Why not AmigaOS4 for x86 Platforms?

[bloodline]

Okay maybe I am getting a little old here, but I remember being on some board a few years ago and some dude was up on there promoting a "transputer" and how it's processing power was gonna make the Amiga the most powerful computing platform on the earth.. They ended up not having a market and their wild claims only came to fruition on the Atari ST..  

Actually the Transputer thing is a great example, which is similar to the Cell hype.

I rather like the transputer idea, but it wasa miserable failure for lots of reasons. But I've never forgoten it and neither did the Chip designers... what we are seeing now with modern CPU's are features of these old failed experiments finding a place in the computing mainstream... one idea a did love about the Transputer was the idea of lots of serial links rather than a single parallel bus... and look... we now have a hyper transport technology and PCI-Express... great stuff!!

May

[bloodline]

Waccoon wrote:

Bloodline:  I rather like the transputer idea, but it was a miserable failure for lots of reasons.

Namely, good compilers didn't exist, OSes didn't support multithreading, and doing it yourself was a huge pain.  The world just wasn't ready for super-computer type code on a home computer.

It really all has to do more with coding practices than hardware.  I sure hope Sony's dev tools for the PS3 don't suck anywhere near as bad as they did on the PS2.

Technology also had quite a lot to do with it... For the Transputer idea to work you needed several chips (ie more than one), but each Transputer chip cost around $800 and performed far worse on it's own than the $100-$200 68030 and 386 of the same era.

[bloodline]

BigBenAussie wrote:

Surely, the speed can't be bad enough for me to notice a difference between an A1 and a high end PC. Am I really going to notice a computational speed difference?

Get yourself a Cheap high end PC... or at least borrow one, before you ask that question.

[bloodline]

terminator wrote:

mdwh2 wrote:

terminator wrote:
Just how long do you think it would survive in the MS sandbox?

It would be dead in minutes. Who in their right mind would waste time developing for an x86 version of OS4 when they may as well just release it for Windows.

Why would anyone buy anything for the x86 version of OS4, when they can buy almost anything at the local pc superstore, today.  Not 18 months from now.

Replace "x86" with "PPC" in what you wrote, and the same points apply.

Would you wait 2 years to buy an Amiga version of a game, and pay $75 for it, when you can buy the windows version right now for $50?

If you answered yes, would you be willing to buy that title for $75 in 2 years, instead of buying the older, now discounted Windows version, which sells on the next rack for $30?

AmigaOS cannot coexist on the wintel platform with windows.  Windows will trump everything.  MS has annihilated  every other operating system that tried to exist in their sphere.  Those who tried to make money from an alternate OS in that ecosystem were eliminated.

Your argument is totally irelevent.

You are saying that an Amiga owner would buy their Amiga and if it were not for the fact that their harware is obsolete and incompatible, they would drop AmigaOS as soon as they got it home... :roll: That's a lot of faith you have in AmigaOS, users have to be FORCED to use it.

[bloodline]

Karlos wrote:

Hammer wrote:

Wow. Any idea what those compatibility issues are?

Recall in AW.net threads about 750FX Vs 750GX and AOS4-Pre issues. There are *minor* different between the two PPC32, but sufficiently enough to cause some issues to AOS4-pre.

“POWER everywhere” is nice IF they freeze ISA standard i.e. none of this 68K style ISA cut&paste. The *minor* incompatibility between PowerPC chips simply sucks from mainstream programmer’s view point.

I'm not disagreeing (somehow disagreeing with Hammer about anything CPU related seems about as sensible as jamming one's tongue in a breville), but what level of incompatibility are we talking about here? I mean almost every member of the 680x0 series had minor differences with those before it - quite majorly so in the 68040 / 68060, but  I don't recall it ever being a major problem.

Is it the fact that existing instructions with the same opcode values have changed implementation or what, exactly?

Had AMD run Motorola, the PPC would have had a 68K frontend, which could be switched in and out like the Athlon64's Long mode... Also the PPC Mode would have been called 64 bit mode (As the ISA would have been 64bit) and the promotional material would have said: "The New '68K64', can run all you existing apps, and ready for MacOS-64 and AmigaOS-64!". :-D

Hammer is right though, had Motorola kept better compatibility between all it's products they would have survived better on the desktop.

Err... Windows 3.11 runs on my Althon64...  :inquisitive: Weird...

[bloodline]

DonnyEMU wrote:

Karlos wrote:

dammy wrote:
by Karlos on 2005/1/26 17:59:19

f it had gone x86, then great, I would have saved money too.

Then what do you call AROS, chop liver?

