Poster | Thread |
WolfpackN64
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 11:11:18
| | [ #21 ] |
|
|
|
Regular Member |
Joined: 9-Oct-2016 Posts: 300
From: Unknown | | |
|
| @paolone
Adressable RAM is something that can be fixed. What I meant is the amount of RAM people have, not how much a stick theoretically costs.
A hybridization with Linux would only be possible through a syscall emulator like Wine or through a hybrid kernel, which is way too difficult and no Amiga or Linux programming group is ever going to try that, let alone making a Wine-like middleware. You might just as well plop AmigaOS in a VM.
The evolution you propose is more of a definitive death of AmigaOS than anything else. Why would you develop for Amiga when it's embedded in Linux, which has far more developer tools?
I see more use in a reverse scenario. In an enviroment like AmiCygnix which makes porting Linux apps easier or an implementation of bits of another kernel. If Redox OS will ever be able to implement Linux syscalls, that will be a good candidate since it's also a microkernel.
I like Linux, but I'm not a fan of it's monolithic kernel structure, which I see as a waste of space and resources.
And you could also ask yourself. Does the world really need another UNIX system? I can (and actually) use Linux just fine, I'm interested in Amiga because I find the classic style OS' interesting. Merging these will make the latter pointless. |
|
Status: Offline |
|
|
terminills
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 11:23:53
| | [ #22 ] |
|
|
|
AROS Core Developer |
Joined: 8-Mar-2003 Posts: 1472
From: Unknown | | |
|
| @WolfpackN64
My main computer has 32 gigs of ram. My laptop has 8 gigs, most tablets now have 2 gigs.
Plus on top of all that you can build a stripped down linux kernel that is very light.
Quote:
A hybridization with Linux would only be possible through a syscall emulator like Wine or through a hybrid kernel, which is way too difficult and no Amiga or Linux programming group is ever going to try that, let alone making a Wine-like middleware. You might just as well plop AmigaOS in a VM.
|
It's much easier than you seem to think.
_________________ Support AROS sponsor a developer.
"AROS is prolly illegal ~ Evert Carton" intentionally quoted out of context for dramatic effect |
|
Status: Offline |
|
|
WolfpackN64
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 11:26:32
| | [ #23 ] |
|
|
|
Regular Member |
Joined: 9-Oct-2016 Posts: 300
From: Unknown | | |
|
| @terminills
Then what do you propose? Maybe a LKM? I still don't see what a Linux Amiga hybridization would do for the Amiga platform. I think you'll be going down the same path that brought OS/2 down. |
|
Status: Offline |
|
|
terminills
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 11:33:41
| | [ #24 ] |
|
|
|
AROS Core Developer |
Joined: 8-Mar-2003 Posts: 1472
From: Unknown | | |
|
| @WolfpackN64
You should think more along MacOS X.
A hybrid linux/Amiga kernel could easily allow cheaper driver development. Less time developing hardware support means more time can be used for userspace.
Sure by default the linux kernel is huge. However there's no reason you need to build a kernel that supports everything.
_________________ Support AROS sponsor a developer.
"AROS is prolly illegal ~ Evert Carton" intentionally quoted out of context for dramatic effect |
|
Status: Offline |
|
|
WolfpackN64
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 11:41:29
| | [ #25 ] |
|
|
|
Regular Member |
Joined: 9-Oct-2016 Posts: 300
From: Unknown | | |
|
| @terminills
Now that's something else. But then I'd still opt for them to implement only certain parts of Linux in the kernel and put as much as possible in userspace. I think a hybrid kernel is a bit of a poor compromise and I'd rather see them keep the microkernel.
