Maybe my message got lost somewhere. I meant to ask Olaf why he thinks it has been said MorphOS for another CPU will be totally different from what it is now?

> Maybe my message got lost somewhere.

It didn't with me at least. Seems Jim thought you were OlafSch :-)

Quote:

because the developers like geit say that:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=9204&forum=11#100844

No PPC integration, no 68k except UAE (like done in Aros), incompatible API (you could also say modernized). I did not find the posting about dropping old stuff like datatypes but I read that.

Ambient certainly will still be there but many other things are changed. Or plans have changed... who knows

"SMP, memory protection, 64Bit support, virtual memory will come with a price. Loosing the compatibly to all thirdparty PPC and all 68K applications, which aren´t in development anymore. It also will split the user/market once again. So a step like this should be made wise and preferable only once. And that is the point, where MorphOS swaps hardware. Such opportunity must also be used to drop insane and old crap the OS is filled with to keep it compatible with old Amiga and the old AmigaAPI simply is stupid in many ways."



[ Editiert durch OlafSch 01.06.2016 - 18:20 ]

Quote:

Andreas, you are clairvoyant.
That is what I get for posting right after waking.

>Or plans have changed... who knows

Only the developers.
But they have a history of dropping surprises in our lap.

>I did not find the posting about dropping old stuff like datatypes but I read that.

Nothing is sure until we see it.

This whole direction (for the thread) is rather pointless to me as I only care about MorphOS apps that don't rely on 3.1 compatibility.
I have Amigas to run 3.1 software, I will probably get stuck with a copy of OS4.1 if I buy an X5000, and I still have PPC MorphOS to fall back on.

>"...Such opportunity must also be used to drop insane and old crap the OS is filled with to keep it compatible with old Amiga and the old AmigaAPI simply is stupid in many ways."

As stated repeatedly, I'm fine with that.

The only things that will be changed/dropped are those things that we have to change/drop in order to reach our goals.

Binary compatibility will be non-existing, which to elaborate means that already existing programs will only run if they're adapted to run on the new system. Luckily from a source code compatibility point of view, we're aiming to be as compatible with the current MorphOS as possible.

The things that will be changed/dropped are minor and on the relatively low level side of things. There's no reason why Datatypes would disappear, nor any other particular current component of MorphOS, and from a user point of view everything will work like you're used to, just better.

Quote:

I agree with that question, and it's not the first time he makes these wild assumptions, and it's not the first time his assumptions are questioned.

Think about it. MorphOS is a huge project. A lot of code, and 1.5+ decade of work by lots of people plowed into its ground. To assume that all this would be scrapped in favor of something new and completely different is both crazy, and frankly unrealistic. MorphOS is MorphOS. It is today what its developers created, and the creation reflects their personal interests. If anything, it will evolve, not be remade. Since MorphOS is an Amiga reimplementation, it means that an evolved MorphOS will be an evolved Amiga reimplementation. A dropped ambition to be as backward compatible as possible to Amiga 68k in favor of an ISA change and more modern features does in no way imply that "it will not have much left from amiga", and since it still will build upon the Amiga reimplementation it is today, it will not be like a "changed Linux with different icons and wallpapers".

[ Edited by takemehomegrandma 01.06.2016 - 22:30 ]

@ bigfoot

Sounds good!

> There's no reason why Datatypes would disappear

There have been recentish statements from your colleagues that could be interpreted otherwise:

"Datatypes [...] are mostly broken by design and Reggae is so much easier and flexible in any way."
http://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=11382&start=86

"basically only some shell commands and libraries/datatypes are based on AROS. Many of those are only present to be backward compatible even that there is a replacement already. Since compatiblity with 68K is no longer an issue in the future those likely will be dropped. datatypes [...] for example are basically already quite obsolete. [...] Reggae is a nice replacement for datatypes."
http://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=11372&start=54

"datatypes suck, imo ;) There's a reason for Reggae."
http://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=11301&start=30

Big Foot, and/or the other members of the MorphOS Dev. Team might have changed their minds about completely dropping support for datatypes, given his more recent statement in this thread. Or, there might be some disagreement between members of the team regarding carrying support for datatypes forward into MorphOS for x64. We should wait and see. The only people it could affect if datatypes are going to be dropped, are programmers who prefer to use datatypes, instead of reggae, and until the MorphOS Dev. Team releases more information and/or an SDK for x64 version of MorphOS, no outside programmers will be working on porting software to the new MorphOS for x64.

