New improved intuition.library version from the Kickstart 3.1

[olsen]

Why pick an arbitrary obsolete version like V3.1 and not, for example, V1.1? Seems pretty random. I don't see what's so special about OS3.1.

Trick question: what's the big difference between programming for Kickstart/Workbench 1.1 and 3.1, other than the number in front of the dot? The APIs evolved big time between 1.1 and 3.1, putting much more power into the hands of the programmer. What required a lot of inside knowledge and fiddling with barely documented data structures back in 1985 became easier to achieve and was a lot less error prone, too. Documentation and example code arguably became better, too, although much of the original documentation was quite good already (even though it was not exactly complete).

Also, the operating system became more robust, thanks to tons of bugs getting found and fixed when QA tools such as "The Enforcer" were widely used within Commodore engineering (which may not be relevant for AROS, though).

[Kronos]

Well, of course. MUI has never been part of AmigaOS. AmigaOS already has a GUI system and it isn't MUI. So I'm not sure why you would put MUI into AROS instead of the official GUI system.

Well maybe you should take a step back to see what really has happened:

Hack&Patch were doing 3.5/9 on the cheap, scamming contributors left and right.
Doing those few actually new tools in 3.5/9 based on just plain GadTools+BOOPSI was no option.
Stuntz was to smart to give MUI away for free.

So they had to find a 2nd class GUI system that could be aquirred at next to 0 cost, and hell did ClassAct fill that bill.

The reality even today is that it is 100% possible (and not even hard) to setup a useable Amiga without ever touching ClassAct/ReAction. Try the same without useing MUI (yes even on OS4)...

[wawrzon]

But then, have you actually run aros on real Amiga hardware?

It sucks horribly (unless you are using a 68060 with graphics card and plenty of ram), and that is why most real Amiga users still prefer AmigaOS 3.1, 3.5 and even 3.9.

for everyday tasks, yes. but this is a vicious circle. too few users or rather testers create too low demand on improvements and deliver too few competent bug reports and the developers go for things that create more interest like ppc emulation on uae.

Aros seems it is the way forward for 68k, but then it also seems that it never reaches that desired state to even replace AmigaOS 3.1 efficiently.

maybe, maybe not. the chance is there, or at least a chance of an improved but legally safe hybrid system such as the one olaf is working on. people must get at least a little engaged to push it forward in order for it to become actually usable alternative on genuine hardware. none will come and serve you full dish for free without the customers cooperation as it was the case when you first bought your amiga from a commercial company back then.

[wawrzon]

Why pick an arbitrary obsolete version like V3.1 and not, for example, V1.1? Seems pretty random. I don't see what's so special about OS3.1.

aros aims at maximal compatibility with the 1.x-3.x range. 3.1 is a reference point, but measures to allow for execution of software that depends of exclusive features of kickstarts since 1.x are being implemented. tonic surely could better comment on this. im only a noob tester. why not ask him on eab?

Report+ and MCE, for example, require BB1 versions of some ReAction components as they use some of the newer functionality.

i seem to remember to try out report+ but i will try to check those if time and patience allows.

OS3.1 never included MUI anyway so I don't see the reasoning behind having it in AROS. If it's missing Installer then it's hardly a "complete system out of the box", even for an OS3.1 clone.

it was quite a reasonable practical decissions in my eyes. there is much more important and complex apps written for mui than for reaction. those could be easily ported having a mui replacement. if im not mistake currently i observe preparations for odyssey, but there is a number of mui dependant software i can just use on aros68k in its genuine amiga version. lets name yam and simple mail, ibrowse and digibooster almost work..

[wawrzon]

Yet, after all these years, aros doesnt manage to implement the functionality AmigaOS had back in 1993. Blame whoever you want, but AmigaOS 3.1 was not even top technology back then, and it is a shame it cant be accomplished in 2014.

So my point is that aros is full of really good objectives, but lacks in its poor execution as a project.

which of amigoid systems is top technology by todays standards? aros is probably least polished on the surface, but it has technical advantages in comparison with mos and os4 in some fields and most certainly in comparison with a bare 3.1 setup. you might simply install a 3.1 system from your floppies and then decompress an aros nightly to another disk on the same machine to try them side by side.you will notice the difference but you will probably notice too that aros will likely run most of the programs ou can run on 3.1 out of the box. in other to run others you will need to download libraries and classes from internet exactly as you need to do on 3.1.

[psxphill]

Please tell me how to do it. Lets say, I want to create generic trackdisk based copier that copies data from drive A to drive B. Source drive could use trackdisk.device and destination drive could use usbtrackdisk.device.

What is the most optimal buffer type application should use? And how to find it out? Ask user for correct memory type?

If you're asking the user for the usbtrackdisk.device name and unit then why not ask them for the type of ram?

Asking both devices is the only way to get the most optimal solution. By telling exec what things need to be able to access the ram then you can implement memory protection at the same time.

Or you could just copy the data using memcpy and pretend it's the best way.

