@In_Correct

The company I work for builds industrial control systems for manufacturing, like welders, hydraulic lifts, etc. One of our most important customers is the #@%$^& *^&%#$ government. I can't really say much more.

I am not familiar with an Amiga x86, unless you mean the machine assembled from standard PC parts that is sold on Visalia (sp?) and to me, that one doesn't need to be that expensive. When you use off-the-shelf parts, I would think you would be striving for a less expensive system. Of course, the demand dictates the price...

If we are talking desktop, without a ton of R&D, i'd say x86 is the way. If we are going to get a concerted effort together and make a team of programmer to work on a "true" ARM port, i'd say ARM. But Aros as a Linux app isn't really interesting. AEROS is a bit more interesting, because you're looking at using Aros as a WM on Linux, while having the added bonus of having a Aros API implementation available as well.

68k isn't outdated, depreciated might be the proper term. As Freescale supports the newer variants of the architecture primarily for legacy support applications. But as 68k is the native architecture of Amiga, it should really never be dropped. The cost of including a 68k VM on whatever architecture Amiga-like systems use in the future will be very low, considering the advances in technology.

ColdFire is a newer (about 10-15 years old) architecture by Freescale that implements the native asm syntax of 68k machines, but the binaries are incompatible. Basically, the asm mnemonics are the same, but the compiled binaries are not. So they share a common lineage, but are different.

And thanks for the welcome, I hope I can really bring something significant to the table to share with everyone.

Quote:

In_Correct wrote: @Cod3r It is nice to see somebody else that might help. I am not sure what your company does but it sounds interesting especially if you say that it might be able to help with Amiga systems. If you know how to code and can code for AROS, then I think you should do so. You can probably do a lot. As for an Amiga X86 running whatever, or any X86 computer running AmigaOS, I don't like that idea especially if the computers are overpriced. The idea to make Amiga X86 and overpriced already exists, and I would not buy overpriced x86 Amiga any more than I would buy an overpriced x86 Apple Macintosh. (Even if the AmigaOS OS Family are much better than Mac OS X) I am not completely against AmigaOS X86 port someday... someday in the very distant future, but I would be completely against it if they port to x86 and drop all other support. My experience with x86 computers has been silly. I would like to see ARM Commodore &/or Amiga Netbooks & Gadgets running AmigaOS, MorphOS, AROS, or all of the above... but I can dream on. You are quite young and you already know how to code. I am young also and still need to learn how to code. ...and I am not sure why 68K is "outdated", and I don't know what a ColdFire is. I am going to try MorphOS when I get G4 PowerBooks. I will put AROS on an empty laptop someday. and I will try AmigaOS when they make a netbook. You are wise in picking AROS if you think that is the best thing you could help program. You would probably be overwhelmed if you did things for all three OS. Best Of Luck.

@OlafS25

Quote:

OlafS25 wrote: @KimmoK "I think both AOS4 and MOS can do things that AROS can not." But the list is rapidly getting shorter

Mainly its integration of 68k libraries and devices, drivers and file systems , that can be used on AmigaOS4.1 PPC and MorphOS, by powerpc native programs, and 68k applications that can use native PowerPC libraries and devices and drivers, AmigaOS and MorphOS is also able to use 68k gui systems, like wizard and bgui systems, also 68k programs that use cdplay.library can use a powerpc native replacement that uses AHI instead of SCSI commands, also AREXX scripts can be written that controls 68k and PowerPC native programs in the same scripts.
None of this can be done on AROS x86/ARM/PowerPC,
(AROS 680x0 only runs one CPU architecture so it’s not a problem.)

Besides this MUI is most updated on MorphOS, 2en most update on AmigaOS4.1, 3rd most on AmigaOS3.9, support the most important stuff on AROS but not all.
AmigaOS4.1 also has large depository of software, MorphOS also has lots of nice software, AROS is a bit behind.
You can always run classic stuff in UAE if OS does not support the software, and this possible on all 3 platforms, Aros has more integrated version called Janus-UAE, AmigaOS4.1 has tool called RunInUAE that aids in starting programs in UAE if you like to do that, you do not need too.
And also the desktops are better on AmigaOS4.1 and MorphOS, (MorphOS uses a more Dopus like desktop)
On 3d drivers AROS has golem 3d, it is only something that is planned on AmigaOS4.2, so the 3d part is better supported on AROS.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@NutsAboutAmiga

Yes of course the 68k integration is better on both PPC platforms and will always be. Aros is designed to be portable to different platforms so similar integration is not possible there (i think AOS and MorphOS use special abilities of PPC for that). On the other side AOS and MorphOS will have a problem when they are ported to another architecture (ARM or X86) and will end with a similar solution like Aros. Or they will both stick to PPC forever.

I am right now testing games and applications (WinUAE, Aros 68k, Kickstart Replacement) and many programs are working. If you want to use games you have to use UAE anyway. And Aros 68k will be in near future in all distributions for 68k support out of the box.

