@OlafS25

Quote:

OlafS25 wrote: @Wumpus why is Aros weak on 68k?

As I've said elsewhere, seamlessly handling existing 68k applicaitons is a huge sticking point for me on Amiga-like OSes, otherwise they don't have much appeal to me and have to stand on their own merits against Mac, Linux and Windows as a new platform.

I'll qualify my answer in that I haven't tried it recently, so my information may be out of date.

The performance running on 68k systems has been pretty bad, so that makes it unsuitable for most 68k systems as a daily OS. That to me was one of the main selling points for AROS, as a replacement open OS for 68k.

Since it's the only one targeting those systems, it's sad to see it not perform.

Also 68k applications on x86 for example aren't first-class citizens, more like single-use emulations with separate OS if I'm not mistaken.

If it can't really run them in the OS like normal applications, then you might as well extend Vamos on Linux, Mac or Windows. At that point there's little special about the host OS.

This is a far cry from Amithlon and especially MorphOS where you can mix and match libraries for different architectures and never notice an application is built for 68k.

I realize those also have a 68k JIT involved, but the integration is far better from an end user perspective.

@remotenemesis

Quote:

@kolla Quote: Rather they had plans to follow Microsoft and license NT as the core for the next iteration of Amiga systems. Interesting! I've not heard that before. Do you have any more information or some references for me to read?

Rather common knowledge... google for Commodore, Hombre, PA-RISC...

https://en.wikipedia.org/wiki/Amiga_Hombre_chipset

Quote:

BeOS would have been a truly elegant choice for a second-wave of mid-90's Amigas.

What do you mean "would have been"? BeOS was there, anyone interested could just get it - I don't see how attaching the name "Amiga" on it would have made it any better, rather the contrary. And BeOS is still here, in the shape of Haiku.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@Wumpus

Quote:

the integration is far better from an end user perspective.

The "integration" involves programs being able to share, read and write, into each other's memory. That's why you can never get any "modern" iteration of AmigaOS that is able to "integrate" with legacy programs - it's a design that by its nature breaks all security. This model was bad enough back in the golden days of Amiga, with all the viruses etc, being able to write to boot blocks etc, when Internet came, it was obvious for every serious developer that AmigaOS as we know it, had no future.

There was an opportunity for someone to recreate the "Amiga user experience" in a modern shape, reimplement the components of the user experience in a clean and secure modern way, and there was a few attempts, but they all failed. And then there were thousands of users who screamed loudly about compatibility, and a few developers who find joy and comfort in programming "the old school way" (or who don't know any other way).

Last edited by kolla on 16-Apr-2018 at 09:59 PM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@remotenemesis

Quote:

Its not a huge stretch to modernize AmigaOS with enough people, a clear vision and affordable _reference hardware_.

If you'e talking about PPC reference hardware, then that's a grim situation. Trevor had to get pretty creative with the A1222, and not everyone is happy with his choices. If there is a PPC Amiga after the A1222, it will probably end up being a re-purposed Samsung toaster oven.

If you're talking about reference hardware for a new pie-in-the-sky Amiga-like system, then you probably couldn't do much better than the Raspberry Pi 3B+. It's not powerful, but it's adequate for many purposes, it's cheap, and there are millions of them out there.

Quote:

I've never really enjoyed Windows and while my main computers are MacBook Pros... they just aren't fun in the same way Amigas were.

I think the same, which is why I switched to Linux almost 20 years ago. It's not as much fun as Amiga -- it's too complicated, for one thing -- but it's a lot better than Windows. If I had to use Windows, I wouldn't -- I'd find another line of work.

Update: learning to spell...

Last edited by bison on 16-Apr-2018 at 10:09 PM.
Last edited by bison on 16-Apr-2018 at 10:08 PM.
Last edited by bison on 16-Apr-2018 at 10:08 PM.

_________________
"Unix is supposed to fix that." -- Jay Miner

@kolla

I agree with you, but this thread is derailing into the what is an Amiga thread. Sorry about that.

My outlook is that if the old applications aren't a priority, then I'll stick to something already modern and stable like Linux.

There nothing wrong with wanting a clean slate with all new applications, but original applications are where I personally get hung up.

I don't expect many new applications to appear and most of the applications that do are just ported from Linux anyway. I could run that on any OS.

The original applications are what makes it unique to me now, not the security features or API.

@Wumpus

Quote:

...is thread is derailing into the what is an Amiga thread.

Any thread that goes past 2 or 3 pages derails into a "What is an Amiga" thread, with the exception of those that start that way.

_________________
"Unix is supposed to fix that." -- Jay Miner

Hmmmm I wonder if a "modern" The Hombre could be built with a "standalone" Vampire on a PCIe card...

Hardware 68k "simulation" would not be slow.

I can dream.

@Wumpus

MorphOS or AmigaOS are not even running on legacy hardware

Yes of course Aros is much slower on old hardware than 3.X but it is a little unfair to compare both. 3.X was optimized for amiga 68k hardware whereas Aros was programmed on modern hardware. But there is still room for improvement. Also you seem to mix desktop and API here. Desktop (Wanderer) was slow but you can also use Magellan and Scalos and Scalos is faster than Wanderer. On API level there are certainly differences too but that has to be seen. You can mix software and libraries compiled for amiga with aros 68k without any problems. I use that in my distribution. So the whole picture is from my view more complicated.

@OlafS25

It really is complicated.

