@mailman2

So you are saying, and i agree, that aros and other ng systems developments are slowed because people like you prefers to use a 25 years old os emulated by a mastodontic os. My compliments!

@Chain-Q

Interesting tweet

I think I read from one of the MorphOS developers that SMP without breaking compatibility is impossible and the AROS devs were trying to implement SMP but it would need major changes. So I think (as you said) that the tweet exactly means what is written, a kind of PowerUP concept so only software that is adapted will benefit from it. And of course you need more than one core what reduces it to X1000 and X5000 users.

Last edited by OlafS25 on 23-Apr-2014 at 12:59 PM.

@ilbarbax

I would say it the other way round... 68k/FPGA development is slowed because developers and users are split on 4 platforms now

But I still not understand why the end of the only open amiga platform would have improved the situation. Then we would not have the rom replacements and Aros 68k either and would have only the choice between emulation like amigaforever (without any chance of development) and two commcercial closed platforms. The situation would propably be better if it would be like in the C64 or Atari community with only one platform (and no NG options) but the situation is how it is. I do not understand why mailman is bashing the only open platform.

@mailman2

I think the biggest mistake, or rather the most unfortunate thing, in this community is not AROS, but the split between MOS and AOS4. I mean they run on the same hardware, almost, have pretty much the same feature set, almost; yet incompatible with each other.

If that split had never taken place, like for example if the driving forces behind PPC was not hindered by Amiga management and stupid lawsuits back in the days the people behind MorphOS would not have had to spend all this time and effort developing MOS from scratch, but could have based it on AmigaOS sources, like OS4, and gained a lot of years of development. And AmigaOS would not have needed its own set of engineers to reinvent the new shiny wheels the MOS guys already had done. And there would not have been a split of engineering forces in two equal, almost, OSes.

The incentive behind AROS is not entirely the same as MOS/OS4. AROS wanted to run Amiga on x86, no matter what, while MOS/OS4 simply strived to develop Amiga further on the already chosen platform. None of us have fast 3D or HW accelerater video decoding, but we could have if we were joined instead of split...

_________________
- Don't get fooled by my avatar, I'm not like that (anymore, mostly... maybe only sometimes)
> Amiga Classic and OS4 developer for OnyxSoft.

@Deniil715

Quote:

The incentive behind AROS is not entirely the same as MOS/OS4. AROS wanted to run Amiga on x86, no matter what, while MOS/OS4 simply strived to develop Amiga further on the already chosen platform. None of us have fast 3D or HW accelerater video decoding, but we could have if we were joined instead of split...

i'm guessing you meant on more recent video cards; because on MOS 3D and video decoding absolutely flies on the older, PCI-only radeon cards.

-- eliyahu

_________________
"Physical reality is consistent with universal laws. When the laws do not operate, there is no reality. All of this is unreal."

@Deniil715

Less platforms would have been better. Sadly it seems that expecially MorphOS and AmigaOS are more and more going in different directions. The similarities between AROS and MorphOS are bigger (on API level and what components are used). But the basic design decisions of the past are not changeable anymore. Besides there are the personal differences between team members of the different camps with a lot of insultings, personal attacks and unproved claims in the past that make it propably impossible to even directly talk with each other. It would need a major change of attitudes and I do not see that happening. And I am not sure if most developers/users really are interested in that. Most have their favorite platform (mostly AmigaOS or MorphOS) and are only interested in that platform. There is no real pressure from the customers to change politics. So I think it will go on like now.

@OlafS25

Aye, tho the OWB/fab fundraiser and kas1e adapting it to AOS4 shows that people can have a constructive dialogue. It kinda depends on the forum for example. Both this and amiga.org can get quite heated, while Amigans.net is very "down to buissniss and on topic" from what I see.

And the newer productions have had cross platform versions too, which is nice.

