Overview Features Coding ApolloOS Performance Forum Downloads Products Order Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
Please visit our Apollo-Discord Server for support.



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 

Wawa T

Posts 695
11 Jul 2019 00:31


@Steve Ferrell

im sorry but i really cannot be bothered to answer in every detail right now. you might want to check the current sources and builds  here:
EXTERNAL LINK 
certainly there are things we lack and what needs to be improved and implemented one by one. but for now we already have several advantages over ks3.x out of the box, like:
-huge poseidon usb stack in rom
-cdrom and several filesystems in rom
-built in rtg system, no third party necessary, cgx compatible, p96 driver compatible
-built in easy to configure network stack, driver compatible
-and so on..
its not that bad, really

 


Wawa T

Posts 695
11 Jul 2019 00:32


and now pardon me that i go back to compiling aros test build for vampire;)


Steve Ferrell

Posts 424
11 Jul 2019 00:42


wawa t wrote:

@Steve Ferrell
  now, what concerns your list, i wont check and answer in every detail
 
  cdrom and fat support is in rom, as it should be, to be able to boot cd;). a lot of stuff is in rom, therefore you wont find it on disk.
 
  there are amiga raster fonts on aros, and it can use what available in this area but it cannot contain the original fonts due to licensing/legal reasons afaik
 
 
 
  there is no overscan support for now and i dont know about pcmcia
 
  edit: aros has a handy editor rather than the once that comes with amiga
 
  the datatypes you mention are im classes/datatypes
 
  no specific cpu libs are necessary, just as you write, no mess.
 
  port-handler is in L:
 
  the concept of monitors has been dropped for a reason and also for lack of documentation, thor himself advised against it years ago.
 
  palette prefs are there but not finished.

You're obviously not reading my posts or not comprehending them.  That list is NOT MY list.  The list comes from here, The AROS Platform Wiki:  EXTERNAL LINK 
And you completely gloss over the fact that at a minimum, an AROS 68K user would have to copy the following software over from their AmigaOS disks to their AROS 68K system to achieve feature-parity with AmigaOS 3.x:

Recoverable RAM disk RAD: device
DEVS/PRINTERS/ (missing all printer drivers but postscript.
All system fonts
DEVS/audio.device
crossdosfilesystem
mfm.device
parallel.device
DEVS/DOSDRIVERS/aux
PREFS/overscan
prepcard
TOOLS/COMMODITIES/crossdos
pallete
printergfx
printerps
DEVS/MONITORS/ (missing A2024, dblntsc, dblpal, euro36, euro72, multiscan, ntsc, pal, super72 and vgaonly)
PREFS/overscan
nofastmem

And I left out quite a few others.

The average user has no idea what pieces are missing and would need to be copied over so as soon as AROS chokes on their system, they will simply go back to OS 3.x

So it's been 24 years since AROS was conceived.  When can we expect these issues to be addressed or should we continue using our gray area copies of OS 3.x to fill all the holes?  Like most users it's just easier and more reliable to stick with my copies of AmigaOS rather than wait for AROS to play catch-up.



Steve Ferrell

Posts 424
11 Jul 2019 00:43


wawa t wrote:

and now pardon me that i go back to compiling aros test build for vampire;)

Good, because you have a lot of catching up to do.


Mr Niding

Posts 459
11 Jul 2019 01:07


Steve Ferrell wrote:

wawa t wrote:

  and now pardon me that i go back to compiling aros test build for vampire;)
 

 
  Good, because you have a lot of catching up to do.

You sure know how to inspire.

Wawa isnt pretending AROS is in brilliant state, but its an OS open for development.
While Im not shy about using Coffin/ApolloOS (especially considering the number of AOS versions Ive purchased), I hope and wish Wawa real progress. And Im sure he hopes for more devs to help out.

There is being constructive, and there is being negative.


Steve Ferrell

Posts 424
11 Jul 2019 01:19


Mr Niding wrote:

Steve Ferrell wrote:

 
wawa t wrote:

  and now pardon me that i go back to compiling aros test build for vampire;)
 

 
  Good, because you have a lot of catching up to do.
 

 
  You sure know how to inspire.
 
  Wawa isnt pretending AROS is in brilliant state, but its an OS open for development.
  While Im not shy about using Coffin/ApolloOS (especially considering the number of AOS versions Ive purchased), I hope and wish Wawa real progress. And Im sure he hopes for more devs to help out.
 
  There is being constructive, and there is being negative.