Zune-Update is in the works and a port of Scalos already available on ABI1 (hopefully with 64bit and multicore-support)

New concepts like aeros now available

And the biggest advantage, it is Opensource and runs on lots of different (and cheap) hardware.

@noXLar

I would welcome the challenge and the offer of course, but I never seen any OS 4.1 code, so I have no clue what I could do for it, or even what I'd be up against.

But I am willing to try anything. What would you want me to do? You'd be picking up the tab for the hardware, so I would guess that would be up to you.

I am best at low-level programming. If you needed a kernel, firmware, drivers or something, that would be me. If you needed something like MS Office... I could do it, but I probably wouldn't be the best to choose to implement something like that.

At work, I do firmware implementations, supervisor/kernel stuff. I am most comfortable with RISC architectures (MIPS, ARM, Blackfin, etc.), and I guess PPC would qualify, as it is a RISC design. I just haven't done any PPC work, at least not assembly level, only C.

I could get some help from my brother-in-law, he was actually at Motorola back in the PPC days. On a side note, that is one guy who can REALLY do some stuff, I mean like impossible stuff. But he never got into Amiga, and he is doing really well with his own company.

Anyway, i'd be willing to try. Tell me what you want me to do, i'll look into it, and let you know honestly if I think I can do what you need me to do.

Quote:

noXLar wrote: @Cod3r If i was able to donate a sam440 with os4.1 for you, would you consider use your coding magic for os4? Would of course buy a new one from acube systems, just now they are cheap. And if you interested, what would you consider doing?

@Cod3r

Quote:

@In_Correct The company I work for builds industrial control systems for manufacturing, like welders, hydraulic lifts, etc. One of our most important customers is the #@%$^& *^&%#$ government. I can't really say much more. I am not familiar with an Amiga x86, unless you mean the machine assembled from standard PC parts that is sold on Visalia (sp?) and to me, that one doesn't need to be that expensive. When you use off-the-shelf parts, I would think you would be striving for a less expensive system. Of course, the demand dictates the price... If we are talking desktop, without a ton of R&D, i'd say x86 is the way. If we are going to get a concerted effort together and make a team of programmer to work on a "true" ARM port, i'd say ARM. But Aros as a Linux app isn't really interesting. AEROS is a bit more interesting, because you're looking at using Aros as a WM on Linux, while having the added bonus of having a Aros API implementation available as well. 68k isn't outdated, depreciated might be the proper term. As Freescale supports the newer variants of the architecture primarily for legacy support applications. But as 68k is the native architecture of Amiga, it should really never be dropped. The cost of including a 68k VM on whatever architecture Amiga-like systems use in the future will be very low, considering the advances in technology. ColdFire is a newer (about 10-15 years old) architecture by Freescale that implements the native asm syntax of 68k machines, but the binaries are incompatible. Basically, the asm mnemonics are the same, but the compiled binaries are not. So they share a common lineage, but are different. And thanks for the welcome, I hope I can really bring something significant to the table to share with everyone.

again, couldnt agree more with whole above post. i see 68k as common denominator, that can be run virtually on all amigalike systems and even beyond that on other machines via uae or other, there is apparently even a solution to run amiga 68k binaries on mac os from the shell or something. therefore, what is most worth to support, is what meets the largest common user base and not the fragment of it, which is 68k.

btw, jason is currently most intensively working on sam440/460 support under aros, which is another example that aros might become the single solution to satisfy all, even if i proffered him to concentrate more on 68k;) really best to talk to him, being as you a newcomer that never used amiga before perhaps two years he is not spoiled by these clan wars and bad memories and luckily stays away from such debates.

@Cod3r

just to clarify things... AmigaOS is closed source commercial product. So you would not have any access to sourcecode (of the OS). And it is a very small community (between 500-700 users worldwide). But you can of course decide what you are interested more.

@Cod3r

what concerns os4 i think there is tiny chance to be allowed into the kernel development, others tried hard to even be accepted as beta testers. took years. i is all pretty much locked away. but why not, accept a sam, you might put aros on it soon as it seems ;P

@Nataline

I think confusion got the best of some of us. But the point is not to dwell on it, the point is to get it done. Infighting will do nothing but dampen spirits and disappoint everyone.

And yes, i'm ready to let the Cod3 do the talking

@Plaz

Thanks Plaz!Quote:

Plaz wrote: @Cod3r Welcome and good luck with any of the Ami projects you chose to contribute to. Plaz

@phoenixkonsole

I'm sorry but i'm lost, my friend. Do you mean the source for AEROS isn't available?

Quote:

phoenixkonsole wrote: @Cod3r You can't get anything regarding aeros.. i repeat me... if i would be your boss i would fire you. ; ) Why? Because someome has a finished design... has software and can save his company a lot of work. An x86 amiga board is ready since 2010. Now slightly outdated but nothing which couldn't be changed in an week.

