@matthey


Quote:

You have complained about lack of documentation for AmigaOS drivers in another thread and how the existing ones were behind a NDA wall.

If AmigaOS3.x sources are released I gain exactly nothing.
If AmigaOS4.1 sources are released I gain nothing either. Because I will not be part of the resulting mess. Why would I? I'm lazy . All those dormant developers are suddendly waking up and doing the hard work for me anyways. I mean, that's the expectation of going open source, right? (and ignore the resulting mess in the process)

But to be completely honest. I was not aware of this possible outcome and the legal mess in general when I bought my Amiga X5k 5 months ago. And I've already dailed down my coding effort as a result. The horror scenario of going open source would mean the end of it anyways. One of the reasons that I wrote my own mixer program instead of extending the available one on os4depot and openamiga.org is that I took a quick peak at the source code. Let's keep it to a lack of my coding skills that I decided to create my own version. Not the mention the alien look of the GUI.

@geen_naam

you keep repeating "open source is horror scenario"

to me "horror scenario" is a platform is closed source and owner vanishes. Then what do you do?

@kolla

Quote:

I don’t care if open sourcing means even more “forking”, we already got plenty of that going on. I just want to be able to do everything myself, for myself, without insane licensing conditions, without having to sign any NDA or adhere to whacky ideas by certain “official” developers. It won’t “hurt” anyone, since I have absolutely zero plans of distributing my own builds, since that would also result in support requests which I have neither time nor interest in. Anyone interested my changes, could just pick commits from my source repos if they wish. This is what open source is about - freedom. Not “business” or “product development” - F dat sh!t.

Thanks for proving my point in advance of possible events. Everyone going there own way. And those who are trying to create programs are left the task of clean up the mess in aftersales why their program doesn't because the customer happens to have some kolla derived work installed on their system.

@geen_naam

at least someone can go somewhere

Assuming the company behind a closed source platform vanishes, what happens then?

You do not answer that

@OlafS25

Up to now, there's always been a new owner. And you might not agree with the direction of the owner. But at least there's a direction.
In the open source scenario. There will be multiple directions. Because all those captains currently not owning the sources all have different ideas and refuse to cooperate. As you already pointed out in your previous post. And expect those captains to create closed source work that only work with their distribution too. You want to have USB3? Works only with distro A. You want a modern browser? works only with distro B. And so on.

I honestly see no indication at all why this would suddenly change when everything is open sourced. Hence the expected horror scenario.
I'm not waiting for yet another morphos vs Amigaos4. Or another Powerup vs Warpos battle. Or another 68k vs ppc vs x86 vs arm vs RISC-V battle. Talking about wasting resources.....

@geen_naam

Quote:

AmiKit is a 68k Amiga distribution.

AmiKit is one of two independent distributions supported by Cloanto's Amiga Forever. The other being AmigaSYS.

Quote:

Noted and corrected. (Maybe people should come up with originals names instad. All the ami- amiga-, -kit combinations are very confusing . )

That probably true. AmigaKit was launched in 2004 and AmiKit was released a year later.


@matthey

Quote:

They would be better off pooling their resources and working together.

Pooling means profit sharing of the end product? I am certain this wouldn't be offered or permitted to happen. A-EON has never be offered or received any additional funding for advancing the OS4 graphics system from the stakeholders. The funds raised by Enhancer Software licences to end-users are used to partially cover the large development costs. The cost of investment over the last six years in the graphics system couldn't be "pooled": i.e. donated or sold off at a token value.

There will be a crossroads soon: the next levels of graphics development needed will require the graphics library to be overhauled. This is very costly and needs expertise. Are these resources available to Hyperion? Any changes to the graphics lib would result in a derivative work. If any third party did the work they would loose their copyright/control on the work unless there was some agreement. The past has demonstrated clearly that agreements are simply ignored or contested by some parties.

So A-EON and AmigaKit Ltd will continue to collaborate in an independent software project, free of the legal entanglements that others have been disputing for years. This is the most pragmatic option to advance our platform given the circumstances.

Last edited by amigakit on 03-Jun-2021 at 10:28 AM.

_________________
Amiga Kit Amiga Store
Links: www.amigakit.com | New Products | X5000

@geen_naam

Ben H. has a direction?

Share to inform me about that... I do not see it. And a business model making money of software that is developed for you for free sorry is no business model or direction (even if those devs do it voluntary)

@amigakit

without legal access to source codes your concept will come to a end

Or you write everything from scratch. Replacing some cli commands is easy or add something new but replacing important components with certainly lots of side effects and without access to internals will propably break a lot of. I imagine that difficult.

@OlafS25

Quote:

without legal access to source codes your concept will come to a end

It is best not to have any access to the source codes.