I'm not being negative.  I just being "real" and tired of all the wine and roses being tossed around in regard to AROS 68K.  It is NOT ready for prime time in any way, shape or form, unlike what some folks here would have you believe.  I've actually tested it on real hardware and it is far from a usable state.  I hope that I've saved some folks here the time and trouble that AROS 68K caused me during testing.  A novice user will certainly have trouble with it and they will get fairly annoyed at having to drag out their OS 3.x disks just to print, edit, or use a high-res screen...not to mention all the other missing functionalities.

Or would you rather that testers like myself just remain quiet and let you find out these issues the hard way like I did?

The x86 branches are much more usable and mature, but for now, I couldn't recommend AROS 68K to anyone other than a dev or a tinkerer.

I hope WaWa gets some help too.  AROS 68K shows promise, but even after 24 years it has a LONG way to go.

So if you believe that testing and reporting the results is negative, I won't try to change your mind, but I will certainly NOT sugar-coat test results....they are what they are. 




Renee Cousins
(Apollo Team Member)
Posts 142
11 Jul 2019 02:03


"intellifont/compugraphic font support (TrueType is indeed the way forward, but it is another resource pig on anything but high-end Amigas)/"

This is everything wrong with AROS.


Renee Cousins
(Apollo Team Member)
Posts 142
11 Jul 2019 02:10


wawa t wrote:

Steve Ferrell wrote:

  So again, I ask why would anyone choose AROS 68K over OS 3.x when it still lags behind OS 3.x in several areas?
 

  because its:
  1. free

So is AmigaOS. Came with every Amiga.
wawa t wrote:

  2. open

You say this like it's a positive.
wawa t wrote:

  3. its lacks all the legal burdens of the genuine os, that will never be fully resolved, we have waited a quarter of the century now for this to happen.

Baggage? That's your best argument.
wawa t wrote:

  4. it clean implementation, only limited by functional compatibility of original concepts, not by outdated code. which apparently is very cumbersome an issue with the original sources.
 

Moot.
wawa t wrote:

  5. its made from the start with portability in mind, so if one hardware platform goes out of the window not all is lost. also evolving features of a hardware platform can be supported with less effort.
 

Really? Why do both the M68K and Pi ports suck so hard then? If it's just SOOOO easy to port.
wawa t wrote:

  6. it is compatible with contemporary toolchains, rather than demanding obscure and outdated tools.
 

gcc is obscure?
wawa t wrote:

  7. its just a few things that come to my mind, but isnt that obvious that aros is a platform we could gather around to improve rather than constantly bragging about opening amiga os source itself?

Maybe. We'd have to fork the M68K and start serious stripping all the shite you seem to be getting all wet for.


Wawa T

Posts 695
11 Jul 2019 02:10


Steve Ferrell wrote:

 
  You're obviously not reading my posts or not comprehending them.
 

maybe. as i wrote im not in mood nor capacity to go through walls of text in every detail nor participate in lengthy discussions, there are things to do.


  That list is NOT MY list.  The list comes from here, The AROS Platform Wiki:  EXTERNAL LINK 

admittedly outdated.
 

  And you completely gloss over the fact that at a minimum, an AROS 68K user would have to copy the following software over from their AmigaOS disks to their AROS 68K system to achieve feature-parity with AmigaOS 3.x:

 
  Recoverable RAM disk RAD: device
  DEVS/PRINTERS/ (missing all printer drivers but postscript.
  All system fonts
  DEVS/audio.device
  crossdosfilesystem
  mfm.device
  parallel.device
  DEVS/DOSDRIVERS/aux
  PREFS/overscan
  prepcard
  TOOLS/COMMODITIES/crossdos
  pallete
  printergfx
  printerps
  DEVS/MONITORS/ (missing A2024, dblntsc, dblpal, euro36, euro72, multiscan, ntsc, pal, super72 and vgaonly)
  PREFS/overscan
  nofastmem
 

im an aros user so to say and so far i dindt felt immediate necessity to have these components at hand, not that some of them couldnt or shouldnt be delivered later.

as i said, there wont be monitor files in aros, the concept has been dropped, aros is not a copy of os3.1 in every respect, just for the sake of it. this needs to be understood.


  The average user has no idea what pieces are missing and would need to be copied over so as soon as AROS chokes on their system, they will simply go back to OS 3.x

there might not be an average user, depending on priorities a user might not even realize that these components were missing. at least i had to check whats there and what not, and you obviously were neither fully up to date.
 

  So it's been 24 years since AROS was conceived.  When can we expect these issues to be addressed or should we continue using our gray area copies of OS 3.x to fill all the holes?  Like most users it's just easier and more reliable to stick with my copies of AmigaOS rather than wait for AROS to play catch-up.

its not 24 years since aros68k is being developed. and very few people help while everybody else complains and makes demands. aros is not a closed source commercial product like os4 or morphos, so this behavior is inappropriate. i wont promise you anything. you can certainly use your copy of genuine amiga sytem as long as you wish, im not trying to force aros upon you, simply wanted to correct some facts.


