Opinions on AmigaOS4 from OSNews

[Floid]

As much as x86 makes all sorts of sense financially (and perhaps even thermally, these days), I don't see it making a lot of sense technically, especially after the horrendous investment made to actually Do Neat Things on PowerPC.

If anything's going to save the ship other than some secret project none of us know about, it'll be one of:

-A third party setting up a 'buffer' company between 'blue' hardware and 'red' software -- Who wants to play Dell for our little scene?

-Any unknown and unrelated manufacturer considering an EFIKA or Vixxen-esque system; after all, these are essentially reference designs, not to denigrate the work that goes into producing working ones.

-Or, my personal favorite if I haven't forgotten major details, someone finding a use for Cell.  This *would* require some further software effort, but given the supposedly lackluster performance of the PPC core, something as light as AmigaOS might be a boon to throughput through the streaming units.  Hard to say, though, as I haven't paid attention to that ecology in a while.

[Floid]

Waccoon wrote:

Linux is free.  I don't use it.  Why?  Because I hate how it works.  Again, the OS is judged by its features.

You should suck it up and try Ubuntu sometime; I've finally found something as decent as OS/2 from a user's standpoint.  Once Xgl and its ilk properly appears (is that slated for 6.4?), some rather interesting "features" will be there.

Floid:   As much as x86 makes all sorts of sense financially (and perhaps even thermally, these days), I don't see it making a lot of sense technically, especially after the horrendous investment made to actually Do Neat Things on PowerPC.

What neat things?

Petunia, which may be of limited utility due to the 'small' library of RTG software, but still plays a role in keeping continuity and satisfying what real legacy users remain...

Per Álmos, "Changing to other type of processor which is not PowerPC machine code compatible is pointless and almost impossible, because all the emulation code were made of PowerPC assembly and tied closely to this architecture."

There's no need to hard-code for VMX (IBM's version of Altivenc).

...Unless you've got a fairly complicated chunk of platform-emulation code that needs to 'compete' on performance with whatever MorphOS has got.  Of course, you could always ditch it...

I think Hyperion is being stubborn about this either to save themselves some work (ie, from actually doing it properly), or they just don't have enough experience outside the 68K and PPC communities to really know what they have to do.  This kind of development would be unacceptable in the "real" OS market, and will probably give Hyperions some major problems down the road.

I think you've missed something; everything points to OS4 being pretty portable, but PowerPC has been the main target for all the obvious reasons (presumed better 68k emulation performance, the fact that hardware was supposed to be available)...  Still, with the PowerPC version wrapped up and running, I imagine it'd be cheaper and more profitable to go out and design some PPC hardware and sell it 'immediately' than to shelf everything and declare x86 the new standard just in time to delay things even longer.