@Overflow

Amigans.net is "on topic" because they remove everything that is not "AmigaOS" related. The same is true for all specialized forums like morphzone or aros-exec. You cannot compare that with f.e. amiga.org. It is nice from Fab to open the sources and he always was and is very helpful. But that is not what I would have meant with cooperation and direct talk. Too much on both AmigaOS and MorphOS is help captive and closed and both sides refuse to open sources or at least offer compiled binaries. And most basic components are different and even when both sides have the same requirements they are reinventing wheels (but of course not compatible to the other side) like OpenURL replacement on AmigaOS. I could easily make a long list just from memory.

@Vistaus

>OS 4.2 will have SMP support, according to a recent tweet by the official AmigaOS twitter account. So I don't see why AROS wouldn't be able to do it (or reverse engineer it at most) after 4.2 lands.

They would not need to reverse engineer. They can do clean new implementation.
AROS lacks 68k binary compatibility / relies on UAE, while AOS4.1 (and MOS) has run 68k apps as natively as possible, emulating only CPU.
So, keeping current legacy binary compatibility is harder for those (when AROS is not directly 68k binary compatible on x86/x64). (but they have the same backup plan as AROS, UAE/68ksandbox)

But as far as I see it AROS "main" goal still is AOS3 API compatibility and that needs to change for NG.

>The only thing x86 (or 64-bit) will solve is the price point of a new Amiga, as the hardware is cheaper than PPC hardware.

x86 HW is not cheaper if custom made for Amigalike OSs.

It is possible to build 1Ghz+ PPC computer for eur200. That's pretty much the price also for cheapest x86 desktops, btw. OS price comes on top of that.

@SMP vs AMP vs BMP vs OpenCL....

Perhaps AOS4 will mix all those.
(btw. it was recently demonstrated by codewarrior people that AMP is a must if you have more than 8 cores for a job at hand) (SMP vs AMP on Qoriq..)

@ssolie

It would be nice to know what you are testing with xkernell and related to multicore support.
(I do not see any good reason to not to tell what is going on. There is no true competition to be afraid of etc. Amigalike flavours are going in different directions after all.)

Last edited by KimmoK on 23-Apr-2014 at 03:27 PM.
Last edited by KimmoK on 23-Apr-2014 at 01:53 PM.
Last edited by KimmoK on 23-Apr-2014 at 01:51 PM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@OlafS25

Quote:

OlafS25 wrote: I think I read from one of the MorphOS developers that SMP without breaking compatibility is impossible and the AROS devs were trying to implement SMP but it would need major changes. So I think (as you said) that the tweet exactly means what is written, a kind of PowerUP concept so only software that is adapted will benefit from it.

Well software like datatypes or 3D softwace could befefit from some kind of (AS)MP feature.

But software supporting this cores are not the main issue behind this. There might be a system for starting processes on additional cores, but as long as they are restricted to 'supporting functions' or 'nuber crunching' you will be slowed down by the maximum speed on the 'main core'.

This kind of system will only give you poor scaling, a system with 8 cores won't have 8 times the performance of a single core system, perfoemance might be limited to 2 times of a single core system.

(AS)MP might be a nice intermediate solution for AROS or systems with only two cores but the trend on x86 or ARM is multi core systems with at least 4 cores (uncluding hyper threading)

Last edited by Arko on 23-Apr-2014 at 02:29 PM.
Last edited by Arko on 23-Apr-2014 at 02:27 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

@Vistaus

Quote:

OS 4.2 will have SMP support, according to a recent tweet by the official AmigaOS twitter account

Good thing they have an "official AmigaOS twitter account" now - with that much information to share, you can't have enough channels to use.

But AmigaOS will not have SMP, because ssolie apparently "doesn't like that term". And if you claim he promised it in the past, you're just intentionally misrepresenting what he wrote and you're a troll.

OS 4.2 has been around the corner for years now. And who knows when and if ever we'll see some of those features implemented...
They also need to port and release OS4 for new Cyrus boards... a lot of work for few guys.

