Amiga.org
Amiga News and Community Announcements => Amiga News and Community Announcements => Amiga Programming and Development => Topic started by: dammy on November 05, 2007, 07:35:03 PM
-
Richard D., of EUAE fame has been assigned to TeamAROS bounty #7,
EUAE Integration (http://thenostromo.com/teamaros2/?number=7). Pull out those electronic wallets, time to start donating to show your support of Rich's Work.
EUAE (http://www.rcdrummond.net/uae/)
-
This is great :-o I will definitly contribute to motivate Rich :-D
-
I would not mind dropping some cash on this too! This project would certainly give AROS an instant boost into useability!
-
Consider another donation going this way!!!
-
:Dancing Kitty: :Dancing Kitty: :Dancing Kitty: :-D
-
Thank you Rich!
-
The day has finally come! Amazing! :-D
-
If I have transferred ROMs from my existing Amiga, they can be used?
-
What is the ETA? I am thinking about a year until a functional release is done, but mine is quite a uninformed suposition.
-
@little
Honestly, for a basic version, not terribly much is needed, just a lot of little things. AROS already can flag an app as 68k-based, you just need to then tie that into launching EUAE. You'd still need the original Amiga kickstart until the AROS Kickstart II bounty is finished. Yes, it's a bit more complicated than that, but you get the idea.
-
Georg do some months ago a Video that integrate e-uae in Linux nice.dont know if he have time to continue...
same should be possible in AROS too he told in AROS ML
http://utenti.lycos.it/stegerg/aros/uae.avi
-
Well this is starting to look promising.
Oh and thanks for the info downix, hopefully the Kickstart replacement will get finished as well as this one that's just been assigned.
Between AROS and the Minimig it's starting to look like the platform(s) is/aren't even going to need to depend on any kind of big company or licensing.
Well maybe I'm jumping the gun a bit here, but I like to keep that frame of thinking.
-
@Fixer
We can only hope
-
Why is a Kickstart replacement still needed for the x86 or PPC versions?
I could see if you wanted to boot directly from a 68k floppy on a Minimig, but that would still be impossible on x86 and PPC platforms without loading the OS first and E-UAE.
Wouldn't this intergration trap and translate Kickstart calls to the OS's version?
-
Amazing news, thank you Richard. A whole new world will open for AROS.
-
@lou_dias
Only if you wanted to spend eyars at it for no real gain. The simplest, and most effective method of integration would be to have a method to identify when a non-native App is being called (already in place) then to have it launch EUAE to then handle said application. Once you have this level done, then you can start getting fancier.
-
Wouldn't this intergration trap and translate Kickstart calls to the OS's version?
Yes, but more than that, it must also "trap" poking/peeking of OS structures since application programming under AmigaOS relies/relied quite alot on that.
-
Wouldn't this intergration trap and translate Kickstart calls to the OS's version?
You need to have a full 68K OS for the 68k program so that they can mess around with system structures (which they are not supposed to but Commodore never said not to...)... Then the higher level 68K OS fucntions can be patched to call the AROS native functions, like the windowing system and the input systems... giving the 68k prorgrams the appreaence of running totally nativly... but in fact they will be safely locked inside a virtual sandbox... think OS9 classic on PPC MacOSX
-
IMHO he'd be mad to any AROS hacking.
I think he should just create a method of utilising the UAE screen buffer, to "greenscreen" the real AmigaOS desktop background and superimpose desktop elements of the real AmigaOS desktop over AROS Wanderer.
There's no reason he couldn't establish a protocol or API for this to work on all other E-UAE capable OSes like Linux, MorphOS or OS4.
He may need to run an app on the real AmigaOS to establish a communication mechanism to adjust the real AmigaOS's visual implementation and call apps as required. You don't want to start multiple UAEs, but just have one. Later he can put some smarts into this real AmigaOS app to merge it seemlessly with the host OS (for instance to run inside hosted windows).
Such a method is used by Mac Virtualisation software and has been coined the term "coherence".
Down the track, if the AROS68k version becomes fully 68k compatable you might think of calling x86 routines to speed up classic apps. While I don't know the state of the AROS68k version I would imagine that this scenario is way into the distant future.
Ok. Going back to my happy place now.
-
You don't want to start multiple UAEs, but just have one.
No one said apps should run in their own instances (unless they take over the virtual Amiga).
-
Coherence mode, ahh the pleasures of being able to run Windows as an integrated part of Mac OSX. To piles of bloat all at once! :)
I don't know much beyond Turbo Basic (he he he), so I'm not sure how this would be done, but what BigBenAussie said sounds logical because the other concept of having multiple instances of UAE running at the same time would be a resource killer, and would prevent the fluid mulit-tasking that we are accustomed to on real Amiga hardware, thus making this whole venture pretty pointless considering that the whole point of e-UAE is to catch up with WinUAE which would finally give us a competent and fast UAE that is not restricted to running on Windows, and therefore held back by the mariad of short comings within the Windows operating system that make UAE seem just a little bit flakey and odd, no matter how well your PC is decked out. :)
Of course there is also the fact that AROS is supposed to be a clone of Workbench, and Workbench was always fluid and resource friendly, so it would not make much sence at all to want integrated 68xxx Amiga support at the cost of all your RAM, and CPU power. I mean I can't see wanting to have a few classic Amiga apps on my screen, when they are going to make my whole system bogg down, if that were the case, I'd just let go of my integration dreams, and just fire up e-UAE again so that I can run more than one thing at a time without having to worry.
-
As I wrote, no one said (wrote) anything about multiple instances of EUAE, *unless* the second/third/fourth.. emulated app would *take over* the virtual (emulated) Amiga, or are you fine with seeing the other apps "sleep" when the "offending" app says "hello" ?
-
He he, you must have replied as I typed.
-
:laughing:
-
@BigBenAussie
It's not that no one has thought about "your" method, the problem is the level of integration this would provide, how would you "greenscreen" out the emulated AmigaOS desktop elements (affectively layers) and "superimpose" those elements (layers) AND integrate those into the host windowing system (every layer must apperar to be a native one) without hacking/patching the emulated KS/OS ?
If you look at Georg's (the AROS dev that did something similar) "greenscreen" method then look specifically at 0:01:05 when he clicks the depth gadget, the window only goes to the rear of the UEA/AmigaOS instance windows, NOT to the rear of host windows, this gives you a fact, with this method all UAE/AmigaOS windows (layers) are only one single "alpha-blended" layer in the host OS.
-
Is there somewhere wehere we can track progress of this? Is there a blog or something?