Overview Features Instructions Performance Forum Downloads Products OrderV4 Reseller Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases
Information about the Apollo CPU and FPU.

Are We Just Nostalgic Or Should the Amiga Advance?page  1 2 3 4 5 6 7 8 9 10 11 12 13 

Anthony Jacques

Posts 9
05 Jul 2019 14:27


Gunnar von Boehn wrote:

  On AMIGA OS if 2 programs "talk to each other" they exchange PTRs and each program knows about the other. And to send a message the program just pokes to value into the other program.
  This is how AMIGA works.
  Direct, easy simple.

Perhaps look at how one of your neighbours evolved. In MiNT on the Atari, when memory protection was added (back in early 90s? or earlier?) they solved these issues.

MiNT added mxalloc() which allows processes to set the memory protection flags on block of RAM when allocating it. So when two processes want to communicate with each other, they can grant the block of memory as readable/writable to other processes, while still keeping the rest of their memory as "not accessible". So they can still exchange pointers and be very lightweight compared to pipes etc (which MiNT also provides) - they simply need to be aware of what memory is to be shared, and what isn't.

Now, the difference here is in the Atari world, this was thought about while software was still being developed, and so plenty of programs were updated to use Mxalloc appropriately. For those that haven't been updated, it does allow you to set default values for all other mallocs, so that you can for example say that "all memory owned by this process is readable but not writable", which gets you a big step towards stability even if it isn't all the way.

I do agree however that doing it now is probably too late for the Amiga world, but perhaps it is worth considering the "two-tier" approach of at least newly built+compiled (eg open source!) apps are secure even if legacy ones aren't.


Gunnar von Boehn
(Apollo Team Member)
Posts 4890
05 Jul 2019 15:36


Thanks for your comment, but how much does this improve about what we have now?

I think the idea on AMIGA OS is not to stop bad applications but to find them and not to use them.

 
 
Let us review what we have.
We have 4096 MB address space.
We use after boot up maybe 8 MB of this.
This means 99.8% of our space is unused or free memory.
These 99.8% of space (all unused or free memory) we can monitor on AMIGA OS for malicious access.
 
This means we have a real high probability to find buggy applications already.
 
The strategy in AMIGA OS is find "broken" aps and remove them.
And then to only run healthy programs.



Kamelito Loveless

Posts 115
05 Jul 2019 16:06


What would be nice is to have Aros Exec being pure hand optimized 68k based on the C code output maybe, also the gfx lib I guess.


Nsklaus -

Posts 63
05 Jul 2019 21:14


Gunnar von Boehn wrote:

  The strategy in AMIGA OS is find "broken" aps and remove them.
  And then to only run healthy programs.

there's no bug-free software. 99% of software contains bugs.
even languages themselves or compilers have bugs. even hardware have bugs.
having to manualy filter programs to choose the less bug-ridden ones to run, is a real lame situation (it cannot be called a strategy, more like a workaround really).

some software bugs don't manifest themselves immediately, in some cases it's a slow degrading process leading to seemingly random crashes while running software that were considered healthy.

this leads to what ? the user having to debug and pinpoint the cause of errors. this is putting on the user the job that should be done by the system.

one shouldn't think it is something to be proud of, it just shows the os is unfinished.

back in the days, i remember we used to make fun of windows for being a crash machine, as windows was crashing more often than amigaos. now consider the situation: most everyone else is crash-free but amigaos. even atari sorted this problem out.

amigaos didn't. all this for what reason ? keeping backward compatibility.
compatibility with what exactly? very old software.

can't you see the circle ? it's plain logic.
no need to explain how amigaos works, it's not the point.
no need to tell me some devs do fall in love with the way it works,
it's a false answer. you're avoiding talking about the end result.
which is bad.

amigaos won't gain new users or new software because it appeals noone to use amigaos for all of its limitations. so you want to keep running at least the old software but in doing so the OS stays unappealing to others. ergo it's a never ending sinking situation.

amigans should realize this and wish for change in the future, but no.
stating the obvious makes those who do, the bad guys instead.
just wanting to discuss about it is considered a sin.
and you wonder why the community is only shrinking and never growing.
i'd suggest, start talking, brainstorming, to see how to make an _optional_ custom kernel, and after that, eventually start recompiling all the apps we can to benefit the new features. but ohh evil me for suggesting that crazy idea. amigaos have no bugs, and if you run healthy software all is fine, right ?
 
well no.




