@nikosidis

there are a lot of Aros devs in the Arix project. I cannot imagine that they suddenly chose the dark (commercial closed source) side . I expect that both sides benefit from it and if they manage to allocate additional ressources and speed up development it can only be good for all sides.

@thread

Obviously something did not work like expected but we should give them the time. What is a couple of days in amiga-land where some measure progress even in decades (really ). Waiting a couple of days to get something that is competitive in todays world is not a long time to me. We all were understandably a little disappointed at first but we should simply give them time.

And to all those who insist on Arix being another Linux distro based on Linux hosted, this project started already a long time ago with lots of Aros developers, do they really think that all those developers are creating just another Linux hosted version?

@cdimauro

Quote:

cdimauro wrote: Quote: EDanaII wrote: Linux Kernel -> Linux OS -> AROS and this Linux Kernel -> AROaS There's no difference, in fact.

This makes me think about intelligence in a fresh loaf of bread again.

Quote:

cdimauro wrote: A Linux distro is an o.s. which uses Linux as the kernel. The userland and/or the window manager has no relevance in this.

Wikipedia says:

Quote:

A Linux distribution (often called distro for short) is an operating system built on top of the Linux kernel and often around a package management system. A Linux distribution is synonymous with GNU/Linux,

Source: http://en.wikipedia.org/wiki/Linux_distribution

So this "Linux Kernel -> AROaS" is not a Linux distro

Otherwise this "ExecSG -> AmigaOS" should never be called AmigaOS

Last edited by Arko on 18-Nov-2013 at 11:02 PM.

_________________
AmigaONE. Haha. Just because you can put label on it does not make it Amiga.

I borrowed this comments from here (#27 & #28):
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=38873&forum=2&start=20&order=0

@clebin

I do not know how the legal status of the project is. We will have to look carefully how the relation to Aros will be. If the kernel f.e. can be used for Aros and is closed source I would have no problem with it. Perhaps we will even forget the opensource branch when Arix is there who knows. What is most important is to me that Aros cannot be closed/destroyed by anyone so we will see what Arix means practically for Aros.

@nikosidis

Quote:

nikosidis wrote: Seams kinda strange that something based on Linux kernel and AROS is closed source. Is that even possible ?

Amithlon really had some parts of UAE and a Linux kernel and IMR it is still closed source.

I doesn't know how much Amiga like Arix will look and feel and I don't know how closed source it will be. But I'm still curious about it.

Last edited by Arko on 18-Nov-2013 at 11:01 PM.

_________________
AmigaONE. Haha. Just because you can put label on it does not make it Amiga.

I borrowed this comments from here (#27 & #28):
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=38873&forum=2&start=20&order=0

@ferrels

Ya know, ferrels, I really hate to say it, but...

Quote:

Quote: EDanaII wrote: @ ferrels Can you tell us what the difference between this Linux Kernel -> Linux OS -> AROS and this Linux Kernel -> AROaS In most people's view, one of these is hosted and the other is not. Can you see the difference? Ed. Actually that's for Dammy to explain. You need to ask him. I already know what it is.

At this point, it's quite apparent you're not interested in having an honest discussion. Not only did you not answer my question, you changed the subject back to Dammy. That makes this look more like you've got some kind of vendetta and are not even slightly interested in anything anyone else has to say.

So be it...

Ed.

@cgutjahr

Quote:

cgutjahr wrote: @ferrels Quote: So in short, Arix is just another Linux distro with window dressing on top to make it look like Aros. It wouldn't be "another Linux distro", it would be AROS with a different Kernel. You probably wouldn't notice any difference when using it, even if you stopped yelling at the screen for a minute.

A genuine question here, from someone who is fairly intimately familiar with: Amiga OS3, OS4 and a user of various Linux distros including maintaining my own specialised Linux distro for the last 12 odd years, having done the odd bit of kernel hacking, etc....

Where exactly does "amiga" end and "linux kernel" begin in this kind of setup? Are "AmigaDOS" process mapped to Linux kernel processes? Or are all "Amiga" process in one single Linux process? If the answer is the latter then a big advantage in Linux is lost. If the answer is the former, then even any type of source compatibility would surely be impossible? What about file system semantics, file locking, case sensitivity issue in filenames, etc? Amiga message ports, and the general way of dong Amiga-style IPC? I would like to understand how all these issues map to "AROS with a different kernel".

