@cdimauro

Money always matters

I read from one of the Morphos devs that it would take years to make a ISA change. If I assume that the resources are very limited (mainly only two devs and some voluntary) a port would not be faster. Who will pay for it and why? If you add that Trevor D. is spending a lot of money for new PPC hardware any change (or even discussion about it) is more than unlikely. AmigaOS on WinUAE might be a chance for more people to test and use it and propably some of those would be interested to buy hardware. In my view a chance for Hyperion. If they see it this way or as a threat has to be seen.

@andres

I think your issue is with terminills' statement then.

Jeremy's comment is only about a right to port to X86 being retained by Gateway Computers.

Anyway, you can write to Jeremy and ask why he was adamant enough to state this over and over in the 3 page thread.

#6

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

@number6

Even in this (in my thought completely irrealistic) vision, then a port to x64 (and no x86) would be possible?

Last edited by andres on 09-Aug-2014 at 06:18 PM.

_________________
A1200/020+68882 - 6 MB RAM - AmigaOS 3.0

Home Recording Audio

@andres

I believe that's what @cdimauro is commenting about after reading the post from Hyperion on that topic.

I'm sure he'll clarify if I'm wrong.

#6

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

@number6

Anyway I've read Jeremy's posts, and there no one document linked to prove what he's saying.
In my opinion, if I don't see an official document, he's completely wrong.

Last edited by andres on 09-Aug-2014 at 06:27 PM.
Last edited by andres on 09-Aug-2014 at 06:26 PM.
Last edited by andres on 09-Aug-2014 at 06:26 PM.

_________________
A1200/020+68882 - 6 MB RAM - AmigaOS 3.0

Home Recording Audio

@andres

I respect your opinion.

And yes, having no documentation deemed acceptable to the masses is a problem.
It's been a problem for a VERY long time.

#6

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

It has been my understanding that Amiga Inc. acquired an exclusive license to distribute AmigaOS for PPC only. They did not acquire the intellectual property for AmigaOS from Gateway. ACER still owns the intellectual property and what few remaining Amiga patents that are still in existance. Their deal is equivalent to the SCO/Novell deal for UnixWare.

But a port to x64 and ARM run deeper than that. Hyperion is a business.

You have to look at the cost benefit of porting and the practicality of it. To port the current codebase would take years of work, hundreds of developer hours and millions of dollars and its a bill someone will have to pay there could be no half-stepping. Look at the age of the AmigaOS code. There are just too many variables that could not and would not be efficient to do a port. BTW, he would buy it? ACER could take it over and produce something but they would have brand recognition they are pursuing an alternative OS strategy with Android which would turn off the remaining Amiga enthusiasts. I think Trevor could do something with Android to enhance his own offerings but thats a topic for another time and place. When Amiga Inc. got the license PPC was a big deal. Apple was on the POWER platform, IBM was developing the Workplace OS and were planning to develop AIX as a workstation OS so there was a method to the madness it just never produced. Unfortunately with Amiga pretty much just being a bright nostalgic spot in people memories, the market viability is not there. Besides they dont even own the kernel and with Hyperions unwillingness to work with AROS and MorphOS their channels are limited at best.

_________________
http://www.pc-opensystems.com

@RobertJDohnert

Quote:

It has been my understanding that Amiga Inc. acquired an exclusive license to distribute AmigaOS for PPC only.

??

Back in 1999/2000 (when Amino acquired these rights), AmigaOS was 68k only. 3.9 (still 68k) was released nearly year after Amino formed new Amiga.Inc. Their new OS (Amiga DE) should have been available for many CPU architectures, this plan never materialised. OS4 project started even another year later.

@pavlor

In 1997 when Gateway acquired from Escom, there were already plans to deploy PowerPC and making the jump from 68k to PowerPC was not as big an investment as jumping from RISC to CISC. The plans were set in motion.

_________________
http://www.pc-opensystems.com

@RobertJDohnert

Can you provide any link to official document about your statement?
If not, I don't understand from where "PowerPC" and "x86" terms come from in the Gateway/AMINO transition.

Anyway, Amiga Inc. publicly stated what they "ACQUIRE" from Gateway:
https://web.archive.org/web/20000302111618/http://www.amiga.de/diary/executive/update03012000.html

Last edited by andres on 09-Aug-2014 at 08:13 PM.
Last edited by andres on 09-Aug-2014 at 08:12 PM.

_________________
A1200/020+68882 - 6 MB RAM - AmigaOS 3.0

Home Recording Audio

@andres

Without picking that document apart, let's look at:

Quote:

Gateway purchased Amiga because of Patents, we purchased Amiga because of the People.