Kamelito Loveless

Posts 115
05 Jul 2019 21:23


The Apollo team is not developing AmigaOS. If you want a feature on it you better ask Thomas Richter or Olaf Barthel. As for a specific Aros version supporting what you ask, you know that Aros is open source.


Nsklaus -

Posts 63
05 Jul 2019 21:36


@kamelito
  you're mistaken, i'm not saying apollo team should do it.
  i'm not even saying it should be done on the real amigaos.
  i'm not even 'asking' anything.
  what i'm doing is advocating for the community to realize the idea.
  those with the know-how and the will to do so can do it,
  on their own OS (be it aros, morphos or other clones), on their own schedule at their pace, their own way.
  i'm just defending the idea. i could be defending this elsewhere, but i talk about it, here, because well there's still a bit of life here, and lots of people read this forum.
  maybe amigaos itself can be patched to produce a custom kernel, i don't know and i don't care much because of the legal status the real amigaos is trapped in, but what i really think about is aros.
  and as for me doing it, because aros is opensource, well i already said it's beyond my level, i'm just a crappy dev, just doing small contributions here and there and struggling, enjoying re-developing retro games from scratch. but that's another subject.
  it's not because someone is not a super skilled dev that one cannot discuss ideas, wouldn't you agree? and again i'm not demanding or asking anything. i just discuss the idea about why amiga ecosystem is where it is today and what could be done about it, in my humble opinion.


Kamelito Loveless

Posts 115
05 Jul 2019 22:10


Watch this
  EXTERNAL LINK


Steve Ferrell

Posts 397
05 Jul 2019 22:41


nsklaus - wrote:

  @kamelito
    you're mistaken, i'm not saying apollo team should do it.
    i'm not even saying it should be done on the real amigaos.
    i'm not even 'asking' anything.
    what i'm doing is advocating for the community to realize the idea.
    those with the know-how and the will to do so can do it,
    on their own OS (be it aros, morphos or other clones), on their own schedule at their pace, their own way.
    i'm just defending the idea. i could be defending this elsewhere, but i talk about it, here, because well there's still a bit of life here, and lots of people read this forum.
    maybe amigaos itself can be patched to produce a custom kernel, i don't know and i don't care much because of the legal status the real amigaos is trapped in, but what i really think about is aros.
    and as for me doing it, because aros is opensource, well i already said it's beyond my level, i'm just a crappy dev, just doing small contributions here and there and struggling, enjoying re-developing retro games from scratch. but that's another subject.
    it's not because someone is not a super skilled dev that one cannot discuss ideas, wouldn't you agree? and again i'm not demanding or asking anything. i just discuss the idea about why amiga ecosystem is where it is today and what could be done about it, in my humble opinion.
 

 
Of course the things you've spoken of can be done, but to what end?  It isn't like any version of AmigaOS will ever become mainstream again to compete with MacOS, Windows or Linux.  Even if a new or enhanced AmigaOS gained process separation, memory protection, virtual memory and multi-user capabilities, there would be no software library for this new OS (just legacy apps sandboxed or ran with UAE).  So who is going to write these new apps to take advantage of the new features of this enhanced OS?  You?  Me? Who?  And if you expect to simply rely on the classic library of apps, then what's the point of moving to a new OS in the first place?
 
