MorphOS on Power Mac G5

[Karlos]

Valid point. I have no interest in playing Flash based games. I wish I'd caught the 64bit version you mentioned, I still regularly dabble with Linux (probably because even with my dissatisfaction with it, I appreciate its UNIX underpinnings).

It's open, I can probably send you the .so

[Karlos]

That is a good point and it would be useful.
Its always good to see that even when we've got open debates raging, we're still (mostly) a civil lot.
I'm not even vaguely aware of how to send a PM on Amiga.org, but I can get you may e-mail address somehow.

Thanks

PM info is in the site FAQ: http://www.amiga.org/forums/faq.php?faq=vb3_user_profile#faq_vb3_private_messages

The simplest way to send me a PM would be to click on my username to the left of this post and select "send a private message to karlos" from the menu that pops up

[Karlos]

@DaNi

I see your altivec and raise you SSE3.

I see my SSE3 and raise it CUDA, but meh :lol:

[Karlos]

That's a very bold statement you make. And I think it's more opinion than fact. Obsolete I would contend is relative. If a single core G5 allows me to do everything I possibly need in a computer, is it truly obsolete?

I still use my classic amiga machines and they are of more use than an arcane hobby. My 1200 is still my main music production system, for example. It doesn't make it any less obsolete.

You may contend it is a relative term but, well, you correct yourself here:

By definition, obsolete means: no longer produced or used. Well, my MacBook Pro Core Duo 2.4 GHz model is no longer produced. Is it obsolete? I would say no since it's still useful to me. By the same token, a G5 would absolutely be useful to me so therefore it is NOT obsolete.

Your entire computer may not be produced and used but the components that it is made from are. However, no desktop systems are built with G5 processors, nor have been for some time.

You say "a G5 would absolutely be useful to me". To do what, exactly? This is what I have been asking. What do you routinely do in MOS (assuming you are a user) that is too slow on your current machine that none of your other machines (core2 macbook included) can't do for you already?

I dunno why people are interpreting my position here as particularly contentious. All I actually asked was what do people need G5 (or PA6T) for when it comes to running amiga apps? And, to be entirely honest, I only asked that thanks to the stupid and utterly predictable "yay G5 mac FTW totally pwns PA6T!!111!" remarks from certain members that have posted in this and related threads. Sure, a 2.7GHz G5 is going to outperform a 1.6GHz PA6T, I'd be surprised if it were not true, but what does it actually matter? It isn't as if G5 is an option for OS4 users and nor is PA6T an option for MOS users, so who cares either way?

Now, I've already said I totally understand the "I want to run my favourite OS on the fastest hw available" motivation. However, there is not one single amiga application that actually requires a G5 or PA6T that you can't get for an x86/64, so if you need to run those apps, cheaper and faster alternatives exist. Logically, you can only justify the former option if you are basically opposed to the latter one.

Now, for a long time, the point of how insanely fast and efficient MOS is on G4 mac hardware has been stressed and I am quite sure of it. Moving to G5 is an obvious and natural progression but why now, exactly? Obviously, I'm not privy to the development plan for MOS, but I am aware that there are plenty of other things that could be focused on instead - support for more graphics cards (especially now that gallium opens up nVidia support, also used in apple machines), support for wifi and other stuff that is surely of more immediate use to more people.

[Karlos]

I hear what you're saying Karlos but you can widen your horizons a bit... Blender for example would benfit greatly from the G5's power. You may not be into GFX but the bottom line is for the price of a nice A1200 030 based setup you gain a really powerfull system that has access to Mac OSX leopard 10.58. The best of both worlds for many of us.

Of course, but given that most people here also seem to have x86 boxes too:

However, there is not one single amiga application that actually requires a G5 or PA6T that you can't get for an x86/64, so if you need to run those apps, cheaper and faster alternatives exist. Logically, you can only justify the former option if you are basically opposed to the latter one.

Perhaps it's just me, but I can't imagine myself sitting there thinking:

"I need to run blender. My time is important! I have a stinky quad core PC already with free OS and blender available, but what I really need is to run it on a considerably slower G5 and buy an OS to run on it so that I can wait several times longer for it to render my stuff!"

Perhaps is my only system was a G4 box, I might think:

"I need to run blender. My time is important! It's not so fast on this G4, a G5 would be nice."