Once gain, if the Redox kernel can succesfully make Linux syscalls, it would be be better to take a look at that since it'll be a lot less code to implement. |
|
Status: Offline |
|
|
pavlor
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 11:58:29
| | [ #26 ] |
|
|
|
Elite Member |
Joined: 10-Jul-2005 Posts: 9593
From: Unknown | | |
|
| @g01df1sh
My list of 2016 Amiga achievements sorted by time:
January/February MorphOS SDK 3.10 (MorphOS) Odyssey 1.23 mini-update (OS4)
March/April Warp3D Nova (OS4) Odyssey 1.25 (AROS)
May/June WinUAE 3.3.0 (emulation) GBAPII++ Cirrus Logic GD5434 based ZII card (classic)
July/August Catacomb 3D port (classic) MUI5 (OS4)
September/October A1-X5000 released (OS4) Vampire 500 V2 first pieces sold on eBay (classic) Tales Of Gorluth II (classic) Prisma Megamix (classic)
November/December MNT VA2000 FPGA based GFX ZII card (classic) WinUAE 3.4.0 (emulation) |
|
Status: Offline |
|
|
duga
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 12:34:56
| | [ #27 ] |
|
|
|
Regular Member |
Joined: 1-May-2012 Posts: 227
From: Unknown | | |
|
| Hopefully OS 4.1 FE update 1 this year. |
|
Status: Offline |
|
|
uponthevoid
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 12:48:29
| | [ #28 ] |
|
|
|
Member |
Joined: 22-Dec-2015 Posts: 15
From: Unknown | | |
|
| I forgot to say....
SSL Update. Please!
|
|
Status: Offline |
|
|
paolone
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 19:50:44
| | [ #29 ] |
|
|
|
Super Member |
Joined: 24-Sep-2007 Posts: 1143
From: Unknown | | |
|
| @WolfpackN64
Quote:
The evolution you propose is more of a definitive death of AmigaOS than anything else. Why would you develop for Amiga when it's embedded in Linux, which has far more developer tools?
|
More or less, as definitive as MacOS X has been the death of MacOS.
Yes, a radical change of paradygms, you know, that some times becomes mandatory when all the world around has changed as well. I guess there is room for a better Amiga API, and there is room for a better Linux as well. That's why I guess hybridizing them would lead to a interesting experiment.
Problem is, unluckily, that a similar experiment is really pricey to do. |
|
Status: Offline |
|
|
WolfpackN64
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 23:10:31
| | [ #30 ] |
|
|
|
Regular Member |
Joined: 9-Oct-2016 Posts: 300
From: Unknown | | |
|
| @paolone
Except that, just as people cling on to AmigaOS, there are also a substantial amount of people who cling to the classic MacOS (up to Mac OS 9).
Because they think it offers something special, perhaps in some cases, something better.
If you don't believe that your platform has something special to offer, why keep following it. |
|
Status: Offline |
|
|
Beans
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 23:11:36
| | [ #31 ] |
|
|
|
Regular Member |
Joined: 26-Aug-2016 Posts: 447
From: Bear Delaware USA | | |
|
| @WolfpackN64
What AmigaOS uses is not in actuality a microkernel, although MorphOS does use a microkernel. The core of AmigaOS is just a heck of a lot smaller than the Linux kernel. A true microkernel only places the most essential services in a tightly protected core, with all the addition crap (that frankly shouldn't be there) in the Linux kernel relegated to layers outside that core. It is a more structurally sound way of building an OS. Think of it as being similar to the layers of an onion, where the most basic and important parts are kept in the center and the OS grows out progressively, in layers, from that.
Nothing prevents a UNIX like OS from being built this way, Linux is not and cannot be structured this way because from its inception, it has been a fairly sloppily constructed device. Sorry guys, but that's just the case, and something it in a way inherited from the haphazard way most UNIX variants themselves grew.
Hybrid kernels exist primarily to provide some of the benefits of a microkernel while assuring compatibility with related OS' based on monolythic kernels.
They are still sub-optimal.
I'm actually not that impressed with Torvalds, or his opinions on microkernel based OS'. They have provided a stable and more secure solution for operating systems that are primarily used in process control and embedded applications since the 80's.
While some of you might want to default to a Linux or Linux like base, primarily out of convenience, that approach is simply not that Amiga like (with all due appologies to our history of using BSD based development systems).
My position? If you want Linux, use Linux. Don't try to pollute my operating systems of choice with your adhoc communally developed crap. Last edited by Beans on 23-Dec-2016 at 11:14 PM.