I can see how a userland could be constructed where Amiga processes sit inside a single Linux process, and that way drivers for things like SATA interfaces, network cards, could be leveraged, but surely that's what AROS on Linux hosted is (never actually used it, but am intuiting what that does).

This is only a technical interest. Not interested in the fighting that seems to be going on round here lately.

_________________
Blog

however, after 30 days, the alpha version had to be ready to download..

_________________

@Aslak3

Quote:

Where exactly does "amiga" end and "linux kernel" begin in this kind of setup?

I don't know any details about this project, I was just making assumptions based on the fact they decided to rehash the older project's nickname.

Quote:

I can see how a userland could be constructed where Amiga processes sit inside a single Linux process, and that way drivers for things like SATA interfaces, network cards, could be leveraged

That's more or less what Amithlon did (and quite well, actually) - but I don't see how they could use that approach and show a screenshot of an AROS application displaying CPU load on several cores.

Quote:

but surely that's what AROS on Linux hosted is

No, AROS hosted is pretty much AROS in a virtual machine. Boot host OS, then boot AROS inside a virtual machine.

If I can buy a fully supported netbook with ArixOS in the 'near' future for a fair price count me in! (assuming we hear what ArixOS is all about )

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

@Seiya

There has been, as far as we know, 2 people using some form of alpha build for a few months (link). If this was a call for beta users then more info would be helpful to guage interest.

After 30 days I would have expected more out of this. In fact I would have expected dammy or someone from the team to be here answering questions by now.

_________________
My personal blog - CritAnime.com

Admin at Commodore Gaming Wiki

@cdimauro

Quote:

There's no difference, in fact. A Linux distro is an o.s. which uses Linux as the kernel. The userland and/or the window manager has no relevance in this.

With that logic Mac OSX is nothing but a BSD distro and android on our smart phones is just another linux distro as well.
The kernel has very little to say about how the user experience is. I would have agreed if it was a linux distro just with a different window manager/theme slapped ontop.

I am still worried though since linux does not have a microkernel which i fear might affect things like how responsive the the system is.

Windows is a Linux distro.

MacOS X is a fork of BEos.

AmigaOS is illegal.

Linux does not exist.












Arix will be released in -92 days, -16 hours, -53 minutes, +4 seconds.

Last edited by resle on 19-Nov-2013 at 04:33 AM.

@Tomas

Quote:

I am still worried though since linux does not have a microkernel which i fear might affect things like how responsive the the system is.

In theory, monolithic kernels should offer slightly better performance - all other things being the same. Microkernels are more secure(most mission-crtitical (RT)OSes are mikrokernel).

Last edited by WolfToTheMoon on 19-Nov-2013 at 08:11 AM.

_________________

@CritAnime

dammy is still banned so he cannot say much here

@WolfToTheMoon

Quote:

WolfToTheMoon wrote: @Tomas Quote: I am still worried though since linux does not have a microkernel which i fear might affect things like how responsive the the system is. In theory, monolithic kernels should offer slightly better performance - all other things being the same. Microkernels are more secure(most mission-crtitical (RT)OSes are mikrokernel).

The micro vs. monolithic discussion is quite useless.

OSX and WinNT(XP, 7, 8) are claimed to have micro-kernels. But for better performances some important parts of the driver system are integrated into the kernel.

Linux, has a monolithic kernel but it has some stubs for drivers that could load hardware driver in a running kernel, a feature usually availe only for micro-kernel systems.

AFAIK there is no major Unix like OS using a pure micro kernel.

----

AmigaOS

Some people here might respond by a sentence starting with "But the Amigas ...."

The AmigaOS is irrelevant when it comes to micro- vs. macro-kernels, because it uses a shared address space.

Passing pointers between the kernel and the drivers is impossible in micro kernel systems, data has to be copied, this will results in lower performance.

So it has the advantages of a micro and a makro kernel because it does not know the difference between user and kernel kernel code.