But I'm much more likely to think

"I need to run blender. My time is important! It's not so fast on this G4, what's the cheapest and best performing kit I can get to do it?"

I downloaded blender for OS4, it works and is fine, but if I needed to render seriously, my PC is far, far faster. SSE3 optimised and blender can use all four cores. I'd have to have real issues to choose the former in preference to the latter if it was something I wanted to do seriously.

[Karlos]

@Karlos,

One thing I don't understand where your coming from is that you keep asking us why we would want to run say Blender on a G4/MorphOS machine when we have a PC sitting here.
when you yourself just said that An A1200 is your main music production machine?

Under your logic, why use an A1200 and not your new x86 PC to do your music production?

Well, that's a case of chalk and cheese. You are comparing very different tasks.

My A1200 runs OctaMED SS as a MIDI sequencer. First of all, OctaMED SS is my preferred tool as I have years of experience with it. There is, or was, a version for windows but frankly, I didn't like the interface and secondly I tend not to run windows. Now, MIDI sequencing is not an intensive task; the 68040 in that machine is already more than fast enough for the job, hence there is nothing to be gained from running it on a faster box. In fact, I did try using UAE but in the end, there was little gain and rigging up my existing MIDI gear to it was a PITA. So, my A1200 is totally up to the job I ask of it. When I'm making music on it using my preferred application, it's already as good as it needs to be and no amount of additional go faster stripes actually help it.

Secondly, my PC does factor into music production. Recording, mixing, post production and soft synthesis. All of which IO/compute bound tasks for which my A1200 is underpowered.

Unlike MIDI sequencing, raytracing, is very much a compute/memory bound task. It's great that blender exists for AmigaOS4/MOS and hopefully time will see it fine-tuned and optimised, but even by then, I don't see it competing with an SSE3 optimised build on a fast x86-64 machine with several GB of high speed memory, running parallel render tasks on a true SMP platform.

In short, it's horses for courses. If you start justifying your choice of hardware by choosing compute bound tasks, why pick underpowered hardware to run them on? As cheap as you can pick up a G5 for, it isn't going to compete with an x86 box you spent the same total on (assuming you factor in the cost of the OS, which if you pick linux, is nada).

[Karlos]

You like Octamed, maybe another user really likes ImageFX?. That program would fly on a G5 and only runs on Amiga's.

I'm sure it would, but again, does it not already fly on a 1.5GHz G4, considering that it was written for m68K machines that never really got past ~80MHz?

I suppose, though, it's a much better example of a potential candidate application, certainly a far better suggestion than any so far (given that it's not a port from another system). Image processing can always use more power; you just have to work with bigger images or more complex processing to get the benefit of it.

[Karlos]

A1200 is obsolete because it doesn't have built in midi-ports and cubase like my Atari Mega ST. Plus the Atari doesn't multitask, so every bit of cpu's prosessing time goes to midi sequencing.

:lol:

However, for the Atari Mega ST:

1) I don't have one

2) As nice as the built in MIDI interface is, it's actually less versatile than the one I already own.

3) It doesn't run OctaMED SS

4) I don't really want to buy another machine to do a job my machine does perfectly well,

5) Irrespective of all the above, the Mega ST probably doesn't the job as well as my existing machine anyway

[Karlos]

I think it's funny that an Amiga enthusiast has to defend his preferred choice of OS and HW for his Amiga hobby here on Amiga.org. This is something I would expect over at slashdot, i.e. "Linux on x86 is cheaper, faster and better". Well it might be, but it won't be MorphOS.

:lol:
Fear not, help is available.

I find it funny that you find a perfectly reasonable question something you need to "defend" against. All I asked was "what do you already do with MOS on your G4 that you actually need a G5 for?"

And frankly, you brought that question on all by yourself with your e-pine rubbing over the G5 slaying all competition

With the G5 support, MorphOS will support the fastest PPC architecture ever made. That's a good thing IMO, not a bad thing.

Nowhere in this thread or anywhere else have I suggested it is a bad thing. That inference is entirely in your going on in your mind.

In fact, I think you will find that the very first thing I did was to ridicule the sorts of arguments I was expecting from certain trolls regarding obvious lack of 64-bit/SMP support. You know, exactly the same arguments that were raised by some people when it was announced that the equally 32-bit, non SMP OS4 would be migrating to a 64-bit dual core processor.

