Interest at bounties for Aros 68k?

[OlafSch]

Aros 68k is quiet usable already when running on UAE. But what about real hardware and FPGA? What are the interests?

Aros offers PCI support and Poseidon (USB) out of the box. What do people want to see so they would use it and where are people willing to donate?

OWB port comes to my mind

[gulliver]

I just wish someone could build an Aros 68k distro that mimics AmigaOS 3.1 functionality with about the same performance.

The problem seems to be that developers target Aros 68k, like if a 68060 at 500mhz actually existed!

I have no problems in added features, but there is really no point in doing it, when the most basic stuff is simply bloated for a 68k, or incompatible with many software. So getting the basics first would definately help, and then of course it can grow from there, but not the other way round.

If someone manages to get similar feature and performance as AmigaOS 3.1, then most 68k users will probably start switching to Aros, as it is free and open source, which means, it is future proof.

If Aros continues the same way it has on 68k, it has no future other than being a tinkerer´s toy on UAE or some FPGA board experiment. But no serious use beyond that.

[OlafSch]

Originally Posted by gulliver

I just wish someone could build an Aros 68k distro that mimics AmigaOS 3.1 functionality with about the same performance.

The problem seems to be that developers target Aros 68k, like if a 68060 at 500mhz actually existed!

I have no problems in added features, but there is really no point in doing it, when the most basic stuff is simply bloated for a 68k, or incompatible with many software. So getting the basics first would definately help, and then of course it can grow from there, but not the other way round.

If someone manages to get similar feature and performance as AmigaOS 3.1, then most 68k users will probably start switching to Aros, as it is free and open source, which means, it is future proof.

If Aros continues the same way it has on 68k, it has no future other than being a tinkerer´s toy on UAE or some FPGA board experiment. But no serious use beyond that.

I am thinking about doing a version that mimics 3.1. (using Magellan still). Of course the current big distribution is bloated because I wanted to integrate as many features as possible. But even a basic system is needing more resources than the original AmigaOS needed.

[deadwood]

Originally Posted by gulliver

I just wish someone could build an Aros 68k distro that mimics AmigaOS 3.1 functionality with about the same performance.

The problem seems to be that developers target Aros 68k, like if a 68060 at 500mhz actually existed!

Hi,

Can you tell me, whether you would be willing to donate to bounties related to improving performance of AROS m68k?

[saimon69]

Originally Posted by gulliver

I just wish someone could build an Aros 68k distro that mimics AmigaOS 3.1 functionality with about the same performance.

The problem seems to be that developers target Aros 68k, like if a 68060 at 500mhz actually existed!

I have no problems in added features, but there is really no point in doing it, when the most basic stuff is simply bloated for a 68k, or incompatible with many software. So getting the basics first would definately help, and then of course it can grow from there, but not the other way round.

If someone manages to get similar feature and performance as AmigaOS 3.1, then most 68k users will probably start switching to Aros, as it is free and open source, which means, it is future proof.

If Aros continues the same way it has on 68k, it has no future other than being a tinkerer´s toy on UAE or some FPGA board experiment. But no serious use beyond that.

Part of the problem is that actually an optimized driver for the original chipsets is missing so wanderer and Worklbook are using sorta chunky-to-planar display; plus don't take distros as reference: usually the nightly are kinda lean as afr as i remember

[gulliver]

Originally Posted by deadwood

Hi,

Can you tell me, whether you would be willing to donate to bounties related to improving performance of AROS m68k?

I have already donated in the past to the Aros m68k port bounty. But will only donate in the future if I see AROS m68k going somewhere, and if that is the case, I will problably be more generous now than before

[gulliver]

Originally Posted by saimon69

Part of the problem is that actually an optimized driver for the original chipsets is missing so wanderer and Worklbook are using sorta chunky-to-planar display; plus don't take distros as reference: usually the nightly are kinda lean as afr as i remember

I see your point, but I kind of expected more for all the time that has passed since the project started.

