Welcome, Guest. Please login or register.

Author Topic: MorphOS ahead of AROS?  (Read 72553 times)

Description:

0 Members and 2 Guests are viewing this topic.

Offline paolone

  • Sr. Member
  • ****
  • Join Date: Dec 2007
  • Posts: 382
    • Show all replies
    • http://www.icarosdesktop.org
Re: MorphOS ahead of AROS?
« on: April 04, 2012, 02:09:40 PM »
Quote from: takemehomegrandma;686775
Yes, like two applications running on two separate computers (with separate OS code), instead of the two applications running in the same environment.

Wrong. AmiBridge is a set of scripts which does a completely different things. When Janus coherency mode was not available, it opened a new UAE/Workbench session for every 68K program you launched, moreover restricting the screen resolution to AGA pal hi res. But once coherency mode was completed by Oliver, this model has been immediately abandoned and Amibridge just became a faster, easier way to integrate AmigaOS and KickStart files into the Icaros installation. Once this has been done, it also assists the user in setting up coherency mode correctly, in a semi-automated way. Done that, you can even run the Amiga M68K virtual machine inside AROS at every startup. It will just sit there until you run any 68K application, and then you'll see applications windows seamlessly integrated in the AROS workbench. You can move and resize them as you wish. Programs will run in a single, shared instance of UAE just like they ran on a single physical Amiga machine. M68K programs can then access to AROS partitions (with the only default limitation of read-only mode for system drive - but the user is free to change it) and read/save the same files. AROS and AmigaOS share the clipboard too, and this ensures there is enough communication between the two world to just forget about messing with 68k and x86 libraries in the same environment (I can't understand why you OS4/MOS people still continue considering this as a feature: I don't even want to imagine how many context-switches and bytecode translations are needed to use, in the same program, libraries written for totally different architectures like m68k and PPC, really!). In the end, I consider this approach safer and better than others, since M68K programs run in a coherent M68K environment, while x86 ones do the same on a coherent and well separed x86 one. User just will see little differences on the screen, due to different fonts used in the systems and maybe some slower refresh in M68K windows. But this is something they can quitely live with, counting they have paid nothing for the OS, less than 10 euros for the Amiga environment (last time I looked into Cloanto's site, Amiga Forever Value - which is perfect for Amibridge! - was priced € 9,99), and surely not € 2500 for the hardware (or maybe they did, but not with AROS in mind).
p.bes

 

Offline paolone

  • Sr. Member
  • ****
  • Join Date: Dec 2007
  • Posts: 382
    • Show all replies
    • http://www.icarosdesktop.org
Re: MorphOS ahead of AROS?
« Reply #1 on: April 04, 2012, 02:23:28 PM »
Quote from: eb15;686847
I follow AROS developments, but its still not ready for me to make my  primary usage platform, like I used an Amiga back in the early 90s in  place of unix variants and windows.  I don't have a machine that runs MorphOS, but as far as I'm aware from what I've read....

Graphics wise, MorphOS has video overlay and some hardware accelerated  alpha channel aware graphics functions implemented which are missing or  incompletely implemented in AROS.  I'm not sure which are HIDD/driver  features, and which are higher levels, but it will be nice to see AROS  continue to improve in these areas.

MorphOS has commercial quality Poseidon USB and TurboPrint printers support.  Other areas AROS is slightly lagging in.

MorphOS allows partitions greater than 128GB with their newer SFS, while  AROS is lagging in Amiga style native file systems that support the  larger disks of today, unless you partition into many slices.

MorphOS has had better MUI and GUI theme integration system wide.
Probably better console and text editor implementations.

AROS has been improving much in the past couple of years, but still has a  ways to go.  The m68k-amiga emulation environment probably needs some  more hooks into the native environment and a gui-less AROS kickstart rom workbench  program that just communicates to the aros-native one to do its work,  with a single shared mui/theme prefs between the two, so a user never  needs to see a second disconnected GUI that behaves differently than the  native environment.

Perfect. I think this is exactly the kind of answers Mazze expected from this topic. You listed some "cons" AROS developers should look at in the future.
p.bes

 

Offline paolone

  • Sr. Member
  • ****
  • Join Date: Dec 2007
  • Posts: 382
    • Show all replies
    • http://www.icarosdesktop.org
Re: MorphOS ahead of AROS?
« Reply #2 on: April 06, 2012, 09:59:11 AM »
Quote from: takemehomegrandma;686989
Maybe I don't understand, but from what you just said there, it pretty much sounded *exactly* what I meant when I said "like two applications running on two separate computers (with separate OS code), instead of the two applications running in the same environment", one host/AROS box, and one 68k box.

Sorry, I misread your statements and thought you were saying AMIGA PROGRAMS were working EACH ONES in ITS OWN Amiga sandbox, while you were really saying ALL AMIGA PROGRAMS work in THEIR OWN Amiga sandbox while AROS programs do the same in the rest of the OS. Which is absolutely true.

Quote
You are talking about making it *look* like the same system in a purely visual/theme sense (and have access to a common clipboard), but in my view it's still a lot more like running WinUAE on a Windows7 machine (where the 68k part also can access the host systems file system, etc)

Yes, it's much like running WinUAE inside of Windows, or even XP programs in Windows 7's XP mode, or MacOS Classic programs in earlier versions of MacOS X, as you can see baci AmiBridge's idea has been used before (Apple), and found very famouse followers afterwards (Microsoft).

Quote
...than what both MorphOS and OS4 offers today, where you simply don't have any HW emulation or separate/shielded off "boxes" at all, but all binaries are run the same way, share the same memory space, the same resources, data, sheduling, messaging, arexx, *everything*, no matter if they are 68k or PPC, there simply is no difference at all (it *is* one and the same, not just visually so)!

Half true. You have to place some boxes here and there, in order to run M68K bytecode on PPC processors. Why are so many OS4/MOS users so sure there isn't any emulation at all in the process? It must be somewhere, hidden to the eyes of users, but 68K software doesn't run on PPC hardware by magic. Translation necessarily introduces overhead and potential flaws, not counting that much software relies on classic hardware specs to run correctly. That's why I prefer emulating the whole hardware and run M68K software on its own, no matter that's the only one option available on AROS: I would do the same on MorphOS and AmigaOS 4, if safeness and stability would get more important than speed.
p.bes