And if it isn't for you, the MorphOS team has showed MorphOS running on a broad spectrum of Mac HW, a whole palette, where each option has it's own individual key benefit:

Mac Mini (Small)
eMac (Cheap)
PowerBook (Laptop)
PowerMac G4 (Cheap, expandable, "real" case)
PowerMac G5 (Powerful)

I am fully aware of the machines MOS is available for. As I have said more than once, I am slightly puzzled as to why migrating to G5 seems to be happening ahead of improving support on the above machines.

To reiterate, there are plenty of G4 class Mac machines that used nVidia GPUs, but you can't run MOS on them thanks to the lack of video driver support. Yet, AROS already has gotten nVidia support thanks to Gallium, so it's no great leap of intuition to contemplate MOS (and OS4 for that matter) to examine the possibility. Or how about getting wireless networking up and running on the boxes that are supported?

I know that it isn't Linux on x86, but for an Amiga enthusiast, this is a lot to choose from depending on your needs and wants. This are the best mainstream machines the PPC had to offer, and none of the options will ruin you. MorphOS doesn't run on x86,

Not yet. But, ask yourself seriously though, where do you go after the G5? Or is the G5 to be to MOS what the 68060 was to OS3.x?

The same argument is true for the PA6T.

and the G5 is the most powerful PPC there is. Supporting it is a good thing, not a bad thing,

There you are again, rubbing your behind as if I just kicked it for asking a perfectly reasonable question.

it makes the picture above complete.

No it doesn't. It would be complete if the major features of each device were adequately supported. They aren't. As I said, fixing wifi and video support would make it far more complete than adding support for one more CPU, which you can still do afterwards.

[Karlos]

I'm slightly puzzled why you think G5 is going to happen before the said machines.

Well, the initial announcement and subsequent excitement might have had something to do with it The key phrase in the above is "seems to be".

Sure there will be things that won't be supported anytime soon (nvidia, wlan, bluetooth), but basic robust support for the announced machines definitely comes first.

Thank you, Piru, that's reassuring to hear.

[Karlos]

And I understand Karlos' argument really well. I fact, since I recognize that I'm not going to be doing anything with a G4 Powermac that I can't do with a 1.25Ghz eMac, the first Powermac I've put together is a 1 Ghz Quicksilver. I'm pretty sure I can overclock this to 1.2 Ghz (which is close to my current 1.25) and if third party CPU upgrades are supported I can go as high as a 2.0Ghz 7448 (the fastest MDD upgrade is a 7447 which would be less powerful).

No doubt you'll barely notice any real loss in performance in everyday usage, since most of the things we do on our amiga/amiga-like systems are not compute bound.

I don't expect to be able to decode HD video without high CPU usage or a video card based decoder. And, as before, there's going to be some applications that the system is either not powerful enough for or that software doesn't exist for under MorphOS.

However, I'm guessing that pretty much every Amiga specific application you can run on it, will run extremely well indeed, right?

Will G5 support be cool? Will we be able to keep pace with AmigaOS4 hardware? Of course, its going to seriously kick ass, and we'll be paying less for our hardware.

Agreed. As I said, the "I want to run my preferred OS on the fastest hardware it can run on" is a motivation I can completely understand.

The thing that Karlos is aware of and that I must admit I'm painfully aware of (as someone who worked for a company selling 68K based computers in the late 80's and early 90's) is that even if we have reasonably powerful hardware we're facing an impossible task.

Its simple, with the number of Windows and OSX computers sold each year the market for software for those machines is very large. As the time and energy necessary to create good software is also quite great, developers gravitate toward those markets with the greatest potential to make a profit.

As such, many great ideas and even better alternatives (to the current dominant platforms) have failed in the past. We need to keep our feet on the ground and remain realistic. The chances of a mainstream revival are slim. Our market does seem to be enjoying a period of renewal and growth. But, our systems and our software still serve a hobbyist market. Whether they will ever serve as primary systems of production remains questionable.

QFT.