PS: I tinker with nightlies anyway

[saimon69]

manpower is lacking, don't ask me why

[AMIGASYSTEM]

I think it's better to focus first on AROS X86 that is the driving force, with some other programmer you could grow much, already distros AROS, IcarOS in particular, with the new version of OWB, it allows you to surf the internet no more limitations. Very little would be enough to increasecompatibility Cards Video, Audio, Wifi e LAN

[deadwood]

Originally Posted by gulliver

I have already donated in the past to the Aros m68k port bounty. But will only donate in the future if I see AROS m68k going somewhere, and if that is the case, I will problably be more generous now than before

Putting the performance topic aside, what other concrete changes you would be interested in (question open to other people as well)

[kolla]

Support for overscan on native chipset

[clebin]

Originally Posted by deadwood

Putting the performance topic aside, what other concrete changes you would be interested in (question open to other people as well)

Do you know of any plans to improve Workbook or has Jason McMullan left the scene now? I thought it had a lot of potential for lower-end machines but you can't do much with it currently, beyond launching a Shell.

[clebin]

Originally Posted by deadwood

Can you tell me, whether you would be willing to donate to bounties related to improving performance of AROS m68k?

Oh, and yes to this, definitely.

[gulliver]

Originally Posted by deadwood

Putting the performance topic aside, what other concrete changes you would be interested in (question open to other people as well)

Then it would be getting the full feature set AmigaOS 3.1 has.

[OlafSch]

Originally Posted by gulliver

Then it would be getting the full feature set AmigaOS 3.1 has.

what is missing? Aside from desktop

[OlafSch]

Originally Posted by clebin

Do you know of any plans to improve Workbook or has Jason McMullan left the scene now? I thought it had a lot of potential for lower-end machines but you can't do much with it currently, beyond launching a Shell.

he left for some time but is active. If asked he is still doing something when there are certain requirements that can be defined.

[wawa]

@deadwood
my current demands (:

- make prefs work (i reported endiannes problems, toni posted hints to the ml. maybe prefs implementation should be cleaned up a bit internally, as toni suggested, also i find it a bit messy where to look for certain settings.. like wanderer and then appearance, but it migght be just me.
-improve the speed loading hunk binaries in comparison with elf. perferably to the same level.
- make scalos from aros contribs work properly on 68k, im trying to debug it but since there is no proper guide how to debug aros on say winuae, im having hard time in it, given a bit limited time currently.
- make skinning work properly, currently there are some flaws, i consider skinning menus impossible, only default works. i have fast iff theme at hand, that would work well for interlace screens i guess.
- we could furtherly investigate the main bottlenecks using librarytimer to get reasonably fast aros on genuine hardware. im certain it is possible, using a limited color iff icon set. unfortunately there is no good icon editor.
- you have to set targets for the bounties, btw, that we know what we pay and what we get. im not sure if im going to contribute to bounties, i habe no paypal anymore but i might transfer some directly.

as proposal to be disputed: bounty to strip alos binaries of startup code in order to reach the amiga binary sizes, or have them compilable under vbcc, introducing vbcc to the build system as optional. i know it is huge, you have dismissed it before, as vbcc didnt seem to grant faster execution, therfore i put it up for dispute.

[clebin]

Originally Posted by OlafSch

he left for some time but is active. If asked he is still doing something when there are certain requirements that can be defined.

Good to know - thanks. It would be easy enough to draw up a list of requirements for Workbook. In the nightlies I've tried, the file-management functions are greyed out, it doesn't support standard .info files and there's no drag & drop. Is disk-space on the boot ADF a problem? You could break up the work something like this:

Step 1: make Workbook a viable view-only environment

* Support standard .info files
* Manually position & snapshot icons and windows
* List-view

Step 2: add basic file-management

* Drag & drop icons to copy/move
* Make new directories
* Delete files & directories
* View/edit file information/tooltypes

Step 3: add the basic niceties
* Leave out/put away
* Background patterns and images

Each step would take Workbook much closer towards being a viable Workbench-replacement.

[deadwood]

Originally Posted by OlafSch

what is missing? Aside from desktop

I would also be interested answer to this question. We can't use "full feature set of AmigaOS 3.1" as validation criteria for a single bounty - that is too broad, you could always find one more thing that is not ok and the developer would never finish the bounty.

[deadwood]

Originally Posted by wawa

- make scalos from aros contribs work properly on 68k, im trying to debug it but since there is no proper guide how to debug aros on say winuae, im having hard time in it, given a bit limited time currently.

Isn't Scalos 68k by origin? The original version does not work with AROS 68k?