Aros x86 is worth nothing crap.
Aros x86 is not source and binary compatible and is not modern for example aros has not memory protection, resource tracking, suport for multicore.

Aros 68k is attempt to clone Amiga Os 68k.
Aros 68k with many patches and libraries made for Amiga Os 68k is enough source and binary compatible with Amiga Os 68k.

With some work aros 68k is nice os almost as good as Amiga Os 68k.

The one and only problem with Aros 68k is speed.
Fast 68k cpu not exist and Aros 68k is usable only on uae.

It is sad but after many years of lies and cheating gunnar von boehn and his followers still does not provide fast 68k cpu.
Natami/vampire crap is still slower than 68060 80 MHz and this is not enough to run aros 68k fast.

Aros 68k is best proof that amiga need better cpu than 68k and it have to be not x86 because x86 is not big endian

If Morphos on x86 be something like aros x86 it will end like aros x86.
Morphos for x86 should be boopsi,mui,datatypes on unix.
Morphos for x86 will be not binary compatible with 68k Amiga Os,
everything need to be recompiled, it is time to drop 90% of old API to get things like memory protection, resource tracking, suport for multicore, posix compatybility.

Quote:

Could. Especially if you have an agenda, and if you narrow down your view from the bigger picture of what has really been said, down to single sentences or words.

Quote:

And all it says is that Reggae is better. Which is kind of true, isn't it? And that's all it says.

Quote:

Well, this is also true. New and better stuff (like Reggae) is already in place and could replace old obsolete stuff (datatypes). Running programs on the new MorphOS requires adaption and recompiling anyway, meaning an opportunity to adapt to the better Reggae as well. And the key word here is "likely", meaning no decision but reflecting the views of the writer.

Quote:

This is also true, datatypes suck in comparison to Reggae, which is the reason to why Reggae exists. And that's all it say.

Personally, I'm all for dropping datatypes in favor of Reggae. That is long overdue. IMO, Reggae has not been getting the attention it deserves despite being so much better, and has had to live in the shadow of datatypes, where it couldn't grow the way it should, mostly because of Amiga compatibility reasons probably.

So when existing programs needs adaption anyway, it would be a good time to remove this and maybe other obsolete parts as well, so that the new MorphOS becomes as good as possible.

SoundBankster is a great example of what becomes possible when using the bundle of MorphOS exclusive, new technologies, instead of trying to rely on similar but obsolete (or non-existing) Amiga stuff, like datatypes.

http://obligement.free.fr/gfx2/soundbanksterreboot_beta.png

Let us see more of that kind of SW, harnessing the new powerful built-in functions of MorphOS!

Quote:

Aros X86 has modern browser all other platforms not have... that are the two sides of the coin, it is almost impossible to port modern software to PPC (or 68k) because of endian issues. Yes Aros basically inherits the conceptual disadvantages of 3.X but that is true for all current platforms. That might be different for MorphOS after ISA change but (like always) it will come with a price. You will then say it is not amiga anymore, then you can use linux too (how true or wrong this might be).

Aros has some problems when running on slow 68k hardware, it is obvious that it was created for PCs with graphic cards, f.e. graphic output is slower with less colours in opposition to 3.X. The reason propably is that it calculates in 24bit internally. It is compiled with GCC (like all versions) but GCC supports 68k not very good today. Featurewise it offers much more than 3.X with many components and patches included by default. Compatibility is good, good means many applications and games work but not all. These problems certainly could be resolved when more people would use it and report problems. Regarding "Vampire crap", talk to the people using it with 3.X. I do not blame Gunnar that Aros at the moment is not perfectly running on it. The only problem is that both Gunnar/Apollo team and most users prefer 3.X so for me it is no longer interesting. I am happy with running it in UAE on modern hardware anyway and do not need "real hardware" to be happy. In emulation I can even outperform some of your beloved PPC platforms.

