) in the attic. I didn't chose for love to the people behind it, i chose it because I knew what they where able to do.

But sadly MOS never got beyond the A-Box, it just became an updated AmigaOS so it never went where it could have been.

I don't know how good the actual MOS runs on AMD-64, they have gotten a new kernel and I don't know if Ralph is still a member of the MOS Team. The silence of the MOS developers does more damage than they might know, IMHO they should spread demo versions of the AMD-64 systems wherever they can, just to show they are alive.

Last edited by OneTimer1 on 19-Dec-2023 at 06:19 PM.
Last edited by OneTimer1 on 19-Dec-2023 at 06:15 PM.

@OlafS25

Good points and sure $1 Million Dollars () and a major backer would help. To me it is far more likely that the bounties will be driving it, via grass roots.

An Amiga Like system won't be a wintel box where drivers are everywhere, etc., but MorphOS have proven they can consistently improve things. 4 years ago I would have laughed if someone said that a modern browser and email client would be available or laugh even harder is someone said good CRM would be developed and offer features found almost no where else and we will see what incremental changes 3.19 will have as that is being worked on.

But look at where we are, donations have fueled development. Will I be able to go buy a brand new pc to use it for all options today, of course not but I don't care at this moment. I can do most of what I need to do all on one platform that is powerful (in comparison to desktop PPC) and cheap to acquire. So I'm very happy with the OS, apps, and most of all the various Team members who have treated me well. Wayfarer 7.2 just came out a few hours ago, the beat goes on.

Last edited by Matt3k on 19-Dec-2023 at 11:30 PM.

@OneTimer1

Sorry you haven't had a great history with the MorphOS Team. I think we all had issues in the 90's when the dirt was flying everywhere.

I actually like the silence of the Team with no empty promises and 2 more weeks stuff. They get it done when they can, and given the sheer number of releases over the last 25 years of MorphOS, I'm impressed they did it.

They talk though what is released, Wayfarer is on a 4 day old webkit as of 2 hours ago. That speaks loud and clear to me.

If I ever had a issue, Frank, Jaca, Fred, and others all have helped and I never felt abandoned.

They also speak via their bounties, the 3d drivers that Bigfoot is working on shows the roadmap if people want that route.

This is only my experience, sorry it wasn't yours...

Last edited by Matt3k on 19-Dec-2023 at 11:28 PM.

@Kronos

Quote:

Kronos wrote: @agami If you think Amithlon was rejected by the market then you couldn‘t be more wrong.

Simply stating it, does not make me wrong.
Even if I didn't personally live through the scene, hindsight analysis also shows that the market conditions were not favourable.

But regarding your heavy SW for high performance 68k remarks, I do get your main point:
If tomorrow the MorphOS team released their beta for MorphOS on x64 + SDK, there would be more developers interested in writing/porting software for that platform than for emu68 on RPi 3+.

A bird in the hand is worth two in the bush.
Since we have no idea if and when MorphOS x64 will be ready, the available über-68k at 800MHz+ with 2GB RAM + RTG and one day 3D graphics for under $200 is not a bad target.

All it needs is the first heavy app to get the proverbial snowball rolling.
You get a half-decent web browser on there, and the user-base doubles overnight.

_________________
All the way, with 68k

@agami

Quote:

A bird in the hand is worth two in the bush. Since we have no idea if and when MorphOS x64 will be ready.

Forget that, the whole point was that the Quark microkernel was supposed to be able to run multiple sandboxed operating systems. This was the killer feature. Yet, what we call MorphOS today is just the ABox, a single quark process. I think this is what OneTimer1 is alluding to.

Quote:

über-68k at 800MHz+ with 2GB RAM + RTG and one day 3D graphics for under $200 is not a bad target.

Damn straight.

There's something of a misconception among the NG crowd (sort of unjustified elitism if you like) that they are the sole inheritors of the "high end" Amiga. They completely forget how many Amiga Forever and AmiKit users there are, all of whom run high end 68K systems that, depending what they run it on, are just as performant as any NG machine.

Until recently, the only realistic option for that has been UAE, but PiStorm has changed that. Even without PiStorm and the quest for "ludicrous 68k speed", there have been a bunch of interesting 68K projects. You've got Warp, BFG and TerribleFire accelerator cards (for the absolute physical 68K purists).

Not everyone cares about "super high end" either. The 68K spectrum is more diverse than that. People have recreated the AA3000, complete with DSP (and devkit for it).

Last edited by Karlos on 20-Dec-2023 at 12:05 PM.
Last edited by Karlos on 20-Dec-2023 at 09:13 AM.

_________________
Doing stupid things for fun...