@jPV

I have at least a stderr utility in my distribution that seems to work, at least it created a logfile when I just tested it. I added it because that is possible on 68k to mix aros and amiga components. Which seems to work too, I tested it several times.

I also tested lha in my distribution and it perfectly worked. So indeed when it not works there seem to be a problem in the aros port of the component, in this example lha. On 68k I more than one time replaced aros components and used replacements I found from amiga side.

But that does not help of course on X86 ;)

the only exception was: *>>ram:output.txt

that seemed not to work

Regarding stderr on 3.1.4 I found a discussion that it is not there because not in 3.1 but introduced later

Last edited by OlafS25 on 26-Apr-2024 at 10:30 AM.
Last edited by OlafS25 on 26-Apr-2024 at 09:59 AM.
Last edited by OlafS25 on 26-Apr-2024 at 09:44 AM.

@Matt3k

in my view finally there were only two chances... either full speed in amiga chipset development from day one to stay ahead always to competition. Or drop chipset development and buy the components. Unfortunately they chose to do neither, instead only limited development. By that amiga lost its position on the top. And also it became too expensive in comparation to PC.

And yes running amigaos on fast hardware not solves the problems. You would need a new rewritten modernized OS but that would hardly differentiate from other platforms and would be a big investment.

@OlafS25
"in my view finally there were only two chances... either full speed in amiga chipset development from day one to stay ahead always to competition. Or drop chipset development and buy the components. Unfortunately they chose to do neither, instead only limited development. By that amiga lost its position on the top. And also it became too expensive in comparation to PC."


Great points and I fully agree.

I would only add one more item to your points...

We all would have much better served if AmigaOS4 never existed and they backed Phase 5 who already had a running OS solution instead of starting over years later making another OS to compete against it that created lawsuits and issues from the very beginning and divided the already shrinking userbase. It would have stabilized the aftermath of Commodore and given a lot more resources to benefit us all. AmigaOS4 would be forever playing catch up and driving talent away and creating instability.

Even to this day, the lawsuits from parties still exist and created a mess for the community...

Imagine if in the mid 90's if they backed the one solution and promoted it? Where would be be today? It would have driven us fully to your number 2 option by dropping the chipset and moving on. It isn't a big stretch that we would have a very capable OS long ago on another platform.

@Matt3k

the new OS4 and the lawsuit was long after Commodore, in the 2000s. I talked about the 80a and beginning 90s. Phase 5 had a new OS? As far as I remember they offered a combined card with both 68k and PPC. And a technology to run software on PPC or at least parts od it. A new PPC AmigaOS did not exist. First step would have been to make it portable. From what people say who saw the sourcecodes it not sounds like they had even started. The problem propably was Commodore run out of money and the easiest way to save money is to cut the development departmenbt. Of course on the long run not the best solution.

@Matt3k

The only "OS" I recall Phase5 producing was the PowerUP kernel. And with the introduction of PPC on the Amiga began what became known as the kernel wars. The only other "OS" competing with it was WarpOS which became somewhat of a de facto standard with support from most game developers including Hyperion.

Then a decade later Hyperion produced OS4. Which lacked support for both PowerUP and WarpOS. That didn't make sense. They ported the OS to PPC. Supported 68K directly. But didn't support any PPC Amiga binaries. How was that meant to work?

In addition they didn't port any of their games to OS4. And none of their games worked on OS4. Not without any 3rd party modifications. They should have worked out of the box. They may have been technically minded but the twin bros didn't know about Amiga marketing after that poor effort.

Last edited by Hypex on 26-Apr-2024 at 02:38 PM.

@agami

Quote:

I agree, porting it to high-performance 68k would not make the OS any better, but it would make it available to a user base orders of magnitude larger.

What would constitute as a high performance 68K? PiStorm? Fast x64 Emulation?

Well, it can be emulated now, but the performance is poor. An easy working qemu would work better without chipset emulation bottle necks. But for some reason x64 struggles to emulate ppc32 at a fast speed, despite it being 20 years in the future. And OS4 ppc is barely up to G5 stands from 20 years ago.

@Hypex

AROS running on 68k is quite slow, which is a surprise for me. I guess it has to do with drivers, but still iI would expect to be faster in RTG. Sadly it has its own gfx system and cannot piggyback on Picasso 96 UAE driver.

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

@OlafS25

Quote:

OlafS25 wrote: @jPV the only exception was: *>>ram:output.txt that seemed not to work Regarding stderr on 3.1.4 I found a discussion that it is not there because not in 3.1 but introduced later

But they've added it at some point, because it's available on 3.2 again...

Hmm.. with a quick test on a very basic and not up-to-date 3.2 emulation setup, *> works, but *>> makes a guru... hmm... IIRC I had an issue with both > and *>> combined in the same line on 3.2 earlier, but it ended up in an eternal loop in certain conditions, so there must be something broken even though implemented. Doing *>< was a work around...