_________________ Amiga! "Our appeal has become more selective" |
|
Status: Offline |
|
|
WolfpackN64
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 23:24:20
| | [ #32 ] |
|
|
|
Regular Member |
Joined: 9-Oct-2016 Posts: 300
From: Unknown | | |
|
| @Beans
That's why I've also been following the Redox project, since it aims to be a modern, light and secure Unix OS on a microkernel design.
I'm largely platform agnostic, Linux will be my main OS out of usability and gaming concerns, but I agree melding AmigaOS with Linux isn't a good idea.
|
|
Status: Offline |
|
|
bison
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 23:28:25
| | [ #33 ] |
|
|
|
Elite Member |
Joined: 18-Dec-2007 Posts: 2112
From: N-Space | | |
|
| @amigang
Quote:
I'm starting to think for os4 it might be better to have a similar solution that Aros platform has been pushing and hopefully Alice when it come out and have linux run in the background but works in a transparent way that you can open linux apps within OS and they just act native, and still have full access to the OS even if its just running in a virtual machine. Could this be done on a PPC linux? |
Maybe, but... if AmigaOS could be modified to run hosted on Linux, then it is no longer constrained to PPC, and x86-64 is a much better option. This is close to what is happening already with 4.1 FE Classic running on FS-UAE running on Linux running on x86-64. Your proposal would of course eliminate the emulation layer, which would be an improvement.
_________________ "Unix is supposed to fix that." -- Jay Miner |
|
Status: Offline |
|
|
bison
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 23:47:20
| | [ #34 ] |
|
|
|
Elite Member |
Joined: 18-Dec-2007 Posts: 2112
From: N-Space | | |
|
| @WolfpackN64
Quote:
I still don't see what a Linux Amiga hybridization would do for the Amiga platform. |
It would eliminate the dire situation with PPC hardware, allowing AmigaOS to run on anything that runs Linux.
This wouldn't be my first choice -- I would rather see AmigaOS running directly on x86-64 or ARMv8. But it that's not possible, then I'm OK with hosting on LInux. It's better than nothing, which is where PPC will be in a few more years.
Last edited by bison on 23-Dec-2016 at 11:48 PM.
_________________ "Unix is supposed to fix that." -- Jay Miner |
|
Status: Offline |
|
|
TRIPOS
| |
Re: What Amiga achivments in 2016 Posted on 23-Dec-2016 23:54:34
| | [ #35 ] |
|
|
|
Super Member |
Joined: 4-Apr-2014 Posts: 1205
From: Unknown | | |
|
| @WolfpackN64
Quote:
Adressable RAM is something that can be fixed. |
Amiga compatibility stipulates 31 bit addressing space (2 GB), as well as an open unified memory architecture, big endian, single core/CPU, etc.
"Fixing it" means de-facto to abandon the Amiga compatibility. And if you have taken that step, it would be foolish to stop with only slightly more addressable RAM, when you can go for full 64-bit computing, true SMP, true memory protection and resource tracking, and small endian, so it could run on x64 machines or ARM maxhines in the future! |
|
Status: Offline |
|
|
WolfpackN64
| |
Re: What Amiga achivments in 2016 Posted on 24-Dec-2016 0:02:29
| | [ #36 ] |
|
|
|
Regular Member |
Joined: 9-Oct-2016 Posts: 300
From: Unknown | | |
|
| @TRIPOS
That doesn't mean AmigaOS4 is stuck in 32 bit mode forever. Windows also has a compatibility layer for it's own 32 bit apps, why wouldn't such a thing be possible under AOS4? |
|
Status: Offline |
|
|
Zylesea
| |
Re: What Amiga achivments in 2016 Posted on 24-Dec-2016 0:34:06
| | [ #37 ] |
|
|
|
Elite Member |
Joined: 16-Mar-2004 Posts: 2263
From: Ostwestfalen, FRG | | |
|
| @Beans
Quote:
Beans wrote:
My position? If you want Linux, use Linux. Don't try to pollute my operating systems of choice with your adhoc communally developed crap. |
Generally I agree to the above. But if you limit an alien kernel to just abstract the hardware (i.e. voiding driver issues, kind of Amithlon way), then why not? I mean look to MorphOS the underlying Quark kernel is de facto invisible to the user/programmer. We know close to nothing what Quark really is... It inits the hardware, starts a few kernel services, and provides some HAL. But more is AFAIK not known to the public. The lowest level we users/application programmers see on MorphOS is Exec.