_________________

@Arko

Propably any Amiga solution will scale worse than Linux/Windows/MacOS because it was never designed to support this. Or you make major changes but then it has to be seen what software still works and where problems with other components arise. I think "silly SMP" was about that, examining what has to be changed and what the consequences are. On AROS (X86/X64/ARM/PPC) there is less software to break and most still adaptable. Old 68k software is not affected. On MorphOS and AmigaOS the situation is very different and breaking compatibility would mean a loud cry. So some sort of PowerUP is most propable from my point of view.

The MorphOS devs indicated that a ISA change would be used in future to make clean implementations of modern components like SMP but that would mean breaking compatibility. And 68k software would run in a box similar to Aros. "SMP" on the current platform and with keeping compatibility is hard to imagine.

Last edited by OlafS25 on 23-Apr-2014 at 04:05 PM.

@thread

None of your responses mean anything as all your statements can be doubted.

Y'all forgot to say 'No doubt', like @mailman2 did,

Three times

Quote:

mailman2 wrote: ... There is no doubt that due to the lack of support for multiple cores and the lack of commercial quality NVidia drivers. AROS is useless and should be canceled many years ago - no later than 2007. There is no doubt that AROS is not and never will be good enough to justify resignation from use Windows/Linux/Os X. ... Resignation of Aros would allow the introduction of a single binary platform - software developer could prepare a 68k exe which would work on Amiga Os/Morphos/Amiga Os 3 on WinUAE. What no doubt have a positive influence on the development of software for our community.

No doubt, there is no one who can argue with that logic. AROS has NO use. Never is a finite period of time. And I wonder what other miracles would come to pass if only we could get rid of AROS?

_________________
All the way, with 68k

@agami

miracles? Perhaps world peace? Man really understand woman? Paradise on earth

@OlafS25
Quote:

Man really understand woman?

The older I get the more sure I am I don't want that!

_________________
MorphOS, classic Amiga, demoscene, and stuff
"When a bridge is not enough, build a Viaduct!"
"Strip the Amiga community of speculation and we can fit every forum on a 720k floppy" (by resle)

@cgutjahr

Or you're just a lousy forum guy? What's up with people calling me a troll? This is like the 3rd forum (and those other 2 are not even Amiga forums) where I get called a troll for misreading something or posting a helpful comment. I wonder if people like you even know what the word 'troll' means in internet slang, because this is getting out of hand.

Anyway: the tweet said "multi-core support" so I assumed that meant SMP, but I was wrong on that part and I'm sorry for causing a stir.

_________________
Proud user of AmigaOS 4.1 on an AmigaONE 500. This is the first Amiga I've ever had so I don't know all the ins and outs of AmigaOS yet, so I'm sorry if I'm asking noob questions and stuff.

@Vistaus

don't worry about it. you didn't do anything wrong, and you weren't trolling.

-- eliyahu

_________________
"Physical reality is consistent with universal laws. When the laws do not operate, there is no reality. All of this is unreal."

@mailman2

Quote:

According to the x86 fanboys this should solve all problems. In fact, it is a scam and an attempt to force people to use a solution that has all the disadvantages of ppc and none advantages of ppc.

With this part, I fully agree.

The often touted "X86 advantages" are based on misassumption, that AmigaOS4 would be allowed to run "any compatible hardware" (like MorphOS runs on compatible Apple PPC systems, and AROS runs on compatible 68k, x86, arm, ppc etc. systems), and not require "special" AmigaOne hardware, like it does now.

Sure, there might be a minor difference in price, when some small manufacturer produces small batch of "custom" X86 motherboards (as opposed to PPC), but they'd still be very expensive. Price increase by this "custom hardware requirement" far exceeds PPC vs. X86 price difference.

Of course with X86 it would be WAY easier to "do the Eyetech thing", take some existing hardware and simply rebrand that as "AmigaOne".

@Jupp3

Unless it would require some very spesific x86 hardware, most of us got one of those at home already.

Kinda like with SteamOS, same hardware, just not Windows. No hardware investment for the ENDuser.
For the OS developers on the other hand...

Tho, this topic has been raised ...God knows how many times, and it leads nowhere

Last edited by Overflow on 23-Apr-2014 at 07:46 PM.