Welcome, Guest. Please login or register.

Author Topic: Request about OS Development  (Read 25655 times)

Description:

0 Members and 1 Guest are viewing this topic.

Offline ferrellsl

  • DavidseltyAJ
  • Hero Member
  • *****
  • Join Date: Jan 2005
  • Posts: 868
  • Country: ph
  • Thanked: 2 times
  • Gender: Male
    • Show only replies by ferrellsl
Re: Request about OS Development
« Reply #14 on: November 21, 2017, 04:22:01 PM »
Quote from: bloodline;833404
AROS is "AmigaOS" on x86_64, and it is a really fun project, I have learnt so much from it and a really recommend trying it out!

Thomas is correct though, from an architectural pint of view the design goals of AmigaOS are far removed from what is needed in an operating system in 2017.

As for adding an inbuilt 68k emulator to a x86_64, I tried it, and hit two problems: endianess (which was quite easy, but very time consuming to solve) and address space size. Unfortunately as soon as the OS and the app operate in different size address spaces they can no longer share data structures, which is a a key architectural design feature of AmigaOS (other operating systems are far more strict about control of the address space).

You can read my issues here: http://www.amiga.org/forums/showthread.php?t=72059

Who really cares about an inbuilt, seamless emulator for 68K apps on x86_64 AROS?  Are the folks who use x86_64 AROS really using 68K software?  That's a rhetorical question and the answer is no.  They're using AROS native apps.  So what's wrong with using the non-seamless UAE on AROS if they really, really want to run 68K apps on AROS?  That's what x86_64 users have been doing on Linux, MacOS and Windows for years and it works great.  So the argument that we shouldn't adopt the x86 architecture in order to retain 68K legacy compatibility is ridiculous.  You have to let go of the past to move forward.  Microsoft did....you can't run DOS apps on the current Windows nor Win16 apps without emulation or virtual machines.  Same goes for Apple.  You can't run classic OS9 or PPC apps on current Macs....both parties now rely on emulators or virtual machines for legacy support....DOSBox, VMWare, UAE, VirtualBox, etc...This obsessing over legacy compatibility has led to a dead end street.
« Last Edit: November 21, 2017, 04:28:52 PM by ferrellsl »
 

Offline ferrellsl

  • DavidseltyAJ
  • Hero Member
  • *****
  • Join Date: Jan 2005
  • Posts: 868
  • Country: ph
  • Thanked: 2 times
  • Gender: Male
    • Show only replies by ferrellsl
Re: Request about OS Development
« Reply #15 on: November 21, 2017, 04:34:43 PM »
Quote from: Thomas Richter;833417
Well, you kow how to become a millionaire? Be a billionaire, and inverst the billions into Amiga development - it's really so easy. Apple had money, funding, and a paying customer base. Now look. See the difference?

You're definitely not a programmer who I'd want on my team if you believe that money is the solution or even a substitute for real analysis and engineering.

Funny how Linux is free and it's been ported to every architecture under the sun.....best thing that could happen for OS3 and OS4 is that they be open sourced, but that won't happen....too much greed going on with what's left of the Amiga legacy.
« Last Edit: November 21, 2017, 05:17:03 PM by ferrellsl »
 

guest11527

  • Guest
Re: Request about OS Development
« Reply #16 on: November 21, 2017, 04:38:06 PM »
Quote from: ferrellsl;833419
You're definitely not a programmer who I'd want on my team if you believe that money is the solution or even a substitute for real analysis and engineering.
But it can be a pretty good incentive to reach out for engineers, and let them do what your product needs.

Quote from: ferrellsl;833419
Funny how Linux is free and it's been ported to every architecture under the sun.....
Then, if Linux suits your needs, why complain about AmigaOs? Don't worry, I'm using linux,and I'm also contributing a bit, so don't worry.

But if you want the answer on the question how Apple could manage that, then this answer is very simple: Money. You don't need to like it, but that's how it worked.
 

Offline psxphill

Re: Request about OS Development
« Reply #17 on: November 21, 2017, 05:33:42 PM »
Quote from: LoadWB;833398
Eh, PPC doesn't run 680x0 natively, does it?  Doesn't MorphOS and AmigaOS4 have an emulation layer for this?  Why can't AROS?


You can get big endian PowerPC chips, you can't get big endian x64 chips.

It's possible to create an x64 compiler which deals with the endian conversion (bonus points if you can figure out how to make it only do the endian conversion for data that really matters).

This is how amithlon added support for native code. Recreating it with the ability to use Amiga OS 3.1 and the 68k PCI libraries to use network/graphics cards etc would be a start. Then duplicate the way of switching from 68k to x86 from amithlon & you can start replacing 68k libraries with x64 versions (potentially from aros).
 