[ Editiert durch OlafSch 02.06.2016 - 11:44 ]

>>> There's no reason why Datatypes would disappear

>> There have been recentish statements from your colleagues that could be
>> interpreted otherwise

> Could. Especially if you have an agenda

Huh? You're really insinuating an agenda when "broken by design", "only present to be backward compatible", "likely will be dropped", "already quite obsolete", "datatypes suck" etc. are interpreted as reasons "why Datatypes would disappear" in MorphOS NG? What's *your* agenda again?

> and if you narrow down your view from the bigger picture of what
> has really been said, down to single sentences or words.

So please enlighten me. What is the "the bigger picture of what has really been said" in geit's and jacadcaps' postings I quoted from regarding datatypes if not that there *are* good "reasons why Datatypes would disappear" in MorphOS NG?

>> "Datatypes [...] are mostly broken by design and Reggae is so much
>> easier and flexible in any way."

> all it says is that Reggae is better. [...] that's all it says. ;-)

...if you ignore the first part of the sentence where geit apparently gives a "reason why Datatypes would disappear" in MorphOS NG. It helps to know that geit's comment was a response to ppcamiga1's proposal to preserve the old datatypes concept in MorphOS NG (comment #81).

>> "Many of those are only present to be backward compatible even that there is
>> a replacement already. Since compatiblity with 68K is no longer an issue in the
>> future those likely will be dropped. datatypes [...] for example are basically
>> already quite obsolete. [...] Reggae is a nice replacement for datatypes"

> Well, this is also true.

Datatypes "likely will be dropped" in MorphOS NG is true? Even I wouldn't go as far as making such definite claim, especially considering bigfoot's statement.
And when you say geit's points are true, how can you say at the same time that he gives "no reason why Datatypes would disappear" in MorphOS NG?

> the key word here is "likely", meaning no decision but reflecting
> the views of the writer. ;-)

Who is talking about any "decision" here other than you? My posting you replied to was about whether or not any "reason why Datatypes would disappear" in MorphOS NG have been mentioned by MorphOS team members (i.e. "views of the writer").

>> "datatypes suck, imo ;) There's a reason for Reggae."

> This is also true, datatypes suck in comparison to Reggae, which
> is the reason to why Reggae exists. And that's all it say. ;-)

And you don't think that jacadcaps gives here a "reason why Datatypes would disappear" in MorphOS NG? It helps to know that jacadcaps' comment was a response to eliot's wish to preserve the old datatypes concept in MorphOS NG (comment #23).

Quote:

You will use it after dropping 90% of API then?

@Andreas

thanks for digging out the statements...

that was what I remembered and not a fool or talking phantasy like some here like to claim :)

The partly opposite statements seem to indicate to me that they not even have agreed about the direction for the ISA change

@takemehomegrandma

dropping components like datatypes makes porting applications even more complicated or impossible. I do not see MorphOS flooded with new software and even less if it splits in "NG" and "NG NG" with incompatible APIs

Andreas certainly has no "agenda" or is citing statements of MorphOS devs that not fit to your views be a problem for you? if you have better informations we all not have then you are free to give them to us. Otherwise I would say you have a agenda too

As I wrote perhaps they have not yet agreed to something so they are expressing the different views. I have explained why I think that "NG NG" is not automatically becoming successful without lots of software support (apps) that people today expect and that current community is more retro orientated and might a new platform even not use just because it is too "modern" to them. Also dropping components that are available and used on all platforms is a double-edged sword because it makes porting to MorphOS almost impossible. You are praising "MorphOS only" software but how much is there and how much will be developed? A number of existing PPC software propably is not available in source so cannot be changed and recompiled. All that might end you with a modern platform but only a basic set of software from MorphOS team developers. That will hardly convince new users to join MorphOS and even pay for it.

[ Editiert durch OlafSch 02.06.2016 - 11:59 ]

> Aros x86 is not source and binary compatible

AROS (no matter the ISA) is source-compatible with AmigaOS 3.1.

> aros has not memory protection, resource tracking, suport for multicore.

Yes, that's because it is source-compatible with AmigaOS 3.1.

> Aros 68k with many patches and libraries made for Amiga Os 68k is
> enough source and binary compatible with Amiga Os 68k.

AROS/m68k doesn't need any "patches and libraries made for Amiga Os 68k" to be source-compatible and binary-compatible with AmigaOS/m68k.