-edit-

Seems I missed this post:

As to your other question about what do we need the G5 power for, and there are no Amiga apps that need that kind of power, you are obviously right for 99% of the apps, but with the old 68k apps needing to be run through a JIT engine and not natively, any demanding Classic Amiga app, such as 3D rendering, will benefit from the extra power. But more important in my minds eye is that the extra computing power will open a few programming opportunities that everyone thought could not be accomplished on any Amiga computer in the past. With G5, or PA6T power and if/when better graphics cards are supported, it will be possible to create better Amiga applications in the future than might be possible with only the current power of a 1.5GHz G4 MacMini, or in AmigaOS4.x's case, a 1GHz G4 Pegasos2, or AmigaOne.

I have to admit, I'd love to see that, for to me that would be a sign of genuine progress. I honestly think that OS4 and MOS are operating systems without software more than they are operating systems without hardware. Sure they run a large slice of legacy applications, but tiny minority of those actually benefit from the step up in performance, which has been my point of contention throughout.

Aside from apps and codecs ported from other systems where the required performance to run is considered minimal at best, neither OS4 or MOS presently seem to have applications that would appear to have been made possible solely by the improved performance they have over classic machines. And in my book, that's a real pity.

Concerning your argument that we all have other computers that can run anything that can be run on a G5 or PA6T Amiga system faster and cheaper, not everyone wants to have to switch back and forth between different systems and many of us hope that with a sufficiently powerful Amiga system, more Amiga programmers will return to coding new programs for us to use and we might be able to do 100% of what we now do on multiple computers on just one powerful Amiga computer. Not everyone here HAS to have all the latest software, or web plug-ins to satisfy our computing needs and I would say that many of us here at A.org are the die-hard few remaining Amiga enthusiasts that would prefer to ONLY use an Amiga computer and OS and not have to boot into any other OS ever again. Maybe that is not realistic, and some will say it is not possible, but everyone has different computing styles and needs, so I say to those nay sayers, don't tell me what I have to own, or have to run to be happy with my computing experience.

A reasonable argument, honestly presented. However, I don't think the various polls over the years (not to mention the server stats) really support your idea that many regular visitors here are amiga-only folk. There is a vocal minority, but the majority would seem to have other machines, at least for internet access.

[Karlos]

G5 on morphos is perfect and the fastest "amiga" never see, you can launch a amiga program, like lightwave 3d, aladdin, art effect etc at the speed of 75% of ppc cpu using jit trance... this is a lot and a lots of mips, 1.8ghz give more than 5.000 mips for amiga native 68k aplications, warpos/powerup and morphos more than 7.000 mips, i think morphos can boot on a micro sencond

Whereas I think you'll find that you could get a similar percentage of native speed for any of the above applications a core i7 just using UAE. Or just run Lightwave 3D at 100% native speed on the same machine, since all versions from about 5.0 onwards are for x86 anyway

Of course, you could say "yeah but you have to emulate the chipset". Funny that, because running it in RTG mode seems to cause that problem to vanish. And, of course, it still can emulate the chipset, so it's even more compatible with original amiga applications.

and with very little memory usage windows is slow loading system, loading, loading loading... and consumes a lot of ram, morphos is ultra fast and mega-optimized, and of course, amiga compatible.

Super effective meme, much?

FYI, not all x86 machines are running super slow slow mega drag lag winblows

[Karlos]

As far as I see it, it would make sense to migrate to a different CPU architecture. ARM or x86.

"But why would they go x86, I mean, how could Amiga *possibly* benefit from the power of modern x86 HW...?" :rolleyes:

It isn't the power it would benefit from, at least not for some time*, it's the fact that they are cheap, plentiful and are extremely unlikely to go out of production at any time in the foreseeable future. In contrast to the G5, which has already been out of production for some time.

-edit-

*Of course, one obvious benefit of all that horsepower is that you'd be able to preserve backwards compatibility for PPC applications via emulation. After all, Rosetta and PearPC have shown it can be done.

[Karlos]

I don't know Karlos, with current AOS and MOS revisions not supporting SMP I'm not convinced that X86 is the way to go yet.

Point taken, but both OS4 and MOS as 32-bit non SMP are moving towards hardware SMP capable 64-bit platforms regardless, so really, that boat has already sailed.

And how would a 32bit OS without memory protection (and in the case of MorphOS a 1.5GB memory limitation) compete against existing X86 OS'?

The same way AROS does - or rather doesn't; by appealing to an existing user base instead of taking on the world. In truth, the only competition OS4 and MOS have is each other and that isn't going to change by moving to a more popular hardware platform. Alternative OS for x86 are legion; none of them are curling up and dying just because windows, osx and linux are in town.