Now let's look at what Gateway REALLY said:

Quote:

It is clear that without you people, we would have had nothing to acquire. So the first obligation of Gateway 2000 is to say thank you. It is the Amiga community that has kept this brand and this OS and this product line and this concept alive

Source

It's everone's right to choose whom they wish to believe, or you can choose to believe none of them at all.

#6

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

@RobertJDohnert

Quote:

In 1997 when Gateway acquired from Escom, there were already plans to deploy PowerPC

That was PowerAmiga project of Escom/Amiga Technologies. Gateway had its own plans (eg. various "AmigaNG" projects). If I remember correctly, final target CPU for this new Amiga was never chosen (there were even rumours about Transmeta or Alpha CPU architectures...).

PowerPC was back then supported by significant part of community and remaining hardware and software companies (Phase5, H a P).

As I wrote in my above post, even initials plans of Amino led Amiga.Inc didn´t centered on PowerPC only. Sure, only PowerPC AmigaOne project ever become reality.

Quote:

OlafS25 wrote: @cdimauro Money always matters

Quote:

I read from one of the Morphos devs that it would take years to make a ISA change.

It depends on how the source code is. If it's almost C, with little assembly parts, like any o.s. should be, I don't think that it'll take years to port it to another architecture.
Quote:

If I assume that the resources are very limited (mainly only two devs and some voluntary) a port would not be faster. Who will pay for it and why? If you add that Trevor D. is spending a lot of money for new PPC hardware any change (or even discussion about it) is more than unlikely.

Talking about AmigaOS4, I read this:

"The AmigaOS 4.x kernel was rewritten from scratch in C but does contain some PPC assembly for some very low level parts."

And it means, for me, that a port shouldn't take years and so much resources. Except for the parts for which there's a PowerPC-only license, that should be re-implemented.

Of course, it also means to say bye bye to Petunia (and to AREXX also), but since many OS4 fans say that 68K isn't important, I think that they can survive to the loss, and focus their attention to RunUAE.
Quote:

AmigaOS on WinUAE might be a chance for more people to test and use it and propably some of those would be interested to buy hardware. In my view a chance for Hyperion. If they see it this way or as a threat has to be seen.

First, Hyperion is a software house and a profit company. As such, it should care about selling OS4 licenses, and not on what kind of hardware OS4 runs. Money always matters, right?

Second, WinUAE can be a chance to test OS4, because it doesn't require a (costly) PowerPC machine. But here the problem is that OS4 is not available as a demo, like MorphOS, so you have to buy a license BEFORE you can even test it.
Quote:

andres wrote: @number6 Even in this (in my thought completely irrealistic) vision, then a port to x64 (and no x86) would be possible?

Yes, it is, in the above conditions.
Quote:

RobertJDohnert wrote: It has been my understanding that Amiga Inc. acquired an exclusive license to distribute AmigaOS for PPC only. They did not acquire the intellectual property for AmigaOS from Gateway.

From the only, legally valid documents available to the masses, there's no such evidence. On the contrary, it seams that Hyperion acquired the full rights to do whatever they want with OS4, for any hardware platform.
Quote:

ACER still owns the intellectual property and what few remaining Amiga patents that are still in existance.

I don't think so. Patents should be well expired. All. Exactly this year if there was some issued on 1994 by Commodore.
Quote:

Their deal is equivalent to the SCO/Novell deal for UnixWare.

Again, there's no evidence for this.
Quote:

But a port to x64 and ARM run deeper than that. Hyperion is a business. You have to look at the cost benefit of porting and the practicality of it. To port the current codebase would take years of work, hundreds of developer hours and millions of dollars and its a bill someone will have to pay there could be no half-stepping. Look at the age of the AmigaOS code.

Looking at the AmigaOS4 code:

"The AmigaOS 4.x kernel was rewritten from scratch in C but does contain some PPC assembly for some very low level parts."

It means, to me, that a port doesn't take years, since it seems to be well written (from a port point of view). IF the above statement represents the reality, of course.
Quote:

There are just too many variables that could not and would not be efficient to do a port. BTW, he would buy it? ACER could take it over and produce something but they would have brand recognition they are pursuing an alternative OS strategy with Android which would turn off the remaining Amiga enthusiasts.

Acer has and cannot have any role on an OS4 port.
Quote:

I think Trevor could do something with Android to enhance his own offerings but thats a topic for another time and place. When Amiga Inc. got the license PPC was a big deal. Apple was on the POWER platform, IBM was developing the Workplace OS and were planning to develop AIX as a workstation OS so there was a method to the madness it just never produced. Unfortunately with Amiga pretty much just being a bright nostalgic spot in people memories, the market viability is not there. Besides they dont even own the kernel and with Hyperions unwillingness to work with AROS and MorphOS their channels are limited at best.