With no consumer demand for this new OS, there would be no consumer demand for any apps to drive and fund further OS or application development either.  I make my living by writing apps to sell.  I don't have time to develop apps for free, nor do most developers.  It took the resources of a multi-billion dollar company and an army of developers and testers and several years of effort to bring those features to Windows. It took an army of open source developers (mostly unpaid university students) years of coding and testing to bring those features to MacOS/BSD/Darwin and Linux.  So who do you expect to conduct the development and testing of this new or enhanced AmigaOS?  Certainly you don't expect AmigaOS to rise from the ashes and compete with established mainstream operating systems, because that's sheer lunacy....so again, what would be the point of bringing mainstream features to a dead OS?  Simply to make you happy?  Who is going to fund the development and testing?  And what is the design goal?  Do you even have one?  Sounds like you want it done just because it your mind you think it's cool....That's not enough justification.
 


Nsklaus -

Posts 63
05 Jul 2019 23:17


@kamelito

- i'm not asking _why_ it is crashy. that is beside the point i'm talking about here.
- but rather, i'm talking about the end result: it is crashy.
- i'm saying this is bad.
- i'm saying that, if amiga-ish OSes wants to appeal new users they need remedy this.

if software compatibility need to be broken to achieve stability then we should do it.
some may think that changing the way it works a little here and there will make it so it is not 'real amiga' anymore. well, it's either that or continued slow death.

how long do you suppose having fun running the very same old, outdated software with ever shrinking community will still outweight the necessary reform the amiga model needs to undergo?

it's been more than 20 years already.
don't you think something else should be tried yet ?
or you are content with the situation maybe ? you think amiga-like OSes could appeal to anyone with all their limitations based on an unfinished design that is probably even older (30+years old) ?


Nsklaus -

Posts 63
05 Jul 2019 23:49


Steve Ferrell wrote:

 
  Of course the things you've spoken of can be done, but to what end?  It isn't like any version of AmigaOS will ever become mainstream again to compete with MacOS, Windows or Linux. 
 

 
  i think the idea to compete on the market share with linux, macos or windows is a dream. it won't happen, not for any foreseable future.
 
  but, does that prevent all the other alternative OSes to keep on going ? no. they continue making their product, better. adding features, updating, it's part of development process.
 
  those who don't are considered dead. users leaving. etc.
 
 
Steve Ferrell wrote:

  Even if a new or enhanced AmigaOS gained process separation, memory protection, virtual memory and multi-user capabilities, there would be no software library for this new OS (just legacy apps sandboxed or ran with UAE).  So who is going to write these new apps to take advantage of the new features of this enhanced OS?  You?  Me? Who?
 

 
  "if you build it they will come", as the saying goes.
  suppose someone take the risk of making such aa modified amiga-ish OS. it's a risk alright, there's no guaranty it will be adopted.
  but if you don't take it, then you have 100% chance nothing indeed will happen.
  now for the optimist view: there's a number of software on amiga ecosystem for which the sources are available. those could be the first apps to be re-compiled to work on such a modified amiga-ish OS. the idea i'm talking about is: retaining mostly, say 80%, of being source compatible. so a recompile of said soft should be fairly easy.
 
 
Steve Ferrell wrote:

  With no consumer demand for this new OS, there would be no consumer demand for any apps to drive and fund further OS or application development either.  I make my living by writing apps to sell.  I don't have time to develop apps for free, nor do most developers.
 

 
  face it, there is no demand for amigaos either.
  most amiga dev do it because they like the platform.
  very few (if any) devs live from development they do for amiga ecosystem.
 
 
Steve Ferrell wrote:

  It took the resources of a multi-billion dollar company and an army of developers and testers and several years of effort to bring those features to Windows.
  It took an army of open source developers (mostly unpaid university students) years of coding and testing to bring those features to MacOS/BSD/Darwin and Linux.  So who do you expect to conduct the development and testing of this new or enhanced AmigaOS? 
 

 
  think aros.
  a handful of talented dev did it for fun on their free time, little by little.
  no army or multi-billions dollars needed.
  same for linux in its early days.
  there's a lot of example of alternative OSes without multi-billions or army of coders.
 