_________________
- The wiki based MorphOS Library - Your starting point for MorphOS
- Software made by jPV^RNO

@OlafS25

Timing is everything :).

Your timeframe is before when I was referring to. Forgot about some of that development.

I think the biggest mistake commodore did in that area was to fire Tom Rattigan. The guy saved Amiga once got things back on track and when they were going into bankruptcy he turned it around and made commodore profitable.

Gould fired him and the rest is history...

That from my memory was the death blow to Amiga.

Commodore went out of business because no one was running the business, that killed the Amiga more so than hardware decisions. Sure, they needed to make a decision at some point but given the poor management and timing, they lost out. If it survived a few more years I figure they would have decided one of the directions you mentioned.

@pixie

What do you mean with slow here? I use it and should know

@Matt3k

I think the mistakes were already done in 80s. They had a fantastic technology with amiga but it was not in the middle of the company. As I understand it from interviews the management not used Amigas but PCs. I am pretty sure at Apple they used Macintosh for work. Amiga was only one brnach, they also produced PCs and burned money there. The only really profitable section was the C64, the cash cow of the company. But of course that could not go on forever.

Perhaps if they had further developed the amiga technology and technical stayed ahead far over the competition. But that would have become more and more difficult.

@jPV

the discussion was about 3.1.4

@Hypex

All valid points and yes it was all very primitive at the beginning. Certainly the OS native development needed to move that forward but given the uncertain nature and volatility in the Amiga universe likely hampered that.

It certainly didn't help that Gateway was forced to dump Amiga by MS, that was probably the last real shot it had for some future regardless of what and who did what afterwards...

Last edited by Matt3k on 26-Apr-2024 at 04:00 PM.

@Matt3k

PowerUP or how it was called was to offer software developers a path to migrate software stop by step to PowerPC. I think the real try to replace AmigaOS and Workbench with something new was p.OS from Prodad. I never installed and used it but at least it looked quiet nice. But the companies at that time could not agree on it.

@OlafS25

Rattigan was the reason for the 500 and 2000, and he understood the importance of the Amiga as well as making a buck. He did restructure and lay off but the results of his leadership was undeniable. IMHO he was capable of dominating the market if he was allowed to stay and not removed because of jealously basically.

He alone had the chops to beat Apple and taken it's place in history.

Last edited by Matt3k on 26-Apr-2024 at 04:01 PM.
Last edited by Matt3k on 26-Apr-2024 at 03:58 PM.

@OlafS25

I also use it, and with themes, when I set it to slow emulation speeds (030) one can see the redrawing of screen happening. Perhaps the gfx aren't accelerated.

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

@pixie

I turn off themes and use Magellan. Magellan uses its own system. Yes Wanderer is slow. Problem is it builds up everything pixel by pixel as I understand it. When running on X86 it is not visible. Still seeing redrawing sounds strange. Where do you use it? Emulation? With JIT or without?

@OlafS25

The issue doesn't show up when you use a fast system, but it doesn't mean the problem isn't there. I am not knowledge enought to say the reason behind it, if it's due to the graphic system not using hardware acceleration.

I use WinUAE, usually at full speed with JIT, but if downgraded, without kit, to real Amiga speeds it just crawls. Where Workbench doesn't.

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

@jPV

Hi jPV, I am always convinced that LHA AROS version doesn't work well, in fact on AROS One I don't use it anymore, instead I use on all App I use "XadUnfile", it works better and supports many formats !

On AROS as reported HERE the Redirection should work fine:

Example: dir >ram:a -> correctly generates the output file !

Also the command c:lha >ram:output.txt *>>ram:output.txt, works fine on AROS, but it generates an empty file !

_________________
AROS One Home Site
AfA One Amiga OS 3.9

@jPV

Quote:

Uhmm... do you mean "MorphDOS" isn't AmigaDOS v4.x compliant, with extensions that's been made after 3.1/3.9? Because in my experience MorphOS is very AmigaOS 3.1 compliant regarding DOS support, and that's been the goal of MorphOS developer always.

I mean with both 3.1 and 4.x. I test my scripts with OS3.1 minimum. I aim to have it working on 3.1 DOS and upward compatible.

Quote:

I have written many 3.1 compatible scripts and never had any issues on MorphOS with them. They've always worked on both MorphOS and OS4, but the only compatibility issues have been with AROS and Hyperion's OS 3.x (3.1.4 - 3.2).

My issue was with strings and collecting output into an env var. I tend to use RequestFile to do that. If it could save direct to an env I would have done that. Like so:

	RequestFile >ENV:Drawer DRAWERSONLY

	RequestFile >ENV:Drawer DRAWERSONLY
	Set Path "`Echo $Drawer`"