Mainstream HW vs. Custom Niche HW

[Iggy]

I, for one, am completely tired of hearing this same phrase "Migrate to an architecture with a pulse".
 
Currently, PPCs are available that completely outstrip all currently available ARM designs (or any that are likely to be introduced in the near future).
That leaves only X86 as a possible direction for ISA change and I DO NOT want to go that route.
 
We have survived this long without adopting the archetecture of our primary competitor.
 
The poster rightly points out that adopting low cost Mac hardware may have influenced increased adoption of MorphOS.
Fine, then there are plenty of G5 machines available at a very low cost that would be far faster then anything offered in the OS4 market or under the ARM archectecture.
 
And, as low cost proprietary design go, I LIKE what A-eon has accomplished with the X1000. Is it overpriced? Yes.
Could a sucessor be built at lower cost?
Read this VERY carefully, I am absolutely sure it (or they) will be.
 
If you want X86, please by all means adopt and support AROS.
 
But have you noticed that AROS is now moving to 68K, ARM, and PPC?
 
Please stop trying to increase the difficulty in supporting and developing my favorite operating system.
 
An ISA change is no trivial matter AND it eliminates backward compatibility with all PPC software without another emulation layer.
 
A move like this would set MorphOS development back at least a couple years. And it would make targeting specific hardware extremely difficult.
 
And you should keep that in mind, because unlike Windows or OSX, MorphOS developers can NOT afford to develop drivers for all the hardware available within a specific ISA and must select specific platforms.
 
You complain that our current hardware is dated? Personally, I think its damned good and I am willing to benchmark my MorphOS system against any OS4 system including the X1000.
I know that will invite the inevitable complaint that I'm not drawing comparisons to AROS.
That is because I have no problem co-existing with AROS and even using and supporting it.
 
You lunk heads do realize that one of the MorphOS developers works on AROS? Don't you? And the competition and cross development with AOS4 doesn't hurt either.
 
So, think about what you're calling into question VERY carefully.
The more I use MorphOS, and the better I get to know its developers, the greater is my faith and conviction that they have made the correct choices for our community.
 
I don't think continuing to move in this direction would be a mistake.
 
But a change of course could be.
 
Jim

[Iggy]

@Iggy
 
OMG, where on earth did all that come from? Having a bad day? Then please don't take it out on me, and please don't try to change the topic already!
 
:whack:

I am in no way trying to change the topic.
You brought it up and you keep repeating it while not even giving credit to your own points.
 
Mac hardware was sucessful?
GREAT lets keep using it.
 
And, btw, I've been following this really close so if you think I'm taking this too seriously its because if the developer take outside advice too seriously they may make a mistake that costs us the future of the OS.
 
I'm not having a bad day (except for the hurricane outside - which is now about at its peak). I'm just aware of a few things you aren't that are going on behind the scenes right now.
 
You really need to leave this topic alone and let the MorphOS developers decide how to continue forward.

[Iggy]

I think I can remember Geit estimating it would need 5 years to port MOS to another platform (X86 or ARM) and that would be 5 years without any updates because all would be busy with the migration. So obviously MOS was never designed to be (easy) portable like f.e. Aros. My final conclusion is MOS will (propably) never be ported to a different architecture because of lack of resources. For people that can live with PPC it is a good choice, when you want something to run on modern new standard hardware you only have the option to use Aros.

Thank you Olaf,
I was being kind.
In the interum our OS would languish.
 
Like you said, want something different? Its already there.
Don't F'up my OS.

[Iggy]

Sorry takemehomegrandma,
Maybe I am wrapped a little too tight today.
 
I am worried that the MorphOS develoers will take this call for an ISA change to seriously when we've still got more we can do.
If an ISA change does occur I'd rather see it be to ARM as it is a more broadly licensed archetecture (and its not X86).
BUT ARM just isn't powerful enough (yet).
 
And an ISA change is going to take a long time and hopefully the development will occur concurrent with support for our current systems.
 