Offline polyp2000

  • Full Member
  • ***
  • Join Date: Jan 2011
  • Posts: 244
    • Show only replies by polyp2000
    • https://soundcloud.com/polyp/sets/polyp-2013
Re: Request about OS Development
« Reply #18 on: November 21, 2017, 06:11:30 PM »
Quote from: LoadWB;833398
Eh, PPC doesn't run 680x0 natively, does it?  Doesn't MorphOS and AmigaOS4 have an emulation layer for this?  Why can't AROS?


Isnt this what JANUS is for on Aros ?

guest11527

  • Guest
Re: Request about OS Development
« Reply #19 on: November 21, 2017, 06:17:43 PM »
Quote from: ferrellsl;833418
 You have to let go of the past to move forward.  Microsoft did....you can't run DOS apps on the current Windows nor Win16 apps without emulation or virtual machines.  Same goes for Apple.
So did most users. They moved forward, ditched AmigaOs and got a serious system. So what exactly is "forward" for you?

Somehow, you seem to believe that we "only" need to port AmigaOs to x86, and everything else will come by itself. That, however, is a big misconception. This ship sailed away a long time ago. Approximately 20 years ago.

Quote from: ferrellsl;833418
This obsessing over legacy compatibility has led to a dead end street.
No, you just haven't understood what this market is about. If you are serious about breaking compatibility, then many users have already made the step and bought a new machine. PC, Apple,  you name it. So what exactly is it that keeps people interested in Amiga, if not old applications?

It's called "retro computing", and it is a completely different market. A market, where legacy compatiblity is of utmost importance.
 

Offline olsen

Re: Request about OS Development
« Reply #20 on: November 21, 2017, 06:53:02 PM »
Quote from: ferrellsl;833419
You're definitely not a programmer who I'd want on my team if you believe that money is the solution or even a substitute for real analysis and engineering.
In my experience, money can help shrinking a fair lot of those tricky bits. You are going to pay for that in the long run, though.

Quote
Funny how Linux is free and it's been ported to every architecture under the sun.....best thing that could happen for OS3 and OS4 is that they be open sourced, but that won't happen....too much greed going on with what's left of the Amiga legacy.

Aside comparing two very different pieces of work, created through very different processes, what always strikes me as odd is that the seeming simplicity of AmigaOS and its hardware platform suggests to many critics that everything about the troubles operating system and platform went through might have been caused by simple events which should be simple to sort out. Such as singling out greed as a significant contributor to the troubles.

If only it were greed, or shortsightedness, management failure, absense of a technically inclined founder, or the initial technical edge slipping away because Commodore's controlling shareholders didn't feel like spending any money at all whatsoever on anything, or just about anything else you could reasonably expect to get your head around. Something, anything that could be measured in humanly understandable terms as a handful of events conspiring against the Amiga.

There is no single cause of failure, or a small set of causes which led to failure. It's a frustratingly large number of events battering the Amiga, death by a thousand cuts, stretched over 9-10 years, followed by more grief and tragedy over the next ten years, etc.

Strangely, the Amiga was tested by tragedy and somehow did not disintegrate. To me that's much more interesting than to look for a convenient choice of possible causes, none of which allow us to learn from the past and find a better way for the Amiga to continue to stick around.
« Last Edit: November 21, 2017, 06:57:58 PM by olsen »
 

Offline LoadWB

  • Hero Member
  • *****
  • Join Date: Jul 2006
  • Posts: 2901
  • Country: 00
    • Show only replies by LoadWB
Re: Request about OS Development
« Reply #21 on: November 21, 2017, 07:49:21 PM »
Quote from: Thomas Richter;833417
Well, you kow how to become a millionaire? Be a billionaire, and inverst the billions into Amiga development - it's really so easy. Apple had money, funding, and a paying customer base. Now look. See the difference?

I already conceded to your point, and that's not what I asked in the first place.  I realize the flights of fancy surrounding the imminent resurgence, rebirth, and market domination of Amiga can get people overly sensitive to the question, but that was not my intention and normally I wouldn't even go through this amount of explanation were it not for the level of respect I have for those still developing on Amiga.

I simply asked if anyone had done an analysis of what Apple did and realistically applied it to what could or would need to be done with AmigaOS.  Apple really hasn't kept things secret since the move was done.  Seriously.  People analyze stupid shyt all the time and make hour-long YouTube videos and lengthy tech blog posts, so it would stand to reason that someone might have done such an analysis, if only to put it to bed forever.

