Critical components to open source

[Kalamatee]

Originally Posted by IridiumFX

So, back on topic, given Exec is tightly bound to the underlying architecture, Full-asm and not extremely useful for long term evolution, if we could get the legal rights to touch the OS code, I'd suggest Starting with a c backport of it.

Do you have any idea of the amount of work involved? And sorry to burst your bubble but some critical parts will still be written in ASM.

I always though that, with an arbitrary simplification of the concept, an Amiga can be fully define by the tuple { Exec, Dos, Intuition, Graphics }

No that certainly doesn't define an Amiga - it defines a subset of AmigaOS that wouldn't even run together without a bunch of other components.

[wawa]

Originally Posted by idrougge

You're guessing wrong. OS 3.9 in ROM eats up 67 kB of chip RAM and 172 kB of fast RAM when booting without startup-sequence. That's still a lot, but astronomically smaller than 1,5 MB.

And no-one has proposed to port OS4 to 68k, so I'll just ignore that.

people repeatedly do talk about it. in here and elsewhere. currently there have been some wishful thinking ot on a1k vampire thread. someone managed even to boot to os4 early startup on uae with only 7mb occupied!

[bubbob42]

You should at least appear to be open about any possible solution, wawa.

When I met Aaron Digulla at the Cologne Amiga fair back in '9x, he and others had just started the AROS project. I personally gave him all the credit he deserved back then and I've been checking out AROS releases regularly ever since. I will accept AROS as my solution of choice once it looks, feels and reacts on my machines like OS3.9 does. Until then, any alternative is welcome, be it a stripped down OS4 on 68k or just the original. Neither I nor many others are willing to spend lots of time (which we don't have) testing or writing bug reports for something which we already have in its original form to build our own projects upon. That does not mean we won't appreciate your efforts or those of the AROS contributors. I look at the AROS source from time to time, comparing implementations and learning how things work. Which is helpful and interesting, but that's about it. Please accept that.

No one is trying to put AROS in a corner. There are different tastes, though.

[IridiumFX]

I am not picking the fight, Kalamatee. Indeed I am an AROS lover almost as much as I am an Amiga lover. That said ... yes, I have an idea of the related amount of work. Yes, I know that Intuition is All C except a few asm and bcpl headers, yes I know that DOS is a mess, and that Graphics is mostly bound to the custom chips, thanks. As for the rest, I invite you to consider the meaning of "arbitrary simplification of the concept". if I could open-source only a limited number of components, I'd gladly get the mentioned ones.

You are doing an amazing job on AROS. Your energies are better spent on creating great stuff in AROS land than bursting our dreamy bubbles in Amiga land

Originally Posted by Kalamatee

Do you have any idea of the amount of work involved? And sorry to burst your bubble but some critical parts will still be written in ASM.



No that certainly doesn't define an Amiga - it defines a subset of AmigaOS that wouldn't even run together without a bunch of other components.

[wawa]

Originally Posted by bubbob42

You should at least appear to be open about any possible solution, wawa.

i always was.

I will accept AROS as my solution of choice once it looks, feels and reacts on my machines like OS3.9 does.

yet you dont want to put any effort towards it, just demands?

Until then, any alternative is welcome, be it a stripped down OS4 on 68k or just the original.

and who do you think is going to do that for you?

aros already works on amiga, yesterday evening i have fired it up with a4000/040/aga and run some stuff for testing.

os4 does not. so its other way around. its not that you can use os4 in the meantime before aros is perfected. you can use aros in the mentime before os4 is even ported to 68k, which likely what never happens.

Neither I nor many others are willing to spend lots of time (which we don't have) testing or writing bug reports for something which we already have in its original form to build our own projects upon.

thats fine. everybody has his own life. but how do you expect os4 for 68k ever to happen that way? who is gonna do that? olsen? thor? all alone? why do you think it will happen just like that?

besides aros is built for compatibility. it shouldnt affect your projects. you just carry on as previous. your prisma megamix driver probably even works on aros as is.. same as other genuine 68k drivers and binares usually do.

Please accept that.

No one is trying to put AROS in a corner. There are different tastes, though.

im fine with it. it is simply your choice. i think im simply offering a viable opportunity. instead of utopic demands. take it or leave it.

[michaelz]

Aros is open source already. This is not the place to be discussing this.

[grond]

Originally Posted by michaelz

Aros is open source already. This is not the place to be discussing this.

AROS being open source sure makes some closed source stuff less critical to be open sourced.

[michaelz]

Originally Posted by grond

AROS being open source sure makes some closed source stuff less critical to be open sourced.

Still a matter of opinion I guess. And it's also a matter of acceptance. I accept that people think AROS is they way to go (and even think there is quite some truth in that). I hope people also accept that if a thread is about open sourcing currently closed source software, they accept that as well even if they don't fully agree to that.