 
Steve Ferrell wrote:

  Certainly you don't expect AmigaOS to rise from the ashes and compete with established mainstream operating systems, because that's sheer lunacy....so again, what would be the point of bringing mainstream features to a dead OS? 
 

 
  as you've seen in my previous replies in this post,
  you do understand i have no delusion of seeing amigaos rising from ashes and competing with mainstream OSes. the point is just improving the situation. like atari did for itself for example. same case, same goal. it would be good to have amiga-like OSes to become stable and trust worthy.
 
 
Steve Ferrell wrote:

  Simply to make you happy?  Who is going to fund the development and testing?  And what is the design goal?  Do you even have one?  Sounds like you want it done just because it your mind you think it's cool....That's not enough justification.
 

 
  it's not for me, i don't even runs any amiga-like OS and i've sold all my amiga gears around 2002. i'm just saying it would be a good time for amiga model to try something else since so far the route taken haven't paid off, at all. i'm saying for amiga itself, not me.
  but i must say, it's also true i'd be tempted to run something amiga-like again, if it modernize a little. but make no mistake, i've been running linux since even before i've sold my amigas, since 2000. i'm happy with linux. i just have good memories with amiga, i've tried all the NGs too, i've tried aros, and also emulation through various means, and,
  i've been dismayed with the poor stability, and various other strong limtations. i just thought 'what a shame. this little OS it's beautiful, responsive, small, it could do wonders but it's crippled.
 
 
 


Steve Ferrell

Posts 397
05 Jul 2019 23:55


nsklaus - wrote:

  @kamelito
 
  - i'm not asking _why_ it is crashy. that is beside the point i'm talking about here.
  - but rather, i'm talking about the end result: it is crashy.
  - i'm saying this is bad.
  - i'm saying that, if amiga-ish OSes wants to appeal new users they need remedy this.
 
  if software compatibility need to be broken to achieve stability then we should do it.
  some may think that changing the way it works a little here and there will make it so it is not 'real amiga' anymore. well, it's either that or continued slow death.
 
  how long do you suppose having fun running the very same old, outdated software with ever shrinking community will still outweight the necessary reform the amiga model needs to undergo?
 
  it's been more than 20 years already.
  don't you think something else should be tried yet ?
  or you are content with the situation maybe ? you think amiga-like OSes could appeal to anyone with all their limitations based on an unfinished design that is probably even older (30+years old) ?
 

 
Who is this "we" that you keep referring to?  Who will pay the developers whose skill and time would be required to do what you're asking? Or do you expect developers to drop their current  projects that pay their bills and take up AmigaOS development without salary and sign up for government assistance in order to feed their families and pay their bills?
 
As for the Amiga's slow death, where have been hiding?  Have you been living under a rock or in a cave?  The Amiga died years ago and it will remain dead.  AmigaOS is a DEAD OS for hobbyists.  IT WILL NEVER BE A MAINSTREAM OS AGAIN.   
 
So I will ask again.  What is the point of updating AmigaOS with modern features?  To compete with Windows or other mainstream operating systems?  I keep asking you very direct questions with easy answers but you're either unable or unwilling to answer them.
 
Consumers don't want nor need AmigaOS, so even if we had a modern AmigaOS available for sale today, do you expect consumers and companies to throw out their existing libraries of Windows/Mac/Linux software, toss out their existing hardware, and wait for this "we" that you keep referencing to write new, enhanced AmigaOS versions of their old apps?
 
So again, what is the design goal?  I don't think you have one.  You're one of those guys who is living in the past and obsessed with the fact that the Amiga failed and you want it to become mainstream again...at least that's my opinion based on the answers you fail to give to the questions here.
 
 
 


Nsklaus -

Posts 63
06 Jul 2019 00:09


@ferrell
  you didn't read my previous post did you ?
  i never thought about amiga-like OSes becoming mainstream again.
  and as for money, there's no money.
  like when i dev on linux, i do it for free.
  like aros folks they do it for free.
  who in his right mind could think of making a living through amiga dev  or thinking amiga would become number one again?
 
  that's not me, you got me wrong here.
just read my previous post, it was replying to you and those specific points. 