_________________ My programs: via.bckrs.de MorphOS user since V0.4 (2001) |
|
Status: Offline |
|
|
Raffaele
| |
Re: What Amiga achivments in 2016 Posted on 24-Dec-2016 21:27:38
| | [ #38 ] |
|
|
|
Super Member |
Joined: 7-Dec-2005 Posts: 1906
From: Naples, Italy | | |
|
| @TRIPOS
Quote:
TRIPOS wrote: @WolfpackN64
Quote:
Adressable RAM is something that can be fixed. |
Amiga compatibility stipulates 31 bit addressing space (2 GB), as well as an open unified memory architecture, big endian, single core/CPU, etc.
"Fixing it" means de-facto to abandon the Amiga compatibility. And if you have taken that step, it would be foolish to stop with only slightly more addressable RAM, when you can go for full 64-bit computing, true SMP, true memory protection and resource tracking, and small endian, so it could run on x64 machines or ARM maxhines in the future! |
So abandon this obnoxious legacy and migrate to 64bit and anytime old software is invoked encapsulate it in sandbox emulation, once and for all... We are all tired of continuing debating on 32bit Amiga limits that prevent us evolving..._________________ "When the Amiga came out, everyone [at Apple] was scared as hell." (J.L. Gassée, former CEO of Apple France and chief of devs of Mac II-fx, interviewed by Amazing Computing, Nov 1996). |
|
Status: Offline |
|
|
Raffaele
| |
Re: What Amiga achivments in 2016 Posted on 24-Dec-2016 21:48:59
| | [ #39 ] |
|
|
|
Super Member |
Joined: 7-Dec-2005 Posts: 1906
From: Naples, Italy | | |
|
| @Beans
Quote:
Beans wrote: @WolfpackN64
My position? If you want Linux, use Linux. Don't try to pollute my operating systems of choice with your adhoc communally developed crap. |
I agree and there are many alternatives to Linux.
For example BeOS-Haiku is a very modern, stable and reliable microkernel and BeOS itself was build for multimedia from scratch, and inherited many from Amiga technology, such as datatypes.
Adapt BeOS-Haiku kernel to give us modern features and wrap around it Amiga APIs, Fliesystem, DOS, etcetera...
It is more or less what people of AnubisOS wanted to made with Linux Kernel and FriendOS are working on today, but using BeOS-Haiku kernel has more sense for us Amigans...Last edited by Raffaele on 24-Dec-2016 at 09:50 PM.
_________________ "When the Amiga came out, everyone [at Apple] was scared as hell." (J.L. Gassée, former CEO of Apple France and chief of devs of Mac II-fx, interviewed by Amazing Computing, Nov 1996). |
|
Status: Offline |
|
|
TRIPOS
| |
Re: What Amiga achivments in 2016 Posted on 25-Dec-2016 8:43:20
| | [ #40 ] |
|
|
|
Super Member |
Joined: 4-Apr-2014 Posts: 1205
From: Unknown | | |
|
| @WolfpackN64
Quote:
WolfpackN64 wrote:
That doesn't mean AmigaOS4 is stuck in 32 bit mode forever. Windows also has a compatibility layer for it's own 32 bit apps, why wouldn't such a thing be possible under AOS4? |
The solution you are talking about is UAE.
@Raffaele
Quote:
Raffaele wrote:
So abandon this obnoxious legacy and migrate to 64bit |
MorphOS announced this in 2011. It will be interesting to see how it turns out in the end.
Quote:
and anytime old software is invoked encapsulate it in sandbox emulation, once and for all... |
UAE...Last edited by TRIPOS on 25-Dec-2016 at 08:50 AM.
|
|
Status: Offline |
|
|