It's easy to dismiss the whole idea as "not enough money, not enough people, not enough time," but the technical aspect alone I would imagine would intrigue someone.  While it does intrigue me, I lack the necessary experience and knowledge to break it down the way it deserves.  I know there are a number of lengthy, in-depth write-ups of the Apple move to Intel, and while I understand while reading much of what gets put out there I would be lying to the universe if I said I had the necessary skill-set to do a proper analysis.

EDIT: I also realize we would be looking at a three-fold move: first from the old OS9 way of things on PPC including running 68k, the move to BSD-based OSX with OS9 compatibility, then the move to Intel.
« Last Edit: November 21, 2017, 07:52:06 PM by LoadWB »
 

Offline ferrellsl

  • DavidseltyAJ
  • Hero Member
  • *****
  • Join Date: Jan 2005
  • Posts: 868
  • Country: ph
  • Thanked: 2 times
  • Gender: Male
    • Show only replies by ferrellsl
Re: Request about OS Development
« Reply #22 on: November 21, 2017, 08:14:31 PM »
Quote from: olsen;833426
In my experience, money can help shrinking a fair lot of those tricky bits. You are going to pay for that in the long run, though.



Aside comparing two very different pieces of work, created through very different processes, what always strikes me as odd is that the seeming simplicity of AmigaOS and its hardware platform suggests to many critics that everything about the troubles operating system and platform went through might have been caused by simple events which should be simple to sort out. Such as singling out greed as a significant contributor to the troubles.

If only it were greed, or shortsightedness, management failure, absense of a technically inclined founder, or the initial technical edge slipping away because Commodore's controlling shareholders didn't feel like spending any money at all whatsoever on anything, or just about anything else you could reasonably expect to get your head around. Something, anything that could be measured in humanly understandable terms as a handful of events conspiring against the Amiga.

There is no single cause of failure, or a small set of causes which led to failure. It's a frustratingly large number of events battering the Amiga, death by a thousand cuts, stretched over 9-10 years, followed by more grief and tragedy over the next ten years, etc.

Strangely, the Amiga was tested by tragedy and somehow did not disintegrate. To me that's much more interesting than to look for a convenient choice of possible causes, none of which allow us to learn from the past and find a better way for the Amiga to continue to stick around.



Interestingly enough, AROS, both the x86_64 and 68K versions, were also written for free.  So again, if anyone thinks money is the cure for the Amiga's ills, they're sorely mistaken.

Offline Fats

  • Hero Member
  • *****
  • Join Date: Mar 2002
  • Posts: 672
    • Show only replies by Fats
Re: Request about OS Development
« Reply #23 on: November 21, 2017, 08:42:37 PM »
Quote from: LoadWB;833414
Obviously there's a difference in man-power and funding, but has anyone done an analysis of how Apple moved MacOS from PPC to Intel and the application of those ideas to AmigaOS?


If I remember correctly they used fat binaries; e.g. the programs just included a PPC and a x86 version in the same file. I'm not sure much Amiga minded guys get warm for such bloat.
Trust me...                                              I know what I\'m doing
 

guest11527

  • Guest
Re: Request about OS Development
« Reply #24 on: November 21, 2017, 09:31:46 PM »
Quote from: ferrellsl;833432
Interestingly enough, AROS, both the x86_64 and 68K versions, were also written for free.  So again, if anyone thinks money is the cure for the Amiga's ills, they're sorely mistaken.

And how many of the legacy amiga applications does it run without problems, with which performance, and what is the user count on this?

Again, nothing against AROS, but you're comparing apples and oranges. AROS is there to make AROS programmers happy. AmigaOs is there to make retro-users happy.

Different goals, different rules.
 

Offline Hans_

Re: Request about OS Development
« Reply #25 on: November 21, 2017, 10:16:50 PM »
Quote from: ferrellsl;833432
Interestingly enough, AROS, both the x86_64 and 68K versions, were also written for free.  So again, if anyone thinks money is the cure for the Amiga's ills, they're sorely mistaken.


Sure, money alone solves nothing. However, imagine what would happen if everyone with AmigaOS development expertise could work on it full-time.** Progress would be a lot faster. What would it take to make that happen? Yep, that's right, they'd need to be paid so they can work full-time & still be able to buy food and put a roof over their head.

Hans

** Of course, I'm talking about focused development with well thought out priorities. Haphazard development rarely gets things moving in the right direction.
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
 

Offline kreciu

Re: Request about OS Development
« Reply #26 on: November 22, 2017, 12:09:28 AM »
Quote from: Hans_;833437
Sure, money alone solves nothing. However, imagine what would happen if everyone with AmigaOS development expertise could work on it full-time.** Progress would be a lot faster. What would it take to make that happen? Yep, that's right, they'd need to be paid so they can work full-time & still be able to buy food and put a roof over their head.