I think that even if AROS was 200% better than the original, Kickstart and Workbench should be released as open source for the history of that. It gives also some perspective about how it was coded, perhaps some insight in illogical choices made, hardware bugs that were worked around and most of all; it's a part of history. In that way, AROS is perhaps "our future" for some or all of us, but it is a new and different product and not the original.

So simply put, I don't ask you to forget about AROS. I ask for the respect that someone else can have a different opinion and view. You can try to alter it, but that is not the point of this thread. Open a new "AROS should be the future for all Amiga's" and I hope I can one day run it on my unmodded Amiga 500 as well. Because that is also why I would like the original code to be open sourced.

[wawa]

Originally Posted by michaelz

AROS should be the future for all Amiga's

thats not what i/we advocate. genuiune amiga is great and fine as long as you dont expect any coordinated progress you can contribute to. i fully support your initiative with open sourcing the genuine stuff. go ahead with it by all means, if you think it has any chance.

[Minuous]

@wawa:

I don't see why you are attacking bubbob42. It seems bizarre that apparently no one is allowed to state any of the many reasons why they don't use AROS without having to volunteer to develop it. It's like saying someone can't criticize Windows without offering to do unpaid work for Microsoft.

I for one have no interest in working on it as I lack the time and motivation, I am happy with OS3.9 and there does not even seem to be any roadmap towards AROS having OS3.9 compatibility, they are stuck in the past with OS3.1. No thanks, I would have no more interest in working on that than I would on a reimplementation of eg. OS1.1. The latest version of AmigaOS for 68K is OS3.9, not OS3.1.

[wawa]

Originally Posted by Minuous

@wawa:

I don't see why you are attacking bubbob42.

fyi. im not attacking bubbob. am i? and im perfectly fine with os3.x, in fact it is and has been my os of choice for many years. but i realisticly dont see an option open sourcing it. and i dont ses an option of open sourcing os4, if there was any advantage to it. what im challenging is the point that something like that happens or may happen for free, without engagement. no matter what solution to choose.

[bubbob42]

Originally Posted by wawa

fyi. im not attacking bubbob. am i?

hm, well...one could interpret a statement like

Originally Posted by wawa

yet you dont want to put any effort towards it, just demands?

accordingly.

But if that had not been your intention, so it it wasn't mine to make demands. I'm simply going to wait for it to happen. Same with backporting OS4 - olsen just dampened the perspective for the latter, though.

Originally Posted by wawa

your prisma megamix driver probably even works on aros as is.. same as other genuine 68k drivers and binares usually do.

Too much credit - driver has been written by scrat; I merely helped with debugging, keeping SAS/C compatibility and built a player based on it.

[kolla]

Originally Posted by idrougge

You're guessing wrong. OS 3.9 in ROM eats up 67 kB of chip RAM and 172 kB of fast RAM when booting without startup-sequence.

How much ram is used depends on how many drives you have attached and what filesystem and cache settings you are using. I see no difference between OS3.1 and 3.9 in this regard, which is no surprise as there are no fundamental difference between 3.1 and 3.9 kickstarts.

[hth313]

Being new to the board I found this interesting thread, but it just ends here. Did something come out of this?

[Marlon_]

Originally Posted by hth313

Being new to the board I found this interesting thread, but it just ends here. Did something come out of this?

No

[Gaula92]

@kalamatee: What's the state of AROS for the Raspberry Pi series of microcomputers? Last time I tried, USB keyboards and mice didn't work at all...
I believe porting to that kind of unexpensive hardware is the way to go for Amiga OS derivatives, other options are dead-ends.

[Samurai_Crow]

Kalamatee has left the team.

[hth313]

Maybe we should consider going to RISC OS instead, it runs on cheap hardware like Raspberry Pi and has an open source model.

Seems to work better there..?

[Samurai_Crow]

After all these years, AROS has only a handful of active developers and 68k and PPC are the focus. AMD64 is still just a skunk works project and x86 has less software than 68k working. ARM works hosted on Linux only so It's basically a small window manager.

[gururise]

Lets face reality. 68k Amiga is for nostalgia purposes only, not too much real work will get done on those computers (even Vampire accelerated), as the software ecosystem is just too old (ie. no modern web-browsing, no modern office suite, etc).

Another reality is x86 is dominated by Windows... and Linux has cornered the free market.

That leaves ARM and Raspberry Pi devices. Currently they run Raspbian (Linux); however, even stripped down, the OS is a bit heavy. This is where Amiga OS (ie. AROS) could potentially make serious inroads.

AROS running on the Pi, and there are millions of Pi's out there, gives the AmigaOS another chance to become relevant again. Apparently, there is a version of AROS that runs hosted on the Pi and runs pretty well; unfortunately, that version seems to be paid. But AROS is OSS licensed, so anyone who purchased the paid version would be able to upload it.

P.S. I fully support open-sourcing the current closed source software on the Amiga too! I would just like to see work put into making AROS viable in the Pi as well! =)