Taking just one example of the graphics library: it is so outdated that a brand new approach is needed. Many years ago parts of Picasso96 API were rolled in the graphics library by Hyperion's devs. Picasso96 is a 1990's relec in itself so it was bolting on old technology upon even older 80's software technology. How is the old graphics lib ever going to support many modern concepts such as Multi Monitor, hot plug detection, monitor DPI information, LCD backlight adjustments, Freesync etc

This is the same scenario when development of MiniGL (with its fixed pipeline design) came to a abrupt dead end and the shaders based Warp3D Nova took over. The rest of the world has moved forward.

_________________
Amiga Kit Amiga Store
Links: www.amigakit.com | New Products | X5000

@OlafS25
The direction was to take Amigaos4 to PPC. And thanks to this, additional PPC hardware emerged next to the powerup cards. Such powerfull hardware compared to 68k also enabled more advanced things like Radeon PCIe grahics card support. Or a webkit based browser.

So at least to me it's clear that there's a direction away from the 68k hardware and 3.1 software.

You might not agree with the direction in which AmigaInc and Hyperion has taken us. Some would have liked arm. Some would have liked x86. Some are still dreaming of AMD zen3+ performance for a 68k.
But it's a fact that there was a direction.

Is is finished? No.
Will it ever be finished? Likely not.
Does anyone care? Only the ones that do not agree with the direction or the people in charge.

@amigakit

but graphics library affects a lot of other components including desktop. Without sources it is in my view impossible to not have unwanted side effects.

And you have lots of software using it.

To do a clean-room rewrite and modernization of graphics library not breaking lots of software and components sounds to me like a huge and complicated project

@geen_naam

that direction was taken almost 20 years ago

we now have 2021 and a different world. What about now?

Yes I do not agree to you but you do not need to agree to me either

@OlafS25

Yes, the amiga world moves (too) slow. That's a fact. But my Radeon RX560 was released in May 2017. I can buy new HD Audio audio cards that are actually on sale in the store next door. So at least on some fronts we're catching up a little.
Modernising is a huge effort and there are simply not enough developers to do it at a pace that everyone likes. But it's all we got. No developer for every line of code like google and microsoft.

And that's not specific to the NG machines. The 68k crowd are cheering over 68060+ speed with their FPGA accelerators on their 35 years old hardware. Sure it's an achivement on its own. And when people are happy with it then that's ok. But alhough they have a larger user base, the actual progress is even less. Not to mention all the captains who have their own view of what an accelerator should look like. The real 68k, FPGA based, emulator on ARM etc).

And talking about direction from where we stand right now? I'd opt to throw away all the classic stuff and create a new landmark. Stop looking back at those classic limitations.
So a 64bit SMT kernel with no respect to the 2GB boundary and other old stuff like permit/forbid. Run the old stuff in sperate boxes for all I care. And ignore all those who will be complaining that "it's not the amiga way!!1!". But even this will be a huge effort.

Last edited by geen_naam on 03-Jun-2021 at 11:33 AM.

@geen_naam

the "68k crowd" has different expectations. People see it as retro, as hobby. They do not expect it to compete with modern platforms. And do not use it for daily work.

On NG that expectations are higher. Skilled devs are rare everywhere. And for commercial development of OS and new hardware you need lots of money, money that cannot be reearned in that market. Only in the classic market you today have growth and money. That is the basic reason why there is a court case between Cloanto and Hyperion. The 4.X line obviously has reached the end. Development is only done by volunteers doing it for free. Amigakit and Trevor D. are investing some money but it is difficult for me to see a real change there.

Regarding SMP, MP and 64bit it can be done (it was done with a special aros branch) but it will break old software. The question always is... who will write new software for it?

Last edited by OlafS25 on 03-Jun-2021 at 11:39 AM.

@geen_naam

Quote:

I remember that Amithlon was gaining a lot of momentum. Not sure why they had to stop.

Bernie is an AW member and has retold the story here several times.

the oldest version of the story on AW

@OlafS25

I agree re:your comment on the back porting. The best thing to do would be to simply ask Camilla.

#6

Last edited by number6 on 03-Jun-2021 at 12:02 PM.

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@OlafS25

"The question always is... who will write new software for it?"

Same ~1,5 people that do it today for alien-to-68k platforms.

With SMP, MP and 64bit, people could continue using their existing SW via UAE or some other sandbox solution.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@OlafS25

Quote:

es I do not agree to you but you do not need to agree to me either

That's for sure But I do not expect to be able to convince anyone anyways.

I returned last December after being away for 15 year. What is see is that AmigaOS4.1 is currently dead in the water with Hyperion and that Amigakit&Aeon is trying to improve the amigaos4 situation regardless with their Enhancer software and PPC hardware. So I really do not understand this anger or mockery towards them that some in here display.