Wawa T

Posts 695
11 Jul 2019 02:28


@Renee Cousins
 
  >So is AmigaOS. Came with every Amiga.
  if you mean the kickstart, then yes it has been included in price. but obviously for many people it isnt enough, because then there wouldnt be any followups or compilation bootlegs like coffin.
 
  >You say this like it's a positive.
  well. its positive in my book, the system can be developed systematicly insted of patching binaries in asm one by one and people can contribute. whats wrong with it?
 
  >Baggage? That's your best argument.
  so you enjoy the legal situation its been under which led to no development or fixes or appropriate distribution being possible for twenty years?
 
  >Really? Why do both the M68K and Pi ports suck so hard then? If it's just SOOOO easy to port.
  i dintnt say its easy. i said it is possible. if the source is not abstracted enough or gets lost, another port will be very hard or impossible. of the amiga like systems aros supports biggest variety of platforms, sure not all of them very well, thats the question of man power. others do not even attempt that.
 
  >gcc is obscure?
  according to thor it was quite an effort even to port all the system sources to gcc, whichever version.
 
  >Maybe. We'd have to fork the M68K and start serious stripping all the shite you seem to be getting all wet for.
  no necessity to be offensive, im not getting wet and you may fork and strip at will. thats the benefit here.
 


Steve Ferrell

Posts 424
11 Jul 2019 02:29


@wawa
 
 
its not 24 years since aros68k is being developed. and very few people help while everybody else complains and makes demands. aros is not a closed source commercial product like os4 or morphos, so this behavior is inappropriate. i wont promise you anything. you can certainly use your copy of genuine amiga sytem as long as you wish, im not trying to force aros upon you, simply wanted to correct some facts.

 
That's funny because every web site that I can find in regard to AROS and its beginnings claims that AROS started in 1995:
 
  like this one:  EXTERNAL LINK 
and this one:  EXTERNAL LINK  and this one:  EXTERNAL LINK  and this one:  EXTERNAL LINK  and this one:  EXTERNAL LINK 
I can go on to cite many more web sites.  So have you forgotten how to count? Or are all these sites including the AROS site hosted on SourceForge simply mistaken about the year? 
 
The AROS site on SourceForge also claims on the same page that the goal was binary compatibility with OS 3.1, so the references here are also in regard to 68K AROS starting in 1995.....24 years ago by Aaron Digulla.
 
 


Wawa T

Posts 695
11 Jul 2019 02:32


@Steve Ferrell
  no i can count simple numbers. i said m68k port is not twenty years old. jason mcmullan picked up the kickstart bounty in 2010, after a number of volunteers failed.

and for what i know aros has initially only envisioned source (api) backward compatibility with ks3.1. the whole reasoning behind aros at that time must have been to bring amiga to a faster more viable platform, and m68k is not such a platform even today, admittedly.
the binary compatibility came into being with kickstart bounty and amiga-m68k target much later.


Steve Ferrell

Posts 424
11 Jul 2019 02:38


wawa t wrote:

  @Steve Ferrell
  no i can count simple numbers. i said m68k port is not twenty years old. jason mcmullan picked up the kickstart bounty in 2010, after a number of volunteers failed.
 

 
So you're saying that Aaron Digulla sat on his duff for 15 years?  I bet he'd like to chime in about now.

I don't know how much clearer that even the devs at SourceForge can make it for you when it says, "In the winter of 1995, Aaron Digulla, who was fed up with this situation, posted an RFC (request for comments) to the AOS mailing list in which he asked what the minimal common ground might be. Several options were given and the conclusion was that almost everyone would like to see an open OS which was compatible with AmigaOS 3.1 (Kickstart 40.68) on which further discussions could be based, to see what was possible and what was not.

And so work on AROS began, and the rest is history..." 

That's clearly 24 years whether it was coding or gathering the tools and resources (compilers, debuggers, documentation, developers, etc....) to start coding, unless you've been studying a new form of math.




Wawa T

Posts 695
11 Jul 2019 02:43


sorry, im not getting it, what you want to say about aaron.



Wawa T

Posts 695
11 Jul 2019 02:46


but if you want to say that aros is still in unsatisfactory state despite all the effort you might as well direct it at him.




Wawa T

Posts 695
11 Jul 2019 02:49


ah:

  Several options were given and the conclusion was that almost everyone would like to see an open OS which was compatible with AmigaOS 3.1 (Kickstart 40.68) on which further discussions could be based, to see what was possible and what was not.

may i ask where you read "_binary_" compatibility here and how aaron implemented it on i386?