Currently, three courses are possible. Stay where we are, move to G5, or adopt a new low volume platform.
I'm satisfied with the first, I could see the arguments supporting the second as being the most affordable, and would really like to see the last occur.
 
Why would I want something like the X1000 when it was Powermac support that drew me to MorphOS?
Because I'd like something new, powerful, and compatible.
We can still support what we have, and have a high end option for both bragging rights and the enevitable desire of a few to own the best.

[Iggy]

How do you mean "no transition"? X86 and PPC in one system?

Not impossible. A PPC add on card for a PCIe slot running with VM.

[Iggy]

Why should i buy an add on if i can emulate it?

You can't emulate a 2.5 GHz quad core PPC with an X86.
That's why.

[Iggy]

OK, you are in the loop, you know the secret handshake and all that. Good for you.
 
It's great if the MorphOS developers are thinking about the future. Otherwise there won't be any, rest assured about that.  

No, I'm not in the loop, I know I few people who have been told more then they should have, and few few others who failed to deny some pretty accurate guesses by Andreas Wolf and myself.
 
Let's just say that, yes, its pretty certain developments are ongoing.
AND that no certain commitments have been made.
 
However, I should refrain from attacking you as I know you support this stuff as much as I do.
 
It is cool that we have a future. All three ng OS' appear to have a future (although the pace of OS4 development is glacial).
 
One thing is certain, until a commitment is made you are not going to hear a peep out of the MorphOS development team.
 
After all, how much warning did we get on eMac support?
Uh, none. I just looked on the web and it was there (before the anticipated Powermac support).
 
So, whatever they do, we're never "in on it".
 
Sorry again about what may have appeared to be flaming.
 
Jim

[Iggy]

AFAIK an architecture change could be done fairly quickly if you "lower your standards" (or rather: altering them). This could also open up for "raising your standards" in other areas, by incorporating features not really possible on Amiga otherwise, like true SMP, true MP, true 64-bit, etc. A whole set of new challanges for the OS developers. Those things would probably take time, but I doubt an actual "migration" would be very difficult if you are prepared to start with a clean slate in terms of Amiga backwards compatibility. And you could always do like AROS, and incorporate some transparent UAE.

I actually wouldn't mind this as a parallel path.
But without the 3.1 API its not MorphOS.
Still, with the Quark kernel, a really tight compact OS is possible.
 
I've used micro-kernel OS' since the mid '80s and that could be pretty cool.
 
On this idea, I wholly support you.
 
I wonder if Ralph would allow Quark to be used in a Open OS?

[Iggy]

I've seen benchmark of the Cortex A15 @ 1.7GHz that give it parity with the G4 (not the G5), so I'm not sure about the benchmarks you're quoting.

But don't get me wrong, I'm a big fan of the ARM ISA.
And once it moves to 64 bits things are really going to take off.

BTW - I haven't killed anybody lately (that I'll admit to). :lol:

[Iggy]

@som99

Yeah, the jump in performance over the A9 (much greater then I expected) has me pretty jazzed.
I've been considering buying a Chromebook, but only if I can get it to run a Linux distro like Ubuntu.

ChromeOS' requirement of always having an internet connection bugs me. There's always going to be somewhere where that won't work.

A15 quad cores should be real killers.
Glad to see something better then the Krait introduced.

I just wish the development boards didn't cost twice what a quad core A9 board cost.

And, btw, not only wasn't I on anybody's list to be killed today, but it looks like Mother Nature has decided to spare me too (should I tempt fate by posting that this early?)

[Iggy]

woof!

[Iggy]

Got a link?

Nope. Mostly PM with people who wouldn't appreciate being quoted. So I shouldn't have mentioned it.

Why not g5?

Not ruled out. But some people (like Bigfoot) do not have a G5. And Powerbook was deemed more important.
Something is likely to happen by the time 3.2 is released.

[Iggy]

The joke kind of went over your head there, didn't it?

:lol:

Probably, then again I'm not Andreas so I don't constantly quote links (I don't even know how he remembers them all).