I also do not understand that people honestly believe that the situation will improve once everything is open source. "WE" can do better than those in charge now is what they say. And all I keep wondering what "WE" have been doing the past 15 years. "WE" could haven been creating a modern browser. "WE" could have been created a new soundsystem to replace the aging AHI. "WE"could have been porting OpenGL/Vulcan, WxWidgets, Pyhton3 etc etc. But instead "WE" are only critizising that "THEY" aren't doing the right thing.

I returned 5 months ago. Saw an issue with the availability of sound cards. The easy way was to demand a driver for modern cards or complaining that everything is moving too slow. Instead, I created a new driver for myself and shared the result for free with every OS4 user. So maybe the "WEs" know it all better. But the "THEYs" are actually getting things done.

Last edited by geen_naam on 03-Jun-2021 at 12:51 PM.
Last edited by geen_naam on 03-Jun-2021 at 12:36 PM.
Last edited by geen_naam on 03-Jun-2021 at 12:34 PM.

@OlafS25

Like KimmoK already pointed out. Sandbox the old stuff where possible. And recreate parts of the OS where not.

But like I said. Even this is not a walk in the park. But it's always better than to spend countless hours to try and circumvent those limitations.

@OlafS25

Quote:

he "68k crowd" has different expectations. People see it as retro, as hobby. They do not expect it to compete with modern platforms. And do not use it for daily work.

I certainly don't have this expectation either. So I think it's more of a persisting myth than reality.

I also think that you underestimate the amount of money that Amigakit&Aeon have invested in the NG market.

Quote:

Development is only done by volunteers doing it for free.

If you mean Hyperion. Maybe yes. But for others developing for OS4 this is absolutely not true.

Last edited by geen_naam on 03-Jun-2021 at 01:51 PM.

@geen_naam

Quote:

have no indication that the captains suddenly come to their senses and start working together.

The idea is to get rid of all the captains, obviously. A lot of the stupid games that have been going on behind the scenes simply would not be possible with an open code base.

Engineers and/or big software projects do need to be managed, of course. But with an open code base, the engineers get much more of a say in what that management looks like. I can't hold them hostage, just because I own the code - I actually need to convince them to work for me: I need to have a convincing vision or amazing people skills, I need to be the most experienced developer or put the most hours in myself, or I need to give them the opportunity to work in an amazing team etc.

Right now, people who openly despise Ben Hermans when you talk to them in private or people who are publicly admitting he pirated Kickstart 1.3 are writing code for him. Because they love working on the AmigaOS code base and he's their only option.

Quote:

But since when is it unusual to pay for quality software in the first place? I remember having to pay for Cybergraphics, P96, MUI, DOpus and so on.

For starters, OS4 used to include (commercial) graphics drivers. Having to pay for drivers now is a sign that things are getting worse.

And while I have no problem paying for quality software, I don't appreciate forcing bundles on customers that try to push products down their throat they have no need for. I haven't checked Enhancer myself, obviously - but the fact that the os4welt guys want to write their own installer strongly implies you do not have the option to select what parts of Enhancer you want to install.

Forcibly replacing system components with 3rd party products is extremely bad design. Whoever decided to implement it that way - or worse, didn't realise that this is a bad idea - is basically holding up a massive sign saying: "please keep me away from any OS development as far as humanly possible".

Quote:

I have no indication that Enhancer coders are being abused. As a matter of facts, some of them are paid quite handsomely considering amiga standards. Part of it is possible due to Enhancer sales revenue.

Yeah, "abuse" was probably the wrong word here. I'm no native speaker, obviously - maybe "misused" would have been a better fit?

Here's what I was trying to say:

In a sane project, once the devs have identified graphics.library as a component that badly needs an overhaul, the project leader approaches one or two guys (Hans...) and asks them to do some research on that topic and come up with a plan - including a vague roadmap and a list of potential (compatibility) issues. Then you try to find a way (money, buttkissing, hardware donations, peer pressure...) to make Hans work on that.

In closed source AmigaOS land, this is what happened instead: Party A owns graphics.library. Party B licenses it and combines it with P96 code acquired from party C. Party D starts porting Gallium to AmigaOS, which is meant to become the base of a next generation graphics.library, but stops after half the work is done because they're no longer getting paid. Then party E comes around and pays Party F and G to write a custom 3D layer which operates completely independent of the existing graphics susbsystem and run a cut down version of OpenGL on top of that. So, fifteen years later, graphics.library is still a completely outdated mess - but we can build and run newer 3D Linux games now. If we pay for that option.

I was simply trying to say that the latter option is not making the best use of the Hans resource. Apologies to Matthew for accusing him of abusing people - he's evil, but not that evil.

Last edited by cgutjahr on 03-Jun-2021 at 03:44 PM.