Steve Ferrell

Posts 397
06 Jul 2019 00:10


nsklaus - wrote:

  think aros.
    a handful of talented dev did it for fun on their free time, little by little.
    no army or multi-billions dollars needed.
    same for linux in its early days.
    there's a lot of example of alternative OSes without multi-billions or army of coders
   
   
 

 
You obviously have no idea how long AROS has been in development and it is no more advanced than OS 3.1.  AROS development began in the winter of 1995 and the design goal was to implement an open source AmigaOS, not to add modern features.  So, it's been 24 years and it is still being developed and debugged just to achieve parity with OS 3.1....So when do you expect AROS to have the modern features you keep crowing about? You keep throwing it out there as an example of progress.
 
So instead of showing up here and and just b@itching about the current state of AmigaOS, I suggest you start learning and coding because I assume this "we" you keep referencing includes "you".  And if it doesn't include "you", then I recommend that you just shut up....because continuing to say the same things over and over is just trolling and not contributing.


  i don't even runs any amiga-like OS and i've sold all my amiga gears around 2002.

Wait, you don't even own an Amiga or run AmigaOS and you're here complaining about stability issues and the need to update it?  That's priceless.  You need to go back to your Linux box and stopp trolling this forum.
 


Nsklaus -

Posts 63
06 Jul 2019 00:21


i'm not "bitching".
if that's your idea of a discussion, we're not on the same page.

the "we" is the "we" as the community of people interested in amigaos and its clones and extented ecosystem.

and as for aros, i've been following all those years.
i'm well aware of the current status of aros and its goal.
so far aros followed aos model and so inherited all its flaws.
i'm just saying it is roughly on par with aos3.1 (i emphase on the rough word) so it could be a replacement for aos, in some not so distant future (5 to 10years). the only big strength of aros is that it is opensource, free of legal problems. and it could be easily experimented upon if one so wishes it.




Steve Ferrell

Posts 397
06 Jul 2019 00:29


nsklaus - wrote:

i'm not "bitching".
  if that's your idea of a discussion, we're not on the same page.
 
  the "we" is the "we" as the community of people interested in amigaos and its clones and extented ecosystem.
 
  and as for aros, i've been following all those years.
  i'm well aware of the current status of aros and its goal.
  so far aros followed aos model and so inherited all its flaws.
  i'm just saying it is roughly on par with aos3.1 (i emphase on the rough word) so it could be a replacement for aos, in some not so distant future (5 to 10years). the only big strength of aros is that it is opensource, free of legal problems. and it could be easily experimented upon if one so wishes it.
 
 

Don't worry.  I don't plan to "discuss" anything further with someone who sold all of their Amiga hardware and software in 2002 and insists that "we" need to fix AmigaOS....clearly, "you" aren't even part of the "we" and you're unable and/or unwilling to contribute, so all you do is point out faults.  That is the very definition of trolling.  Have a nice day!



Nsklaus -

Posts 63
06 Jul 2019 00:39


Steve Ferrell wrote:

  Wait, you don't even own an Amiga or run AmigaOS and you're here complaining about stability issues and the need to update it?  That's priceless.
 

 
  correct, i don't even run amigaos nor its clones anymore.
  i don't think it's "priceless", i think it's a bit sad, and it shows what the situation of amigaos model and its limitations does to people who might be slightly interested in it.
 
 
Steve Ferrell wrote:

  all you do is point out faults.
 

 
  correct. i never said otherwise.
  amiga is not worth more from me at the moment. i just share an idea and i feel i'm being generous with explications volume since you people apparently have difficulties to read, talk, process information and make use of plain logic. but since i'm a nice guy, i did try to explain again differently a few times so you could get a chance to grasp the point.
 