@Plaz

What I made is a 68k emulator that runs on the ColdFire before the OS is booted. I can pass through ColdFire instructions (only some, not all because of duplicate opcodes) to be run native. A recompiler could work, but I believe that would be unacceptable, as you can't really count timing and cycles accurately with a dynamic recompiler. Well, I will say you could, but you'd need more horsepower on the CPU, and you'd have to carefully time your recompilation routines. Would be more trouble than it would be worth.

I changed some parts of Aros to better fit my solution, for performance and stability reasons. Unless someone followed my implementation exactly, my some of my adapted Aros code wouldn't even work.

Quote:

Plaz wrote: @Nameless Yes, there are tools that will recompile 68k code for use on CF. You can take any source for 68k and compile it for 68K. Cod3r has done just that with AROS code if I understand all the previous thread posts. And that's the problem for classics, If your want to run AOS 1.2-3.1 or any classic apps/games they would need to be recompiled. And that goes for kickstart too. Also possible to start with binaries, pick through finding the problem code and then rewrite it. But you can see how deep the rabbit hole starts to go. Hopes were CF could be a drop in CPU replacement supporting classic hw/kickstart and apps, but t quickly revealed itself as code writing behemoth. Might as well start from scratch. I personally felt back in the day that Minimig was the end of the CF story for classics. Plaz

@jingof

Thanks. I know now to wear chainmail and armor before I decide to log on!

Quote:

jingof wrote: @cod3r, Don't get overburdened by the squeaky wheels around here. The majority of us welcome you and your efforts and hope your intentions to contribute to this community materialize.

@Nameless

You're right, emulation was my solution. I only pass through certain native instructions after parsing them.

Just a note: stay cool and refrain from name calling.

Cod3r: asking you for a proof is not "attacking you". We have seen many, many similar posts, from both "recognized companies" and new forum members. Many people have been taken for a ride, some parted with money and got short stick in that deal.

You, being new here, have to understand the dynamic of a *miga community. This community is fragmented, at war with itself, you can have 3 people holding 5 positions that exclude each other. We have arrogant loudmouths and polite backstabbers, we have people who don't care anymore and people who could make King of Vatican jealous of religious zeal.

Cut down on sarcasm, it's not helping here. People are asking valid questions, answer them as best as you can and move on. It's not a witch hunt, it's a proper response from a community that has been burned many times by its "enthusiasm".

I welcome anyone who can contribute some work to AROS source-code even if I'm not going to get CF board.

Once again: we're all adults here. Talk. "Oh yeah?!" is not a discussion starter.

// moderator hat off

_________________
OpenWindows Initiative. Port PS3 hardware to bananas. For free. Join today and receive expired $50 cupon from AI!

@KimmoK

I'm am learning about Morph OS and Amiga OS 4.x more each day. And I was unaware than any Amiga-like OS had multicore CPU support. I am curious about its implementation...

@OlafS25

Quote:

just to clarify things... AmigaOS is closed source commercial product. So you would not have any access to sourcecode (of the OS).

You don’t need a NDA to write devices, I write the Cat weasel driver whit out one and source code is available, and there are open source code driver of sana2 network drivers as well, the SDK is freely available for MorphOS and AmigaOS4.1, its just low level stuff or if wont to contribute to the OS then you need a NDA.

You might need a NDA to write kickstart modules, as there is no documentation on that, as I have seen, so loading drivers at boot can be a problem.

AmigaOS3.x 68k is common dominator but does not have all new API’s provided in AmigaOS4.1 and MorphOS, some examples are DOS functions for reading file size, and seeking beyond 2 or 4gb file offsets.

Quote:

And it is a very small community (between 500-700 users worldwide). But you can of course decide what you are interested more.

AROS has only 5 to 7 users :-b

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@Cod3r

Quote:

@KimmoK I'm am learning about Morph OS and Amiga OS 4.x more each day. And I was unaware than any Amiga-like OS had multicore CPU support. I am curious about its implementation...

It was an announced feature that has never been shown publicly as of yet.

_________________
Support AROS sponsor a developer.

"AROS is prolly illegal ~ Evert Carton" intentionally quoted out of context for dramatic effect

@OlafS25

Thanks again. I'm about to leave for work now. Just trying to figure out the best way to ask my boss about my request. I'll hopefully have a positive update later today regarding my work on the board. Wish me luck!

@Cod3r

It sounds very reasonable. Welcome to the community.[/quote]

@Cod3r

Multicore support scheduled for AmigaOS4.2 and it well known, I guess AROS developers are trying to get there before, we will see, as closed source project like AmigaOS4.x and MorphOS are not so open about that they are working on and how long it’s going to take.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@NutsAboutAmiga

"AROS has only 5 to 7 users :-b"

I could start to make some comments but better not

You are "Nuts" about AOS and I am the same about Aros. But he will certainly decide for himself what is more interesting for him.