[OlafS3]

Yet, after all these years, aros doesnt manage to implement the functionality AmigaOS had back in 1993. Blame whoever you want, but AmigaOS 3.1 was not even top technology back then, and it is a shame it cant be accomplished in 2014.

So my point is that aros is full of really good objectives, but lacks in its poor execution as a project.

You talk in general phrases... what is missing in your view and I will make a screenshot showing you it running on my distribution. Perhaps you never used it, haven´t you? I saw this already often in the community, the less experience the stronger are the opinions. As I said Aros 68k is really the last chance for the 68k community to get a working AmigaOS successor that is in development. If people are not interested in it and prefer fiddling around with their aging AmigaOS then it is ok for me and I do not need to invest time in anything for real hardware and can concentrate on emulation (what is enough for me personally).

[kolla]

Hm, AROS does have "installer", evindently.

http://kolla.no/aros-installer.jpg

[OlafS3]

Yet, after all these years, aros doesnt manage to implement the functionality AmigaOS had back in 1993. Blame whoever you want, but AmigaOS 3.1 was not even top technology back then, and it is a shame it cant be accomplished in 2014.

So my point is that aros is full of really good objectives, but lacks in its poor execution as a project.

A simple question to you... What functionality is missing in my distribution?

[OlafS3]

Hm, AROS does have "installer", evindently.

http://kolla.no/aros-installer.jpg

On 68k you can use a standard installer by copying the file in C. I use the official installer from aminet and it perfectly works. I only include it not in my distribution because of legal reasons.

BTW examples of GUI toolkits in Aros 68k

http://www.aros-platform.de/html/gui_toolkit.html

[Ball000]

Hm, AROS does have "installer", evindently.

http://kolla.no/aros-installer.jpg

If you're satisfied with it being able only to run the provided demo (and even then, it sometimes crashes...), then yes. Really, I would guess it's only about 50% finished and still needs a lot of work. But perhaps you were kidding?

Until some developer is interested in finishing it, I would advise people who need those functionalities to use a shell script of course: it is quite stable and evolved on ABIv1 and ABIv0-on-trunk, and so is able to make many tests on installed files and the system in general, and even allows quite comfortable interactions with the user.

[OlafS3]

If you're satisfied with it being able only to run the provided demo (and even then, it sometimes crashes...), then yes. Really, I would guess it's only about 50% finished and still needs a lot of work. But perhaps you were kidding?

Until some developer is interested in finishing it, I would advise people who need those functionalities to use a shell script of course: it is quite stable and evolved on ABIv1 and ABIv0-on-trunk, and so is able to make many tests on installed files and the system in general, and even allows quite comfortable interactions with the user.

What are you talking about? I really use it so i say you tell not the truth or your experience is based on old nightly builds. Nightly builds are NOT what people use or should use and are misleading. They are a backbone for testing new features for devs, not more not less. For users and application devs are distributions.

[wawrzon]

What are you talking about? I really use it so i say you tell not the truth or your experience is based on old nightly builds. Nightly builds are NOT what people use or should use and are misleading. They are a backbone for testing new features for devs, not more not less. For users and application devs are distributions.

you did not understand. he is talking about the aros installer utility, and is being correct. therefore we on amiga 68k use the commodore installer. would be cool to have aros installer finished, but since there is free available 68k utility i dont consider it a priority. the x86 is naturally missing on these functionalities, but then to my knowledge there is currently no x86 software that requires it.

[OlafS3]

If you're satisfied with it being able only to run the provided demo (and even then, it sometimes crashes...), then yes. Really, I would guess it's only about 50% finished and still needs a lot of work. But perhaps you were kidding?

Until some developer is interested in finishing it, I would advise people who need those functionalities to use a shell script of course: it is quite stable and evolved on ABIv1 and ABIv0-on-trunk, and so is able to make many tests on installed files and the system in general, and even allows quite comfortable interactions with the user.

Some raytracer I have tested on it:
http://www.aros-platform.de/html/raytracing.html

But of course not even demos are running. Perhaps I am just daydreaming and making screenshots of it

[OlafS3]

you did not understand. he is talking about the aros installer utility, and is being correct. therefore we on amiga 68k use the commodore installer. would be cool to have aros installer finished, but since there is free available 68k utility i dont consider it a priority. the x86 is naturally missing on these functionalities, but then to my knowledge there is currently no x86 software that requires it.

Yes I know what he was talking about. But we are talking about 68k here and he is saying it is 50% because of important parts missing and example here is installer that can be easily solved. I read that people on MorphOS and AmigaOS use the original 68k libs to add AREXX, then both are only 50% finished too? People here partly are talking nonsense, based on non-knowledge or halftrue based on very old experiences to have reasons to stay where they are and not needing to contribute. That may sound a little harsh but that is my view in the meantime.

[Ball000]

Sorry Olaf I may not have been clear. I was only answering to Kolla's post and his screen-shot of AROS' installer. I am totally convinced that your distribution is perfectly usable on emulators, and polished even. Congratz BTW.