Amiga OS 3.1 Source code leaked yesterday

[redblade]

So if it was to be ported to the ARM architecture.
Keep Exec
Redo DOS but start from Kick20 as standard?
Redo GFX
What about the rest?

[Minuous]

@absence:

Unfortunately AROS is much too far behind recent versions of AmigaOS, and that gap is only widening. Also its performance leaves much to be desired.

[Thomas Richter]

Originally Posted by redblade

So if it was to be ported to the ARM architecture.

Why?

Originally Posted by redblade

Keep Exec

Is in assembler, thus useless for arm.

Originally Posted by redblade

Redo DOS but start from Kick20 as standard?

Is about 30% in assembler, and has a questionable design with BPTRs over the place and quite some legacy cruft around BCPL and its structure.

Originally Posted by redblade

What about the rest?

Intuition is quite ok. The layers that came in with 3.1 are mostly ok.


But still... Why, what's the point of an outdated Os ported to a new platform with no new software becoming available? Os 4.x is not exactly a commercial success, and neither is Morphos. It is a niche of a niche - is there any reason why that should suddenly change with arm?

[masteripper]

While not an assembly programmer (higher level languages... ) from a little i read the x86 platform seems much more attractive...for start it has some heavy heavy documentation and the design given the fact is CISC means a lot less registers swapping.
As for the why migrate an ancient OS....well....i am not a hard core Amiga lover (i sold my Amiga almost 30 years ago...sold but not forgotten) so it was a bit of shock after discovering - by accident - a live breathing ecosystem that keeps producing NEW products for 30+ years machines (am i wrong ?...with all the accelerators i see on sale)
I don't see much love for Windows 3.1 or Apple System 7.6.1 as someone mentioned them as Amiga rivals and all i see from my perspective is beautiful crafted desktops,old but still capable applications eager to consume the extra power...utilities/applications that try to keep the OS up to date for daily use...even development of new applications/games that would probably "explode" if there was a solid background.
And given the outrageous benchmarks of Amithlon on ancient "trash level" systems ...maybe it would cause a stir.

[Bruce Abbott]

Originally Posted by absence

It's funny that you should mention RTG, because competing operating systems from the same era as Amiga OS 3.0, like Windows 3.1 and Apple's System 7.1, already supported this natively without hacks like P96. They had their own flaws for sure, but were really quite capable, both in native features and extensibility.

What I remember about the Mac was that if you didn't have the exact OS for your model it wouldn't work, because it was full of patches specific to that particular machine.

Windows 3.1 was so capable that high performance games stuck to using DOS due to poor Windows driver performance. To fix this Microsoft developed WinG, and later DirectX.

WinG was a technology being built by Chris Hecker in the research group, and at the time it was one of the small Microsoft Skunkworks projects, very low profile and off-the-wall. Basically it was fixing broken Windows drivers to make them run faster and more acceptably. Using it, we were actually able to create a video API that could run DOOM almost as fast under Windows as it did in DOS...

WinG was shipped on September 21, 1994

Then a few years later they threw it all away in favor of a completely different code base (Windows NT). Apparently Windows 3.x/9x wasn't that extensible.

Amiga OS didn't have RTG because it didn't need or want it. Tuning the OS to the native hardware permitted getting closer to the 'metal' for improved performance, which is how we were able to beat PCs and Macs in multimedia and OS friendly games.

The Mac infamously forbid this, and DOS PC apps almost universally bypassed the OS to get directly at the hardware for better performance (I had my first introduction to this when trying to get a serial mouse working on my IBM JX. If only they they had gone through the BIOS it would have worked perfectly. But no, they had to fiddle with the UART registers!).

[Thomas Richter]

Originally Posted by masteripper

While not an assembly programmer (higher level languages... ) from a little i read the x86 platform seems much more attractive...for start it has some heavy heavy documentation and the design given the fact is CISC means a lot less registers swapping.

That's really an academic discussion, you know. I don't think AmigaOs is going to another CPU model anytime soon, simply because it is not central to the theme. The reason why those people left still use the old machines is not because their technology is advanced, but because they grew up with them.



A new CPU platform cannot address that. Amiga is a piece of cultural heritage. That is also the reason why any attempt to migrate to another platform (PPC) was not a breakthrough, but a very minor niche.

Originally Posted by masteripper

I don't see much love for Windows 3.1 or Apple System 7.6.1 as someone mentioned them as Amiga rivals and all i see from my perspective is beautiful crafted desktops,old but still capable applications eager to consume the extra power...utilities/applications that try to keep the OS up to date for daily use...