Dammy
TeamAROS

I have downloaded AROS periodically and taken it for a spin on the old PC. And I have to confess, it's about as much fun as one can have with x86 :-)

However, and don't take it personally but for me, AROS is not AmigaOS. It is source compatible and I do intend to code on it at some point but I prefer to wait until it has matured further. Also, I want to see where the OS is going as opposed to just where it has been. In this regard, I feel AROS and AmigaOS will diverge considerably. So, which route to follow? For me, that depends on which of the two is presently the most compatible.

It's probably not a good quantifiable reason, but one of the reasons I feel more of a connection with OS4 is that it already runs on my classic PPC, which has several bootable versions of AmigaOS. It runs the old 3.x applications (well the 680x0 ones), even the hardware banging ones, without UAE , as well as the newer OS4 native ones. I appreciate that the A1 does not have this level of compatibility, but it will always be more compatible with existing software than AROS (that is, until, AROS develops some kind of internal 680x0 emulation).

This leads me to the question just HOW compatible is OS/4 with old Amiga software.. I know that OS 3.9 enabled some PPC compatibility, but honestly to me emulation is emulation. Whether you have to run it through software or some OS later.. If you were to run a "classic" mac application on OS X it would load an entire copy of the old Mac OS just to run that application. At the chip level the PowerPC really has nothing in common with the 680x0. Most people today run a version of "UAE" just to be compatible with applications for compatiblity with old custom chips.

I think karlos's argument revolves around being able to run the OS (Be it AmigaOS4 or AROS) on his existing hardware. Funnily enough that is my argument too... the only difference is that I have invested in newer hardware, which I had to do since I now have a large x86 software investment spanning nearly 5 years. Despite exclusivly owing an Amiga for 10 years (and inclusivly for 16 years), I probably only have maybe 3 or 4 years worth of software invesment, 100% of which is not irrepalceable with modern software.

Like karlos, AmigaOS 4 only becomes interesting to me when I can run it on my existing hardware, but unlike karlos I can't get OS4 to run on my BlizzPPC... yet. But I can run AROS on my existing hardware.

What does interest me is what happens when Karlos's BlizzPPC dies... :-(

[bloodline]

OS4 and MOS are excellent in this respect in that they transparently run 680x0 stuff, they don't need to do a MacOSX style "load the previous OS" in order to do it and they certianly don't need to use a full blown UAE style emulation.

I think the MacOS approach is actually the best all round idea... removes the legacy dependacy.

[bloodline]

I absolutely guarentee you would not say that if you ever had to use it

I have used... it made me want to upgrade the os9 software to osX software... which makes good business sense, and perpetuates the platform :-)

[bloodline]

Waccoon wrote:

Bloodline:  I have used... it made me want to upgrade the os9 software to osX software... which makes good business sense, and perpetuates the platform

Yeah, that's my beef, too.  If it was cheaper for me to get an OS4 platform (hundreds and not a full thousand), it would be easy for me to start writing software for it, which is what Amiga really needs to survive.

All this talk about what hardware matters, and after we have it, still no software.  It's tough getting used to a new platform when companies are terrified of making anything available to developers because they're worried about piracy.  That's a sure-fire way to kill a platform.

There is too much talk of hardware... it seems like Amiga users have never realised that hardware is now a commodity... it's irrelavant to the equation of a computer platform, you just buy the cheapest that does the job.

What is important is software... the REASON why you buy the hardawre in the first place. And 10-15 year old apps, regardless of how good they are have nothing to do with people buying software.

[bloodline]

DavidF215 wrote:

terminator wrote:
(Linux is an exception, it isn't a commercial product, nor does it really have any traction outside of a few specialized zones.  It costs nothing but your time to install, so it's no big loss if you don't use it.)

Say what? Then why is Microsoft so concerned about it?

Because Linux is free and Windows is not.

[bloodline]

I'll download AMOS and try it out. Don't know how well it'll do on my 1900+ AMD chipset.
 

It should run great, if you have a Nvidia gfx card (and a SBLive! sound card), then it will be brilliant!

[bloodline]

AmigaMac wrote:
@Waccoon

Like it or not, PPC and non-secured filesystems are the future of Amiga -- because Hyperion either didn't think about the future, or didn't want to.

Not sure about the thoughts on the filesystem, but as far as PowerPC is concerned, Hyperion did very well to think about the future and the latest news on the Cell processor will attest to that phenomenon.

Cell processor will attest to what? That Computer architecture is still plodding down the same old path?

[bloodline]

AmigaMac wrote:
Cell processor will attest to what? That Computer architecture is still plodding down the same old path?

Old path?  Well if you consider PowerPC old, then I guess x86 falls under ancient (and eventually history)!

"Cell" is just a flashy buzzword... it will take time before anything general purpose will come from it.

[bloodline]

AmigaMac wrote:
"Cell" is just a flashy buzzword... it will take time before anything general purpose will come from it.

Sorta like "Extreme" under Intel's Pentium line moniker, nothing really exciting beyond the hype marketing that drives the name?!

Exactly... though more like their HT hype...