Steve Ferrell

Posts 424
11 Jul 2019 02:54


wawa t wrote:

but if you want to say that aros is still in unsatisfactory state despite all the effort you might as well direct it at him.
 
 

Why would I want to bother him when he obviously has better things to do right now and you clearly stated you're working on the latest 68K AROS Vampire builds? So my comments are most definitely pointed in the right direction unless you were just joking about your involvement the AROS 68K project.  I never said you nor anyone else was wasting time, I pointed out that there's a lot of catching up to do if AROS 68K is expected to be a drop-in replacement for OS 3.x or ApolloOS...and I still stand by that statement.


Steve Ferrell

Posts 424
11 Jul 2019 02:57


wawa t wrote:

  ah:
   

    Several options were given and the conclusion was that almost everyone would like to see an open OS which was compatible with AmigaOS 3.1 (Kickstart 40.68) on which further discussions could be based, to see what was possible and what was not.
   

    may i ask where you read "_binary_" compatibility here and how aaron implemented it on i386?
 

 
 
It's in one of the same links I gave you ealier at: EXTERNAL LINK 
 
but since you've decided to make me your research assistant, here's what you're looking for and it's in regard to 68K binary compatiblity, not x86.

bullet point 3
 
 

Goal
The primary goals of the AROS project are to create an open source OS which:
 
1. Is as compatible as possible with AmigaOS 3.1 where appropriate;
2. Can be easily ported to different kinds of hardware architecture and processors, such as x86, PowerPC, Alpha, Sparc, HPPA;
3. Is "binary compatible" on Amiga, and as source compatible as possible on other hardware;
4. Can run as a stand-alone 'native' version, bootable directly from hard disk - or hosted, opening a window on an existing OS, to develop software and run Amiga and native applications at the same time;
5. Improves upon the functionality of AmigaOS.

 
I don't give a rat's @ass about x86 binaries.  We're talking about 68K binaries and AROS 68K.


Renee Cousins
(Apollo Team Member)
Posts 142
11 Jul 2019 03:00


wawa t wrote:

@Renee Cousins
 
  >So is AmigaOS. Came with every Amiga.
  if you mean the kickstart, then yes it has been included in price. but obviously for many people it isnt enough, because then there wouldnt be any followups or compilation bootlegs like coffin.

You don't have to use it. I don't.
wawa t wrote:
 
  >You say this like it's a positive.
  well. its positive in my book, the system can be developed systematicly insted of patching binaries in asm one by one and people can contribute. whats wrong with it?

Too many cooks? Systemd? Unity? Stability is more than just whether you crash or not.
wawa t wrote:

  >Baggage? That's your best argument.
  so you enjoy the legal situation its been under which led to no development or fixes or appropriate distribution being possible for twenty years?

I really don't care at all. I literally give zero shits. My Amiga 1000 and 1200 run perfectly fine whether Cloanto and Hyperion burn the IP into the ground.
wawa t wrote:
 
  >Really? Why do both the M68K and Pi ports suck so hard then? If it's just SOOOO easy to port.
  i dintnt say its easy. i said it is possible. if the source is not abstracted enough or gets lost, another port will be very hard or impossible. of the amiga like systems aros supports biggest variety of platforms, sure not all of them very well, thats the question of man power. others do not even attempt that.

It's not impossible on AmigaOS -- in fact, aside from the CPU and CIA chips, AmigaOS is one of the MOST hackable operating systems on earth. Name another operating system where the kernel can be swapped out while running. DraCo proved you could almost entirely unbolt AmigaOS from ANYTHING resembling a Commodore Amiga with little more than a little boot ROM.
wawa t wrote:
 
  >gcc is obscure?
  according to thor it was quite an effort even to port all the system sources to gcc, whichever version.

It's not obscure though. And it's been working fine for YEARS. Vbcc is there even if it's not that great and we can always grab a copy of SAS/C if we're okay with abandonware.
wawa t wrote:

  >Maybe. We'd have to fork the M68K and start serious stripping all the shite you seem to be getting all wet for.
  no necessity to be offensive, im not getting wet and you may fork and strip at will. thats the benefit here. 

Yeah, I have zero interest in this.


Wawa T

Posts 695
11 Jul 2019 03:03


fyi aaron isnt involved with aros since years anyway.
  look. im not even a coder. i do what i can, but it may not be much. yes, im involved to certain extent and i dont dispute that aros needs improvements especially on m68k front. you wanna talk to real devs? then join the slack channel.. but not just to complain. we are aware things are missing.

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