I do not know that. I believe if you start looking, you'll still some fan base of DOS based games, old MacOs software, or probably win3.1 software. In a thinkpad forum I use to visit, people are still trying to use Os/2 nowadays, or at least, it has some fans. I recently found even a fan base of the old "Joyce" Z80 machines if anyone remembers them...



I believe it is a matter of observation, and where to check. Still, I would not call these "rival systems". It is not that an Amiga could replace an old Dos machine, or vice versa. These are just different niches, with almost distinct users basis.

[masteripper]

Originally Posted by Thomas Richter

That's really an academic discussion

I know its of academic nature ...and its almost 1000% certain that i won't involve in any way but nonetheless there are 2 questions
1. What if source were made available...how the computer industry would change...do remember that Linux was also a tied close niche OS that almost none knew about it when it was kind of release.
2 (This is from what i read) ...what if those $25,000 that were paid for the Intelectual rights of Amiga were somewhat paid from someone else ...or whatever.

[Thomas Richter]

Originally Posted by masteripper

1. What if source were made available...how the computer industry would change...do remember that Linux was also a tied close niche OS that almost none knew about it when it was kind of release.

Somewhere between "not a bit" and "not at all". I don't really see a "novel idea" the industry might be keen to know. The industry already "did change because of Amiga", in a sense. It is a system that made a non-zero number of people interested in information technology. It certainly influenced my career at least, and likely that of many others of my age.


So yes, it provided certainly some value at its time, but not in the form of superiour intellectual property.

Originally Posted by masteripper

2 (This is from what i read) ...what if those $25,000 that were paid for the Intelectual rights of Amiga were somewhat paid from someone else ...or whatever.

I have no idea, really.

[absence]

Originally Posted by Minuous

Unfortunately AROS is much too far behind recent versions of AmigaOS, and that gap is only widening. Also its performance leaves much to be desired.

Perhaps its lagging behind is proof that open sourcing Amiga OS wouldn't really change anything, because there aren't enough competent people interested in working on it? As for performance, surely the incredible difference in speed between a modern x86-64 chip and even a 68060 goes a long way to compensate for it?

[absence]

Originally Posted by Bruce Abbott

Windows 3.1 was so capable that high performance games stuck to using DOS due to poor Windows driver performance.

Hardware abstraction doesn't come for free, and it's only natural that early attempts, despite being extensible, were not as performant as direct hardware access on the relatively slow CPUs at the time.

Originally Posted by Bruce Abbott

Then a few years later they threw it all away in favor of a completely different code base (Windows NT). Apparently Windows 3.x/9x wasn't that extensible.

As you say, even the higher extensibility didn't prevent those early consumer systems from having to be redesigned from scratch in order to become robust professional operating systems. Where does that leave Amiga OS?

[gimbal]

Originally Posted by Bruce Abbott

Then a few years later they threw it all away in favor of a completely different code base (Windows NT). Apparently Windows 3.x/9x wasn't that extensible.

No, they simply got rid of the OS-line which relied on DOS.

[S0ulA55a551n]

Yeah from memory windows 3.1 was a GUI ( ish) that ran on top of DOS.

[coldacid]

@S0ulA55a551n: True so long as you're running in standard mode; in 386 Enhanced mode, it and DOS run in virtual-86 boxes on top of what we now know as a hypervisor.

[masteripper]

Sorry for reviving an old thread but if I haven't misread a critical question is not answered.
Is the code "complete" ?
Suppose that someday someone hands over a key to one of the "Gurus" here and says "go to that address and enter to room ABC" and finds a perfectly organized development environment with all the machines running what ever tool is needed (compilers and such) and you had this source code in a form that you could easily loaded...IF you start compiling would you end with a rom image of the Kickstart ready to flash it with a programmer and all the files of the Workbench ready to write them to the floppies. ?

[daxb]

As far as I know the answer is no.

[masteripper]

Originally Posted by daxb

As far as I know the answer is no.

Do you know what is missing?

[kamelito]

Originally Posted by masteripper

Do you know what is missing?

Olsen already gave details about that maybe in this thread I don’t recall exactly where.

[Matt_H]

Originally Posted by kamelito

Olsen already gave details about that maybe in this thread I don’t recall exactly where.

Short answer: a lot. Olsen is probably one of the three or four people on the planet who could turn the archive into buildable code.

[gulliver]

Originally Posted by Matt_H

Short answer: a lot. Olsen is probably one of the three or four people on the planet who could turn the archive into buildable code.

This is quite accurate.

[redblade]

I think I remember reading that you needed Green Hills C compiler to compile Intuition because it was designed with 16 bit ints not 32 bit.