The original point though was that if the different factions had worked together, they would all have been better off by incorporating each of their strengths rather than each spending developer time on reinventing the same wheel and splitting the user base.

OS4's access to the source code
MorphOS 68k compatibility
AROS's hardware support and openness
Amithlon's JIT and native CPU performance/integration trickery

All their developers working together would have been a bigger software team than Commodore ever had.

And the community would have one place to focus their attention, develop applications and possibly spend money.

Hell, the time wasted on OS camp flame war posts alone could have bought us a web browser to beat Chrome and Firefox lol

@Wumpus

Aros has no big problems regarding 68k compatibility as I should know ;)

Yes all working together would be good but for that to happen you would need to brainwash a lot of people ;)

other than that it would be more propable that they would start a fight

There is no chance for that

@Wumpus

You are not going to convince people to give up on their chosen Amiga platform, most of them stick to it with a religious ferver. Even if you could convert them, you have various companies figthing over what remained of the Amiga. Everybody is trying to be the big fish in a small pond. Sorry if this sounds depressing, but it is what it is.

_________________
This is just like television, only you can see much further.

@BSzili

i think it is the attitudes that divide the audience. the categories are maybe not very just, but somewhere around
1. do you want elitist quality, then morphos is your choice
2. do you want maximal independence and possibility to bring in yourself, then aros is your choice
3. do you want to try to preserve the label in a progressing it context, then os4 is your choice
4. would you have to realize all this wont get far and you would like to preserve the common denominator as long as it lasts, choose amiga.

my choice has been a combination of 2 and 4. but others may at least seem to be mutually exclusive. whatever..

@BSzili

Quote:

BSzili wrote: @Wumpus You are not going to convince people to give up on their chosen Amiga platform, most of them stick to it with a religious ferver. Even if you could convert them, you have various companies figthing over what remained of the Amiga. Everybody is trying to be the big fish in a small pond. Sorry if this sounds depressing, but it is what it is.

It is depressing.

I probably won't convince anyone but I do see several that agree. At best maybe it's food for thought.

In the context of the original question, I do think this hurt us more than anything else.

@wawa

Quote:

wawa wrote: @BSzili i think it is the attitudes that divide the audience. the categories are maybe not very just, but somewhere around 1. do you want elitist quality, then morphos is your choice 2. do you want maximal independence and possibility to bring in yourself, then aros is your choice 3. do you want to try to preserve the label in a progressing it context, then os4 is your choice 4. would you have to realize all this wont get far and you would like to preserve the common denominator as long as it lasts, choose amiga. my choice has been a combination of 2 and 4. but others may at least seem to be mutually exclusive. whatever..

I've been accused of being a MorphOS shill and I'm far from elitist. I don't like the fact I can't buy new hardware for it or that it's closed source and destined to be abandoned eventually, but it runs my old apps well and is pretty slick for native ones.

I've been accused of running down AROS, but I'm all for platform choice and openness. It has at least partial ports to more hardware than any of the others and I've said open source was a priority of mine in almost every post.

These two also share a lot of code, so I have to give them credit for that.

Amithlon is maybe half-closed source, but runs on commodity hardware and is crazy fast. It hits a pretty good balance for my daily use, but it's abandoned and still way too closed.

I can't afford to have an opinion on OS4.

We've all got different priorities and they don't all fit cleanly into categories, but following or hating based on which camp wrote it and not how it suits your usage isn't smart or helpful.

The heart has reasons reason knows not.

@Wumpus

If discussion is the only goal then sure, you can get several people to agree. The "unity" threads tend to blow up to dozens of pages, but and the end of the day they won't change anything. Don't get me wrong, I'm not trying to shoot down discourse, this is just a reality check. The users and the companies are far too invested in their platform, but I'm repeating myself.

_________________
This is just like television, only you can see much further.

@Wumpus

just lately i have put up the repository to update osyssey code after deadwood has left. my intention is to keep it working for all platforms. sure i concentrate on aros, and my preferred platform amiga-m68k, but i want to merge code for morphos and os4 as well, in particular so much sought for endian fixes to js engine. currently x86 and ppc targets compile, but i dont know if ppc works because i dont yet have the way to emulate a sam460 properly.

now, there is a thread about it i have opened here, but the interest was moderate and it fell from the front page while members are furiously disputing "why there is no cooperation" and "what is blocking amiga" and their dreams and visions and how they are waiting for some next update or some killer app.

go figure.

Last edited by wawa on 18-Apr-2018 at 11:28 AM.

@bison

Maybe Hyperion already implemented it, who knows?

_________________
retired

@Wumpus

I think 68k applications are fairly "integrated" and seamless in AROS, provided you configure the emulation layer properly for Coherency Mode: http://vmwaros.blogspot.com/p/amibridge.html
Naturally it's always going to be running an emulator underneath, since it's a completely different hardware architecture.

@OlafS25

Quote:

Yes all working together would be good but for that to happen you would need to brainwash a lot of people ;)

Well said.

This thread is mostly a rehash of what has been said before, many times, but the interesting thing to me are the poll results, which are spread fairly evenly among all seven choices. The Amiga community is small, but it is also very divergent as far as opinions and goals, which makes it even less likely that anything will come of it in the end.

@Wumpus

Quote:

It is depressing.

The fact that there's an Amiga community at all, after all these years, is more amazing than depressing, at least for me. If I were a sociologist I think I would find this phenomenon interesting.

_________________
"Unix is supposed to fix that." -- Jay Miner