Hans

** Of course, I'm talking about focused development with well thought out priorities. Haphazard development rarely gets things moving in the right direction.


... and this will never happen, since AmigaOS is nothing anybody (with exception of hobbyist) cares about and wanted to pay money for. AmigaOS is nothing to stand in line overnight to get, there is no feature that makes it better then popular OSes.

Financial support can only be provided by Amiga hobby st/fanatics ;), that are willing to pay for greatly limited OS and hardware. End of story.
Re-A1200inE/BOX/3.2/AmigaOS3.2/TF1260@66Mhz/256Mb/MediatorTX/R9200SE/SpiderUSB/LAN/SB128/16Gb-CF/DVD-ROM/FDD-HD
 

Offline kreciu

Re: Request about OS Development
« Reply #27 on: November 22, 2017, 12:10:53 AM »
Quote from: Hans_;833437
Sure, money alone solves nothing. However, imagine what would happen if everyone with AmigaOS development expertise could work on it full-time.** Progress would be a lot faster. What would it take to make that happen? Yep, that's right, they'd need to be paid so they can work full-time & still be able to buy food and put a roof over their head.

Hans

** Of course, I'm talking about focused development with well thought out priorities. Haphazard development rarely gets things moving in the right direction.


... and this will never happen, since AmigaOS is nothing anybody (with exception of hobbyist) cares about and wanted to pay money for. AmigaOS is nothing to stand in line overnight to get, there is no feature that makes it better/wanted then popular OSes. Even non-user friendly Linux gives more options and software that people could use.

Financial support can only be provided by Amiga hobbyist/fanatics ;), that are willing to pay for greatly limited OS and hardware. End of story.
Re-A1200inE/BOX/3.2/AmigaOS3.2/TF1260@66Mhz/256Mb/MediatorTX/R9200SE/SpiderUSB/LAN/SB128/16Gb-CF/DVD-ROM/FDD-HD
 

Offline ferrellsl

  • DavidseltyAJ
  • Hero Member
  • *****
  • Join Date: Jan 2005
  • Posts: 868
  • Country: ph
  • Thanked: 2 times
  • Gender: Male
    • Show only replies by ferrellsl
Re: Request about OS Development
« Reply #28 on: November 22, 2017, 12:32:29 AM »
Quote from: Thomas Richter;833436
And how many of the legacy amiga applications does it run without problems, with which performance, and what is the user count on this?

Again, nothing against AROS, but you're comparing apples and oranges. AROS is there to make AROS programmers happy. AmigaOs is there to make retro-users happy.

Different goals, different rules.



They run just fine using Janus UAE for AROS:   https://sourceforge.net/projects/janus-uae/

or E-UAE:  http://aros.sourceforge.net/documentation/users/applications/euae.php

And for seamless emulation on AROS, the best emulator is AmiBridge which is part of the IcarosDesktop distro:  http://vmwaros.blogspot.com/p/amibridge.html

So no, I'm not comparing apples to oranges.  Someone asked why the Amiga hadn't moved away from a dead CPU architecture and the response was "We need a sh@t load of money to make that happen" and "We need to maintain 100% legacy compatibility".  Those are both fallacies perpetuated by folks who still want to monetize a dead OS called OS3.x and OS4 isn't far behind it.

Offline ferrellsl

  • DavidseltyAJ
  • Hero Member
  • *****
  • Join Date: Jan 2005
  • Posts: 868
  • Country: ph
  • Thanked: 2 times
  • Gender: Male
    • Show only replies by ferrellsl
Re: Request about OS Development
« Reply #29 from previous page: November 22, 2017, 12:37:20 AM »
Quote from: Hans_;833437
Sure, money alone solves nothing. However, imagine what would happen if everyone with AmigaOS development expertise could work on it full-time.** Progress would be a lot faster. What would it take to make that happen? Yep, that's right, they'd need to be paid so they can work full-time & still be able to buy food and put a roof over their head.

Hans

** Of course, I'm talking about focused development with well thought out priorities. Haphazard development rarely gets things moving in the right direction.


That's pretty funny because people pay a LOT of $$ for OS4 and NG hardware yet both are  still stuck in 2003.  OS4 has rough feature parity with Windows 98 and I have cell phones that outperform NG PPC hardware.  So yep, that's right, charging boutique prices for OS4 and its hardware hasn't panned out like you suggest.

As for trying to make a living coding for dead platforms one should lower their financial expectations or look for another line of work.  FYI, no one is making a living coding for the C64 or the Spectrum ZX-81 anymore either.
« Last Edit: November 22, 2017, 12:50:39 AM by ferrellsl »