It doesn't matter. The problem with OS4, but for the post-Amiga products in general, is that the survived Amiga community is too small. Too little numbers. Too little money. It's not possible to expand the market, especially being linked to a dying architecture like PowerPC.

And this is the only point now: they are just surviving, remembering the "good old days"...

@andres
Quote:

Anyway I've read Jeremy's posts

I once pointed out some inaccuracies in one of his Ars Technica articles. The reply I got was: "accuracy is relative". I can't find the Reiner quote everybody's referring to, but I wouldn't put too much emphasis on what some guy on the internet once said.

As for "PPC only license": both AInc (Amithlon, Amiga Games Inc.) and Hyperion (layers.lib v54 on Aminet) have issued licenses for 68k code.

The Ben Hermans quote about "some licences being tied to PPC only" most certainly refers to the licenses he has from the OS4 developers for their *new* code. He couldn't buy the code (too expensive), he licensed it. And we have no clue what these licenses say and what limitations they enforce.

If I had to guess who only gave Ben a license for PPC, the Friedens come to mind immediately.

@cgutjahr
Agree.

_________________
A1200/020+68882 - 6 MB RAM - AmigaOS 3.0

Home Recording Audio

@cdimauro

Quote:

cdimauro wrote: It depends on how the source code is. If it's almost C, with little assembly parts, like any o.s. should be, I don't think that it'll take years to port it to another architecture.

Ah the old "just run it through a different compiler" chestnut !! Didn't work as good as a certain "lawyer" expected back when it was going from 68k to PPC....

Well actually it might kinda work, only you end up with something like AROS-x68.

Only with even less SW, 0 drivers and not even the possibilty to run hosted.

As to ClownCar.inc only buying a PPC-AOS-licence from GateWay, those 2 had zero interest in AOS nor porting it to PPC when the bought/licenced/leased/?? the Amiga name.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

Quote:

Kronos wrote: @cdimauro Quote: cdimauro wrote: It depends on how the source code is. If it's almost C, with little assembly parts, like any o.s. should be, I don't think that it'll take years to port it to another architecture. Ah the old "just run it through a different compiler" chestnut !! Didn't work as good as a certain "lawyer" expected back when it was going from 68k to PPC.... Well actually it might kinda work,

Actually the situation is a bit different, right?
Quote:

only you end up with something like AROS-x68.

x86 (or, preferably, x64 which has better performance).

Anyway, yes.
Quote:

Only with even less SW

Sure, applications needs to be ported. Well written code I think has no problem to be recompiled, except if there are parts in assembly.
Quote:

, 0 drivers

Please, enlighten me: is it so much different programming a PCI/AGP/PCI-express card simply changing the architecture?

AROS coders deserve a great ovation for their invaluable work, if it's that hard to do, since AROS supports different architectures...
Quote:

and not even the possibilty to run hosted.

How does AmigaOS4 run? Hosted? Nope. So, why does it matter having OS4 running hosted somewhere? It doesn't.

BTW, I tried AROS Windows hosted a couple of years ago, just to see if it works and if it can be a good solution for AROS development. But they were the only exception: all other times I ran AROS natively (after installing it on an USB-pen through VirtualBox).
Quote:

As to ClownCar.inc only buying a PPC-AOS-licence from GateWay, those 2 had zero interest in AOS nor porting it to PPC when the bought/licenced/leased/?? the Amiga name.

For the Amiga o.s. the licenses are for AmigaOS, Amiga OS, AmigaOne and AmigaOne.

The Amiga name is used for different things. And it's also logical: Amiga, itself, was not only the o.s, but the whole platform.

Aside this, an AmigaOS4 port has some value for post-Amigans.

@thread

Not directly related to the current license discussion, but still one of the most interesting, if not convoluted stories of Gateway (who kept the technology) under Amiga Development LLC:

Amiga Development LLC v Hewlett-Packard Company

Somewhat interesting that this was going on while the AmigaOS revival here was in full swing.

#6

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

So why doesn't someone just buy Amiga Inc.? It cant be worth a whole lot more than the licencing fees, at present.

@Marky_D_Sahdd

Quote:

So why doesn't someone just buy Amiga Inc.? It cant be worth a whole lot more than the licencing fees, at present.

Only those who have tried can answer that. Perhaps it's a question of how they value it vs reality. And what if they wished to address their debt in such a sale?

#6

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