> vampire [...] is still slower than 68060 80 MHz

That's not true from what I can see from reports and videos. Vampire V2 seems about as fast as an accelerator with 120 to 175 MHz 68060.

http://www.apollo-accelerators.com/wiki/doku.php?id=minibench
http://minibench.apollo-accelerators.com

Quote:

I might even be tempted to take up coding again.
Frankly, I never wanted to try to familiarize myself with the 3.1 API in the first place.

@Jim

I did ask him... if I remember right he was praising AmigaOS and now he recommends MorphOS to drop the API

And yes if you make a clean start and accept a very different API you can make something more modern

But I doubt that most existing amiga fans would like it, porting would be more or less impossible and I do not see developers rushing to develop new MorphOS software then

But of course all speculating we do not know what team will make, perhaps they not even know it themselves

The point is that since it will be incompatible anyway, why keep all the legacy problems too? Give people a new reason to use the OS besides using 20 year old software or program like you did 20 years ago.

There are a lot of people saying the same thing: they want an AmigaOS but modern so let's drop the legacy software and be done with it. Those who are only interested in 3.1 will never go NG anyway so their opinions don't count.

I don't see why dropping datatypes would mean that this new OS would become basically a Linux with an AmigaOS skin. Most things would still be there, including the kernel. I think you are reading in too much with what has been said.

@bigfoot

I so hope I will experence this new OS someday not in the far future

Quote:

Not a lot more complicated, and certainly not impossible; if you have enough knowledge to understand how to use datatypes for loading/saving media, it shouldn't be that much of a deal to replace it with Reggae (many well written examples exists), and it opens up a lot of possibilities as a bonus.

Datatypes is obsolete. It should go.

Hopefully it does.


Quote:

I meant *you*, not Andreas...


Quote:

The definition of "success" is kind of crucial in that statement. I think nobody is expecting MorphOS to be a general desktop competitor to Windows or Mac anytime soon, so that is probably not a goal to measure success for anyone. But today MorphOS is tied to PPC, and that platform is rapidly dying. MorphOS has already covered everything worth supporting regarding PPC HW, and MorphOS hit the Amiga legacy boundaries many years ago. MorphOS has come as far as it possibly can on PPC. The alternative to "NG NG" is stagnation and death, and because of this I think most of us will define a continued life and evolution of MorphOS on a new ISA as a success in itself.



Quote:

This may apply to the Amiga retro crowd, but hardly to the MorphOS community. I doubt any MorphOS users ever chose MorphOS for retro reasons (the same goes for OS4 I think). We all had Amigas already, what we wanted was the Next Generation, not the past. The retro segment is a completely different one. The problem with current MorphOS is that it turned out to be neither really retro, nor really modern either. It's stuck in the middle. On a sinking 2005 level PPC ship. A MorphOS NG will hopefully cure this.


Quote:

I'm praising SW that makes use of all the new features introduced by MorphOS, since this was the reason I chose MorphOS over a Workbench 3.1 based Amiga retro system in the first place. Advocating that SW should restrain itself and only use old standards goes totally against everything I want from MorphOS! SoundBankster is a shining bright example of SW that depends on things like MUI and Reggae as well as the upcoming MIDI and isochronous USB. I don't care if Amiga 68k don't have these features, I'm glad MorphOS does (will have). This is an example of how new and evolved features are enabling development of SW that wouldn't be possible otherwise. And hopefully more SW will be developed for a MorphOS that is alive and evolving, than a MorphOS that is stagnating on a dead PPC platform. Your option of just rolling over and die "because it will probably not be successful" seems less feasible to me. I'm glad the MorphOS Team is trying, and hopefully good things will come out of it.


Quote:

Quite possible. So MorphOS should just roll over and die, instead of at least trying to go forward?

MorphOS will have all important Internet applications, it will play movies and music, it will have competent text editors, gfx editors, etc, from the start. It will have a new SDK with a new IDE, and it will have new options and possibilities for non-retro oriented Amiga SW developers that simply aren't here today. Hopefully developers will appreciate this and find good use for it.

> using 20 year old software or program like you did 20 years ago

Amiga software from the 1970s? ;-)

> Most things would still be there, including the kernel.

And the Quark kernel isn't even Amiga-ish ;-)