Nice idea, but a bit beyond me. Would be cool to be able to build your own mini A500 though.
Printable View
Nice idea, but a bit beyond me. Would be cool to be able to build your own mini A500 though.
They started a thread over at Amiga Org, to see how many were interested. I cant even find the original thread now, so im quessing nothing will come of this. Really liked the idea of having a mini-mig.
The problem with building and selling it would be
a) Kickstart ROM, although I think this could be solved by bundling a copy of Amiga Forever with each one sold which adds about £15 to the cost. I'd imagine giving a licence for AF would allow you to include the KS on a chip.
b) Electronics licencing - Not sure on the exacts of this, but don't electronics have to pass some safety certifications before they can be sold?
However these two are bypassed by making it a 'build it yourself' project. The end result if you have the skills really is quite impressive:
http://home.hetnet.nl/~weeren001/files/Minimig_rev1.JPG
all i can say is
AOS for x86 :D
ZetrO wrote:
Going x86 would for sure make things a bit easier. The hardware is cheap and readily available, and it would be easier for most people to try out the OS if they are interested. Not the worst move that could be made! :)Quote:
all i can say is AOS for x86
I too have wished for years that Amiga OS developers would just ditch the now dated and no longer viable PPC architecture (PPC chip development is no longer focused on computer development, but instead is now completely geared towards integration within appliances and other SFF technology), and instead move forward with an Amiga OS for x86 architecture.
Doing this would instantly remove the need for proprietary hardware, which is always expensive, suffers from delays and short manufacturing runs, and therefore is always hard to source. In contrast x86 architecture is readily available off the shelf, is cheap, and current Linux driver architecture could surely be used to improve compatibility.
It would also suddenly open up the Amiga OS to current PC technology, rather than that of the past. This would surely entice more game and productivity developers to start making products for the Amiga again. Much like Windows, Linux, and now Mac OS, the technology behind the OS is completely the same. It is just the OS and what it can do that makes the difference. This brings great choice to the user, who only has to invest in a single hardware platform, but then has access and the ability to run whichever OS and software they wish. Much better than having a load of custom proprietorial hardware platforms, each incompatible with each other, as was the case in the past.
To be honest i think we should all admit the AOS for anything other than Amiga hardware is a bit of a dead end. AOS on a PC is fine but there would be no point as there wouldn't be any software to run on it. You may be able to run the old Amiga games but at the end of the day thats what UAE is for. The only reason PPC achitecture was used is because it was sold as 68000 code could be easily ported to PPC's as the both chips were made by Motorola. Why would anyone build a x86 accelerator board for either classic amigas or Amiga Ones? The market is extremely limited, the boards would be extremely expensive and by the time a board i sdesigned and new software is written the x86 would be out of date. Then there would be the cooling issues. The board would have to be either zorro (which would be pointless as the bus speeds would be too slow and the achitecture is outdated or PCI. With PCI you would be taking up at least the next 3 slots with the heatsink and fan.
The reality is that OS4 will be the last AOS and the Amigas will then be truely relegated to Fanboys and Nostalgia freaks like us.
You can't continue to support and keep ageing technology in the present, especially as the platform it is all connected into gets older and older. Therefore a future version of Amiga OS should be based on the x86 hardware platform and JIT and other emulation techniques used to support backwards compatibility.
I know the budget isn't comparable, but look at Apple. They moved their complete platform from PPC to x86 and all programs needed to be re-written to make then native and therefore run faster, or at all. Many moaned, but now the move is complete they are happy again.
The truth is that the PPC processor is dead as a computer platform. It has gone off in a different direction and only stayed in development as a CPU for the desktop market for so long because of Apple. Now they have ditched it there is no viable desktop computer market for PPC to continue its development down that route. Therefore the main alternative x86 has to be the logical move for any OS looking to still be here in another 5 years.
In truth i have argued an X86 aproach for years on other forums,
alas the die-hards seem to think that by including an intel based co-processor perverts the Amiga cause (one of the reasons I am not on A.Org)
One could easly run a C7 1ghz x86 companion chip passively cooled or with a laptop cpu cooling unit, the tech has already been designed and implemented all it needs is to be constructed for another use...
one could easly add to this a 68k compatible chip and SDRAM interface
if anyone here remembers the PC286 card for the A500, it had 1MB of ram on board that it could share with the miggy, as well as use chip and fast.
The way i see the x86 AOS, would only really work for me atleast with a transparent 68k emulator that would function in seamless motion as the code-header chunks are inspected and then executed for the right cpu (similar to the warp-up code).
On another more direct x86 AOS (legacy almost free ;) ) I would of hoped Amiga.inc would of adopted a basic linux and bade it thier own, from this there own implementation of OpenGL with enough development it would offered another gamming platform.... hell theres not that much to it these days...
grab a cheap grafix solution,
slap a USB base controller,
yam on a simple CPU
proprietary is for large chip producing corps.
The problem is what would be the point of AOS on a x86 processor? Linux works because it is free and has software support. AOS would cost money and have no software support. As for supporting old architecture, you're right there is no point in supporting the classic or even Amiga one archetecture and as AOS hos nothing to offer anyone the AOS for x86 is pointless. As an OS it offers no innovations and doesn't measure up to any of the newer OS so why bother? It would just be an exercise in futility and a backroom project for fanboys.