I actually think that an x86 port of MOS (or OS4) would face far more of a threat from AROS than it does any of the "big" ones.

Regarding the memory limitation (which I wasn't actually aware of TBH), the observation applies equally to the PPC mac hardware it already supports.

With MorphOS, assembly language is still within my reach (as the number of PPC instructions is still manageable). With an X86 processor? No, I'm going to need higher level programming tools.

Actually, the assembler for x64 is quite clean. You'd be surprised (I know I was), it's a far cry from the arcane old x86 stuff. And of course, you can dispense with the horrid old stack based x87 nonsense and go straight for SSE. Of course, there's not as much call for it, but if you must, it is still there.

That said, I still prefer writing m68k assembler over everything else

There are a lot of reasons I can't support another platform change just yet.
And AROS is there's for those that want to switch. Right now we're spread across 68K, PPC and X86 platforms. That diversity may be of benefit to us.

Sure, that's understandable and FWIW, despite how it may sound, I actually enjoy  PPC. Variety is the spice of life. What worries me is the long term viability of supporting hardware that is no longer in production (not the PPC per se, but desktop systems built on it).

[Karlos]

Granted, as Karlos likes to point out, most of those apps that we can run on MorphOS (AmigaOS4 is in the same boat) exist on other platforms that run orders of magnitude faster than any MorphOS2.x or AmigaOS4.x machine at a fraction of the price, but I fail to see the logic of why any MorphOS or AmigaOS4 user would want to keep using underpowered hardware that limits the kinds of things we can do on the OS we prefer to use, not are forced to use because the software we want to run is only available on other OSes that we do not enjoy using as much as an Amiga-Like OS. Does Karlos use either MorphOS2.x or AmigaOS4.x? I know he uses AmigaOS3.x as he uses his A1200 for music

Actually, a MorphOS 2 machine is the only "amigoid" system missing from my collection. They're not all listed in my signature, you know

Regarding the application situation, I can only reiterate what I said in #153:

As to your other question about what do we need the G5 power for, and there are no Amiga apps that need that kind of power, you are obviously right for 99% of the apps, but with the old 68k apps needing to be run through a JIT engine and not natively, any demanding Classic Amiga app, such as 3D rendering, will benefit from the extra power. But more important in my minds eye is that the extra computing power will open a few programming opportunities that everyone thought could not be accomplished on any Amiga computer in the past. With G5, or PA6T power and if/when better graphics cards are supported, it will be possible to create better Amiga applications in the future than might be possible with only the current power of a 1.5GHz G4 MacMini, or in AmigaOS4.x's case, a 1GHz G4 Pegasos2, or AmigaOne.

I have to admit, I'd love to see that, for to me that would be a sign of genuine progress. I honestly think that OS4 and MOS are operating systems without software more than they are operating systems without hardware. Sure they run a large slice of legacy applications, but tiny minority of those actually benefit from the step up in performance, which has been my point of contention throughout.

Aside from apps and codecs ported from other systems where the required performance to run is considered minimal at best, neither OS4 or MOS presently seem to have applications that would appear to have been made possible solely by the improved performance they have over classic machines. And in my book, that's a real pity.

[Karlos]

IOf course AROS would be the bigger threat as it is closer to appearance, and function and it is free.  Have you ever considered that fact might be a big contributing factor for why MorphOS2.x and AmigaOS4.x have resisted moving to x86 so far?  Why would they want to compete with not just one free OS, but probably dozens, when both teams of developers are just tiny groups of programmers that could never compete with other OS choices that might have ten to a hundred times the man-power to work on an x86 alternate OS?

It has occurred to me many times and I'm sure I've said as much in the past. However, the key functionality required of any MOS x86 / OS4 x86 implementation would be PPC emulation, something which I doubt AROS developers are really interested in. You have to admit, the prospect of moving to x86 and sacrificing all compatibility with your existing software base is not an appealing one.

I hate to say this but if OS4 and MOS plan to stay tied to PPC forever, then enjoy it now, since with the G5 mac, you have reached the pinnacle of what the architecture had to offer for desktop systems. It's been 4 years since the line was discontinued. I admit, we're still using m68k amigas 25 years later, but do you expect to be using a PPC machine for as long?