---

Responsiveness

I don’t know how people here are measuring responsiveness. On RTOS systems you have a guaranteed time for a reaction of the OS. On Amiga you can spoil this, using the forbid() call in a wrong way.

You should define responsiveness by “Time used until a response” the TCP/IP drivers in Linux are built into the kernel and will give quit good responsiveness. IMR Windows has done the same fot TCP/IP and parts of the GUI.

TCP/IP and file operations are very important on these systems and they deserve a huge grade of priority and responsiveness, but waiting some minutes when the Explorer tries to access, not available resources over the network, will not make the system look fast in the eye of the user.

So until there is no real definition for “responsiveness” a WindowsXP frozen after a network problem might still have a higher responsiveness than an AmigOID trying the same.

_________________
AmigaONE. Haha. Just because you can put label on it does not make it Amiga.

I borrowed this comments from here (#27 & #28):
http://amigaworld.net/modules/newbb/viewtopic.php?topic_id=38873&forum=2&start=20&order=0

@Arko

Quote:

AFAIK there is no major Unix like OS using a pure micro kernel.

MINIX? Although, IIRC, it also runs some drivers in kernel space...

_________________

@EDanaII

Quote:

EDanaII wrote: @ ferrels Can you tell us what the difference between this Linux Kernel -> Linux OS -> AROS and this Linux Kernel -> AROS In most people's view, one of these is hosted and the other is not. Can you see the difference?

I sure can't. Hint: there is nothing called "Linux OS", Linux is the kernel including drivers.

@OlafS25

Quote:

OlafS25 wrote: @nikosidis there are a lot of Aros devs in the Arix project. I cannot imagine that they suddenly chose the dark (commercial closed source) side . I expect that both sides benefit from it and if they manage to allocate additional ressources and speed up development it can only be good for all sides.

How can that be when AROS have just a few developers? Heck even counting all Amiga-ish system and application developers there are just a few.

This makes a difference - if this project is closed source it's dead already. If it's open source (not likely when Dammys involved) than it's probably dead anyway . The target audience is too limited and active developers just aren't there.

IMHO of course.

@megol

look at the list... I do not talk about hundred or thousands of developers but "many" in amiga sense. Dammy is certainly not the head of it and the person that does all the work or owns everything so I do not care about if Dammy is in or not. We should simply wait what they will show us and what the concept will be, f.e. how support mobile devices, how to motivate developers to port f.e. games, expecially how make it easier to support the platform. Finally the content will decide if it has a chance or not. I know some glimpse more than you here and that sounded very interesting but finally we all have to wait for the full picture.

@Arko
Quote:

Arko wrote: The micro vs. monolithic discussion is quite useless. OSX and WinNT(XP, 7, 8) are claimed to have micro-kernels. But for better performances some important parts of the driver system are integrated into the kernel.

Neither of those are microkernels.

Quote:

Linux, has a monolithic kernel but it has some stubs for drivers that could load hardware driver in a running kernel, a feature usually availe only for micro-kernel systems.

LOL! No dynamically changes to the kernel environment isn't a microkernel feature. Nor is support for user mode drivers.

Quote:

AFAIK there is no major Unix like OS using a pure micro kernel.

What is traditionally called UNIX like can't per definition be a microkernel (this is using the definition as derived from the UNIX codebase - not the current "certified Unix" definition). But POSIX systems have microkernel designs - QNX is the most well known. If that counts depends on what you mean by major, QNX is widely deployed.

Quote:

AmigaOS Some people here might respond by a sentence starting with "But the Amigas ...." The AmigaOS is irrelevant when it comes to micro- vs. macro-kernels, because it uses a shared address space. Passing pointers between the kernel and the drivers is impossible in micro kernel systems, data has to be copied, this will results in lower performance.

Oh. I guess K42, Nemesis etc. aren't microkernels then. Either that or the above is wrong. Even modern L4 kernels avoids copying in most cases.

Also explicit copying can actually increase performance by reducing coherency traffic, enabling higher performing security checking and a lot of other effects.

Quote:

So it has the advantages of a micro and a makro kernel because it does not know the difference between user and kernel kernel code.

The main advantage with microkernels is limiting the trusted computing base.