Andrew Miller

Posts 229
06 Jul 2019 01:27


@Nsklaus
Just a quick question, are you going to buy any of the apollo products?


Nsklaus -

Posts 63
06 Jul 2019 01:29


that video, viva amiga, with Bryce Nesbitt.
EXTERNAL LINK 
between 1:57 to 2:03 and again from 22:01 to 22:20

the guy tells it just like i've been saying all this time.
memory protection/memory management was indeed planned.
plain loud and clear.
so is he a troll too ?
and you can bet he is no longer running amigaos nor its clone either too.
so was i talking nonsense ? no. why ? because, you can see and hear a dev from the original team telling exactly like me.




Nsklaus -

Posts 63
06 Jul 2019 01:31


@andrew
  i might. i'd be interested in the standalone.
  by the way i did buy a v600 and an amiga 600 it was 3 years ago,
  i've used it less than two weeks, i immediately sold it.
  don't get me wrong i think what this project does is really good, especialy the integrated gfx card in addition to all the rest. but, running old amiga hardware.. all clunky and the software too, the OS limitations, the instability and all those it was like a cold shower, i instantly remembered why i did sell all my amiga stuffs in 2002.
  it was good in 199x, but not so much after 2000 especialy once you tasted stability and reliability in linux, the weaknesses of aos became unbearable.
  i still think the os looks good, it's responsive, small, customisable, easy to understand, logic nice and all until it crashes on you ..and all the weaknesses... it cannot be justified in 2017 let alone 2020.
 
 
 
but even if i don't buy the standalone, that doesn't change a thing.
what i'm saying about memory protection/memory management is true.
in that video, the guy even says: "we knew how to do it, but we couldn't do it before the clock ran out (commodore dying)".
 
and everyone were telling me, it's not true, it cannot be done, this is not the amiga way, blah blah blah. you can see clearly in this video they were about to do it, and they were going in that direction.



Mike Kopack

Posts 268
06 Jul 2019 03:13


Just realize that there were a NUMBER of things that they had planned for Amiga (like AAA, etc.) that ran out for them. And they totally expect that they would break backwards compatibility when doing so.

And that's the problem - you can do something like that if you have a major company backing it (like CBM was) to do the development and push the updated platform, providing developer support to get critical apps ported over in a timely manner, etc.

Today, that's just not a reasonable expectation on an old, largely unsupported, niche OS. We're already fragmented enough between AROS, 3.9, 4.x, 3.1, 3.1.4, etc.

You see how things went with the x1000 and X5000 with a new OS4.x - very little uptake. Most Amiga users these days either do it with old classic hardware (plus accelerators), or they do it via emulation. VERY few use the Amiga for any sort of productivity anymore - it's mostly to play the classic games, maybe do some renderings, etc. In that sort of use case, it's very VERY hard to justify the investment needed to make a new incompatible OS, port apps over to it, and get enough uptake of new hardware + OS support to justify the investment necessary to make it happen.

Sorry, but those days have long since sailed.

The nice thing about the Vampire is that it's acting as a way to bring ALL the various Amiga models up to a much higher performance  than was previously possible on classic systems, giving them all SAGA, while maintaining all the backwards compatibility.

One could argue that the V4SA will be more of an Amiga than the X1000/x5000 ever were given their departure from the previous Amiga OS / architecture. I see it almost like a classic style A5000 that never existed.

But do I see it as a replacement for my modern macbook Pro or linux boxes or Windows gaming machine? HELL no... Those are all WAY WAY WAY more powerful and useful machines given their modern OS'es, modern hardware, storage support, powerful CPUs, and oodles more memory and graphics capabilities, modern software that can do tons more than I ever could with my Amiga.

In it's day, the Amiga was AWESOME. But those days have come and gone. If CBM had survived and brought some of what they had planned to market it could have been a very different mid 90's and 2000's. But alas, coulda shoulda woulda....

posts 244page  1 2 3 4 5 6 7 8 9 10 11 12 13